Created: 1st December 2000, last updated: 30th December 2000, © 2000 ABRF

ABRF 2001

THE NEW BIOLOGY

Technologies for Resolving Macromolecular Communications

An International Symposium Sponsored by The Association of Biomolecular Resource Facilities

Town & Country Resort
San Diego, California
February 24-27, 2001

POSTER ABSTRACTS

P1-S

Compatibility of the RapXtract Dye Terminator Removal Kit with purification of sequencing reactions for the applied biosystems Prism 3100 Capillary Sequencer.

A.L. Springer, L.R. Booth, K.A. Hughes, R.J. Kaiser, D.A. Spicer; Prolinx, Inc., 22322 20th Avenue SE, Bothell, WA 98021

Successful automated DNA sequence analysis requires that sequencing reactions be purified of excess unincorporated dye terminators. Common methods for purifying sequencing reactions involve several steps and require processes that are not easily automated. Prolinx,^® Inc. has developed the unique RapXtract™ Dye Terminator Removal Kit for this application that is rapid, easy to perform, yields high quality product and does not require modified primers. The RapXtract kit uses a magnetic particle format to remove dye terminator contaminants by mass action. The protocol has three steps: 1. Remove storage buffer from particles, 2. Add sequencing reactions and mix, and 3. Remove purified extension products. The ABI PRISM 3100 Genetic Analyzer is a 16-capillary instrument that utilizes technology from other ABI capillary sequencers in a medium-throughput format. RapXtract-purified samples analyzed on the ABI PRISM 3100 yielded sequences with good Phred quality scores and signal strength. The RapXtract kit has been optimized to minimize the use of expensive sequencing reagents and can be performed on reactions containing as little as 1 µl of Big Dye Reaction Mix and as little as 5 µl total volume. The RapXtract kit has been used successfully with double-stranded plasmid, PCR product or single-stranded DNA templates. RapXtract purification can also be used for dye primer reactions. The ease of use and versatility of the RapXtract kit make it a good choice for manual or automated purification of sequencing reactions.

P2-M

Serial analysis of gene expression using the Applied Biosystems capillary electrophoresis platforms.

K.M. Gunning, B. Nutter, A. Swei, J. Zon; Applied Biosystems, 850 Lincoln Centre Dr., Foster City, CA 94404

In today's genomic environment, the draft completion of the human genome has allowed researchers access to a wealth of information with respect to gene identity and gene expression profiles. Scientists have developed novel ways in which to elucidate this information and SAGE is deemed one of the more comprehensive methods available for rapid, detailed analysis of large numbers of cellular transcripts. To further enhance the rapidity at which one can determine this global gene expression profile, one needs more efficient and precise methods for analyzing these transcripts and for assessing their abundance. Automated capillary electrophoresis provides a platform whereby the sequence of these many transcripts can be determined with speed and precision. The ABI PRISM.^® 3700 and the ABI PRISM.^® 3100, the most recent addition to the CE family of platforms, provide the SAGE researcher with a means by which they can determine quickly and reliably, the transcript information necessary to build these gene expression profiles. These platforms partnered with the uniformity of signal strength and length of read of BigDye™ Terminators, are the labor saving answer to large scale transcript analysis using the SAGE method. We will demonstrate the ease of use and flexibility of these platforms as a preferred method for the sequence analysis of ditags in the SAGE process.

P3-T

Improved signal strength on ABI Prism 377 DNA Sequencer and 310 Genetic Analyzer by using a new sequencing run module.

S-M. Chen, H. Zielke, G. Amparo, S. Spurgeon; Applied Biosystems, 850 Lincoln Centre Dr., Foster City, CA 94404

In many sequencing labs DNA sample template quality varies greatly due to the diverse sources of these samples and the different DNA preparation methods used. This variable DNA template quality results in variable sequencing data quality. Many low quality data are associated with low quality or quantity of DNA templates. This frequently results in low signal strength and low signal to noise ratio. A new sequencing module was developed for the ABI PRISM 377 and 310 systems that will generate greater signal strength during DATA collection. With the use of this module in data collection we were able to improve signal to noise ratio and were able to improve success rate in data analysis.

P4-S

Two steps cycle-sequencing improves base ambiguities and signal dropouts in certain BigDye DNA sequencing reactions.

L. Wen; San Diego State Univ., 5500 Campanile Drive, San Diego, CA 92182-4614

The use of automated fluorescent DNA sequencer systems and PCR-based DNA sequencing methods play an important role in the actual effort to improve the efficiency of large-scale DNA analysis. While dideoxy-terminators labeled with energy-transfer dyes (BigDye; PE/ABI) provide the most versatile method of automated DNA sequencing, premature terminations results in a substantially reduced reading length of the DNA sequence. Premature terminations are usually evidenced by base ambiguities and are often accompanied by diminished signal intensity after that point. I studied a two-step protocol for Taq cycle sequencing using the ABI BigDye terminator for reducing premature terminations in DNA sequences. I demonstrated that combining the annealing step with the extension step at one temperature (60°C) reduces premature terminations in DNA sequences that regularly contain premature terminations when the three temperature steps is used. This procedure can be significantly improved sequence quality in our core facility. Sequence results from initial studies are still forthcoming and will be presented.

P5-M

Using high speed data collection on the ABI Prism 377XL to improve core facility DNA sequencing throughput.

F. Lach, J. Medalle, M. Randesi, B. Imai; Rockefeller Univ., 1230 York Avenue, Box 105, New York, NY 10021

The variety of samples subjected to a core facility presents a challenge when it comes to achieving high throughput DNA sequencing. Any given run can exhibit different degrees of success based on template type and size, template base content and arrangement, primer design, and overall sample quality. Although the ABI Prism 3700 DNA Analyzer excels in high throughput capabilities, the instrument still requires stricter sample management that is harder to achieve in the core facility environment. The ABI Prism 377XL continues to be an important resource since it requires less sample stringency than the 3700. Modification of the 377XL module parameters and new polyacrylamide premixes give the potential for increased sample throughput with minimal loss of resolution. This study will investigate the conditions required to obtain quality data using the 4X high speed data collection module on the 377XL. A data comparison will be made between the 2X standard speed and 4X high speed modules using various samples and polyacrylamide premixes.

P6-T

Update on the continuation of the DNA Sequencing Research Group 2000 Study: an evaluation of methods used to sequence and isolate bacterial artificial chromosomes.

T. Thannhauser¹, L.S. Hall², J. Hawes³, T. Hunter⁴, E. Jackson-Machelski⁵, K. Knudtson⁶, D. Leviten⁷, M.A. Robertson⁸; ¹Cornell Univ., Rm 149 Biotechnology Bldg, Ithaca, New York 14853-2703, ²Albert Einstein Col. of Med., 1695 Poplar Street, NewYork, NY 10464, ³Indiana Univ. Sch. of Med., ⁴Univ. of Vermont, ⁵Washington Univ. Sch. of Med., ⁶Univ. of Iowa, ⁷ICOS Corp., ⁸Univ. of Utah

Recent emphasis on BACs as a primary source of template has created significant challenges for those who operate DNA sequencing facilities. Due to their large size (80-350 KB) BACs behave differently in sequencing reactions than do plasmid clones. To achieve optimal performance, standard sequencing reactions have to be modified. Moreover, due to their low copy number it is difficult to isolate enough BAC template using standard miniprep protocols. Using data submitted to part1 of the original DSRG 2000 BAC study, the DSRG extracted a recommended protocol for BAC sequencing and performed carefully controlled experiments in an internal study to test the efficiency and reliability of the protocol. This protocol was made available along with a standard BAC template and primer for testing in the community.

Part 1 of the study examines the effect of a DSRG recommended sequencing protocol on the quality of the results obtained on a standard BAC template sent to participating laboratories. The sequencing results are analyzed for quality and length of read on the wide variety of instrumention used by the DNA sequencing community. The success rate of the standard protocol is compared to the internal study results and to the original results where applicable.

Part 2 of the study continues to assess and evaluate the quality of BAC isolation methods commonly used in DNA sequencing facilities with the goal of providing a recommended isolation protocol. The isolated BAC DNA is sequenced using the DSRG recommended protocol. Results will be analyzed for quality and length of read on standard instrumentation to determine the success of the different BAC isolation procedures.

P7-S

Automated transfer of DNA sequencing files to an ftp site.

K.M. Ivanetich, R. Taylor, W. Yan, F. Sandifer, D. Wolber; UCSF, Box 0541, San Francisco, CA 94143

The Organize Samples program has been developed to automate the renaming, consolidation, and transfer of sequencing data from Applied Biosystems 373, 377 and 3700 DNA sequencers directly onto the facility's ftp site. The program accesses (1) information in the Applied Biosystems sequencer's analysis file, (2) the Applied Biosystems Tool Kit, (3) a file containing a list of user's names paired with their laboratory's principal investigator's names, and (4) the compression program DropStuffit. Using the above, the program compresses all sequence files for one user into one document, names the document with the user's name and the date, and places the compressed file into the appropriate laboratory's folder on the facility's ftp server. The laboratory's folders are password protected. Users can access their laboratory's folder, download the file, and uncompress it with a single click. The result is a set of uncompressed files, with one file for each sequence analysis file. The uncompressed files are named with the run and lane numbers or coordinate and capillary numbers plus the user's template and primer names. Separate sequence text files are handled equivalently by Organize Samples, and are differentiated by a ".seq" suffix. Users can access the ftp site and download their DNA sequencing data from their own computer 24 hours/day-7 days/week.

P8-M

Sequencing performance of the ABI Prism 3100 Genetic Analyzer.

P.A. Baybayan, B. Johnson, K. Roy, R. Pingue, Q. Liwei; Applied Biosystems, 850 Lincoln Centre Drive, Foster City, CA 94404

The new ABI Prism 3100 Genetic Analyzer is the latest addition to the suite of capillary electrophoresis platforms offered by Applied Biosystems. This 16-capillary electrophoresis platform employs complete automation designed to meet the demands of mid to high-throughput sequencing and genotyping laboratories.

The ABI Prism 3100 Genetic Analyzer offers flexibility by providing options for a variety of applications. In de novo sequencing where throughput, consistency and efficiency are important, several arrays (including 36 cm, 50 cm, and 80 cm) are available. These arrays require different run parameters and therefore, perform differently. The user will have the option to an array depending on their requirements (readlength and throughput). In comparative sequencing where accuracy and sensitivity are required for the detection and identification of polymorphisms, a 36-cm array is sufficient. Here, we will present the performance (readlength, accuracy, consistency) of the ABI Prism 3100 Genetic Analyzer in the area of de novo and comparative sequencing and demonstrate how this instrument is the platform to best accommodate to your research needs.

P9-T

A comparison of alternative methods for sequencing through difficult templates.

D.A. Bintzler, Y. Song; Univ. of Cincinnati, 231 Bethesda Avenue, Cincinnati, OH 45267-0524

Template DNA that has a seventy percent or greater content of guanosine and cytosine bases (GC rich) has been a potential problem for automated fluorescent DNA sequencing. Two well-known strategies for improving the base read length of the GC rich template include the addition of dimethyl sulfoxide (DMSO) or betaine to the PCR amplification of the template. Applied Biosystems Incorporated (ABI) has offered a third method with dGTP big dye chemistry. The dGTP chemistry has been significantly effective for extending through templates that contain a chain of ten or more guanosine bases. The dRhodamine chemistry (ABI) has also been effective for some problematic templates. However, no single method has been effective for all problematic templates in this facility. Therefore, research has continued to find an alternative method. A PCR additive available from Qiagen, Q buffer, was introduced as a potential method for improving the base read length of the GC rich template. Preliminary studies showed that Q buffer had improved PCR extension through difficult templates that were not successfully sequenced with the other methods previously mentioned. However, a thorough investigation of the Q buffer compared to DMSO, betaine, dRhodamine and dGTP big dye chemistry was required. The results of the investigation are reported here.

P10-S

Fragment and sequence analysis in Portugal: a survey of facilities and applications.

L.M. Souto; Univ. de Aveiro, Portugal, Departamento de Biologia, Universidade de Aveiro, Aveiro, Aveiro 3810 Portugal

Since 1992, with the first unit of Automated Sequencers starting to work in Portugal, a significant improvement in the number of equipments and institutions dealing with sequencing and fragment analisys has been registered.

The leading groups are the forensic institutes which deal with a heavy routine work, namely in paternity cases. These groups are using very standardised techniques and markers, in accordance to the EDNAP (European DNA Profiling Group) and the Portuguese and Spanish Working Group of the ISFH (GEPISFH).

Recently, the universities as well as some private research and diagnostic labs began to use DNA Sequencers, opening the spectrum of applications.

We present in this poster the state of the art of Automated Sequence and Fragment Analysis in Portugal and its relationship with other iberian and latin-american countries, with a special emphasis onto the forensic genetics field.

P11-M

Integration of new sequencing technologies into the W.M. Keck Biotechnology/HHMI Biopolymer Lab. at Yale.

K.M. Hager; Yale Univ., 295 Congress Ave., New Haven, CT 06536

In the past year, we have successfully integrated the Applied Biosystems 3700 DNA Sequencer into our DNA sequencing service. This machine utilizes 96 capillaries and a proprietary polymer rather than polyacrylamide slab gels to separate fluorescently-labeled DNA termination fragments. Because of 3700's unattended robotic sample loading from as many as four sample plates per operator interaction, we have increased the lab's overall sequencing throughput by 25% with no increase in staff and lowered the average sample turnaround time. In addition, the 3700's automation has allowed us to initiate Genescan and SNAPSHOT single-nucleotide polymorphism (SNP) detection assays as new services. At present, about 75% of our total DNA sequencing samples are run on the 3700 with the remainder run on one 377 (slab gel). About 2-5% of the samples run on the 3700 give a poor separation and these samples are rerun on the 377. Such poor separations are due to either improper loading of the separation polymer into the capillary or overloading of the capillary with excessive amounts of DNA. On the vast majority of runs, our pgem standards yield 650-750 bases of sequence data at 99% or greater accuracy (POP-6 polymer).

Recently, we have begun an evaluation of Pyrosequencing, a non-electrophoretic DNA sequencing technology using an enzyme coupled assay where light is ultimately generated by luciferase from PP_I released following nucleotide incorporation (Science 281, pp. 363-364 (1998)). Our initial applications for Pyrosequencing will include sequence tag (20-40 base) determination and SNP detection.

P12-T

Investigation of the entire spectrum of genes induced by staphylococcal enterotoxin B in human lymphoid cells using differential display-PCR.

C.A. Mendis, C. Sanchez, R. Das, M. Jett; Walter Reed Army Inst. of Res., 503 Forney Drive, Silver Spring, MD 20910

Since its introduction in 1991, Differential Display has evolved into a powerful and efficient technique in analyzing differentially expressed genes in various cells under altered conditions. Here we used differential display not only to screen the entire gene population in human lymphoid cells induced by SEB but also to understand the specificity of the expression pattern of altered genes by comparing it to the expression by Cholera Toxin (CT). In order to approach our goal in a systematic manner as well as to increase the efficiency of cDNA product recovery we used a set of arbitrary and anchored primers to further subdivide the gene population. Out of the 750 altered genes, at least 200 genes showed a SEB specific expression pattern when compared to CT exposure. The altered genes were then subjected to a high throughput cloning procedure and sequenced. Out of the 250 genes that were sequenced, 150 genes matched with available sequences in the Gene Bank and EMBL databases. Unique to SEB are genes involved in ancillary functions as adhesion molecules, regulators of vascular tone, wound healing, inflammation, heat shock, cell death and T-cell proliferation. Altered genes that matched with known sequences were verified by northern blots, RT-PCR or real time PCR. Genes that were specific for SEB were then placed on glass chips for use with other cell lines or to understand the gene expression profile of other toxins. In the quest for finding a set of genes that can be used as surrogate markers, differential display has shown to be a proven technique as it has the capability to look at the whole gene spectrum irrespective of wide use or availability of individual genes.

P13-S

Microsatellite analysis using fluorescent PCR primers synthesized in tandem.

S.L. Wegener, G.J. Wiebe, S. Yu, R.T. Pon; Univ. of Calgary, 3330 Hospital Drive N.W., Calgary, AB T2N 4N1, Canada

A new procedure for solid-phase synthesis of multiple oligonucleotides linked end-to-end (tandem synthesis) on the same solid-phase support has recently been developed in our laboratory. Upon cleavage and deprotection, the oligonucleotides are released from each other yielding a mixture of two or more sequences in the same solution. This procedure has practical benefits for applications requiring large sets of oligonucleotides, especially PCR amplification, since the number of individual oligonucleotide syntheses to be set-up and the number of individual oligonucleotide samples handled is reduced by half. Fluorescently labelled primer pairs for automated genotyping are also possible by incorporating a fluorescent dye phosphoramidite onto the 5'-end of the terminal oligonucleotide. PCR primer pairs with FAM, HEX, and TET fluorescent labels have been prepared by tandem synthesis and used to amplify known microsatellite markers from (C57BL/6 X NOD) mouse genomic DNA. Automated microsatellite analysis using GeneScan and an Applied Biosystems Prisim 377 DNA sequencer was then performed.

P14-M

Economical usage of value-added phosphoramidites.

J.B. Hobbs; Univ. of British Columbia, #237-6174 University Boulevard, Vancouver, BC V6T 1Z3, Canada

A method is described for economical use of value-added (dye, biotin, etc.) phosphoramidites. The method was developed on an Applied Biosystems 380B synthesizer. The amidite is placed in a glass autosampler vial which is located in a plastic jacket inside a 10 ml glass vial. The amidite is dissolved in situ using a user-variable microdilution adapted from a bottle-change procedure. Typically 5 mg. of amidite are dissolved in 100 microlitres of acetonitrile. The amidite is used in conjunction with Applied Biosystems LV columns and synthesis cycles adapted for use with LV columns which use a similar "multiple-coupling" approach to that seen in "LV cycles" on later (392, 394, etc.) Applied Biosystems synthesizer models. These cycles are also described. Good tagging efficiency can be obtained for 40 nmole or 200 nmole scales using 5 mg. of amidite in this way, and the wastage of large quantities of expensive amidites is avoided. While the methods are applicable to later model synthesizers, the relatively modest flow rates, gas pressures and line volumes of the 380B render it a good platform for processes of this type.

P15-T

Quality control in a core oligonucleotide synthesis facility.

R.R. Muhlhauser, C.G. Miller, A. Yeung; Fox Chase Cancer Ctr., 7701 Burholme Avenue, Philadelphia, PA 19111

Our facility makes about 400 oligonucleotides each month. For the past 16 years, we have done quality control on each oligonucleotide prior to delivery. Our users do not want the quality of oligonucleotides to ever be a variable for troubleshooting in their experiments. Even a failure rate of 1% in oligonucleotide synthesis would have caused four experiments to fail each month. Such a low rate of synthesis failure is difficult to isolate in a statistical approach of quality control by random sampling. Analyzing every oligonucleotide also allows us to catch instrument and reagent problems more quickly. By analyzing each oligonucleotide immediately after deprotection, we reject 1 to 3% of the oligonucleotides we make that do not meet our quality standard. These oligonucleotides are resynthesized without delay. The oligonucleotide analysis method we use is anion-exchange at pH 12.5, using the Mono Q system on an FPLC (Yeung, A. T. and Miller, C. G. Anal. Biochem. 187:6675, 1990), assisted by a home-made autosampler and autoinjector. At that pH, both n-1's and deprotection problems are visible. We highly recommend this practice because it does not add appreciable expense or effort to the cost of the oligonucleotide, but provides the facility personnel with assurance in the quality of their products.

P16-S

Incorporation of pseudouridine and 4-thio-uridine into RNA oligonucleotides using 5'silyl-2'-ACE-orthoester chemistry.

S.A. Scaringe¹, D. Kitchen², J. Qui²; ¹Dharmacon Res. Inc., 3200 Valmont Road #5, Boulder, CO 80301, ²Dharmacon Res., Inc., Boulder

The ability to incorporate a wide range of modified ribonucleotides into RNA oligos is an essential requirement for RNA chemical synthesis methodologies. Many modified bases have already been incorporated into RNA oligos using 5'-Silyl-2'-ACE-orthoester chemistry. We report here the incorporation of two more important modified bases, pseudouridine and 4-thio-uridine.

Pseudouridine was converted from the nucleoside to the protected nucleoside phosphoramidite in five steps in 45% overall yield. During synthesis the amidite coupled in 99% stepwise yields. The high yields for both the amidite synthesis and the oligonucleotide coupling have allowed pseudouridine to be incorporated into RNA on a routine basis. For example, a series of RNA oligonucleotides containing 1-3 pseudouridines were synthesized to study the 1920 loop region of E. coli 23S RNA.

4-Thio-uridine was incorporated into RNA oligos using the convertible nucleoside approach. The protected 4-triazole-uridine phosphoramidite was synthesized from uridine in six steps in 32% yield. The amidite coupled in 99% yields. Following oligo synthesis the support bound oligonucleotide was treated with thioacetic acid, buffered to pH 8.0, for 6 hours at 55°C. The support was washed and then treated with 10% DBU methanol at room temperature for 24 hours. The oligonucleotide was desalted and 2'hydroxyl protecting groups were removed using a pH 3.8 aqueous buffer for 30 minutes at 55°C. The resulting oligonucleotides contained 98% 4-thio-uridine at the desired position as assayed by anion exchange HPLC analysis. This is significantly higher than the results using cyanoethyl protection of 4-thio-uridine. All oligonucleotides containing 4-thio-uridine clearly exhibited the characteristic 330 nm absorption peak. 4-Thio-uridine is now being routinely incorporated in RNA oligonucleotides for use in several collaborative studies.

P17-M

New methods for the synthesis of multiple oligonucleotide sequences either singly or linked end-to-end in tandem.

R.T. Pon¹, S. Yu¹, Y.S. Sanghvi²; ¹Univ. of Calgary, 3350 Hospital Dr. NW, Calgary, AB T2N 4N1 Canada, ²Isis Pharmaceut.

Solid-phase oligonucleotide synthesis requires attachment of the 3'-terminal nucleoside to the support, usually through a 3'-ester linkage. Phosphoramidite methods for performing this attachment through a phosphate linkage on "Universal" supports have been developed. However, these methods introduce a terminal 3'-phosphate residue which is difficult to remove. We have developed new linker phosphoramidite reagents which eliminate this problem by incorporating a cleavable 3'-ester linkage within the linker arm. Thus, only products with 3'-OH ends are produced and no extra conditions for 3'-dephosphorylation are required. These new reagents provide an alternative to our previous method for automated nucleoside attachment using HBTU and DMAP (Pon and Yu, Synlett 1999, 1778). The linker phosphoramidites are especially suited to high-throughput synthesis in 96-well plates since inexpensive, underivatized CPG can be used as a "Universal" support. Nucleoside attachment to the 5'-OH group of a prior sequence is also possible. This allows multiple oligonucleotides linked end-to end in tandem to be prepared in a single synthesis. The linked oligonucleotides are released from each other upon cleavage from the support and used together as a mixture. Sequences prepared in tandem can be PCR primers, forward and reverse sequencing primers, or duplex DNA fragments. The total number of bases is only limited by the support's pore size. We have prepared strings of oligonucleotides up to 96 bases long (4 X 24 base primers) and combinations of up to ten shorter sequences.

P18-T

Performance of high-speed oligonucleotide synthesizer in 96-well format.

D. Luk, A. Tuyet-Doan, S. Hall, J. Koh, R. Guettler; GeneMachines, 935 Washington St., San Carlos, CA 94070

Oligonucleotides are essential elements of genomic research. The PolyPlexTM oligonucleotide synthesizer makes 96 different oligos in standard 96-well format, which is amenable to downstream high-throughput processing and handling. Operator time is minimal and synthesis time is less than 3 hours for a 96-well plate of 20-mers. By eliminating flushing of reagent lines through a parallel dispensing technology, optimal synthesis time and reagent consumption are achieved. PolyPlex's low reagent consumption generates oligos quickly, at low costs and small scales as low as 10 nmol. Synthesis costs, including all consumables, are less than $0.10 per base. The PolyPlex synthesis chamber provides an inert-gas environment where synthesis progress can be monitored by using full-plate trityl collection after any base addition. PolyPlex utilizes fully licensed chemistry and generates oligos with greater than 98% coupling efficiency. Versatile, easy-to-use software provides powerful synthesis control along with extreme simplicity.

P19-S

Production of antigen specific MHC class I tetramers as a core facility service.

P.S. Adams, T.B. Miller, M.J. Dobrzanski, R.J. Hogan, D.L. Woodland; Trudeau Inst., 100 Algonquin Avenue, Saranac Lake, NY 12983

Major Histocompatibility Complex (MHC) Class I tetramers are composed of a tetrameric complex of the MHC Class I molecule and a specific antigenic peptide. The Class I heavy chain has been modified to contain a BirA sequence that allows biotinylation and subsequent tetramerization of the complex with a fluorescently tagged strepavidin. These tetramers can be used to identify antigen specific CD8+ T cells involved in immune responses using flow cytometry. The Molecular Biology Core Facility prepares these tetramers as a core facility service to aid the researchers at the Trudeau Institute. One project involves the characterization of the underlying mechanisms and immunoregulatory roles of tumor antigen-reactive Tc1 and Tc2 effector cell sub-populations, which can provide insight into the modification and/or development of novel immunotherapeutic approaches to disseminated malignancies. Another project involving viral immunity has demonstrated that a substantial population of antigen specific CD8+ T cells can be found not only in the secondary lymphoid organs, but also in the peripheral tissues after resolution of a primary respiratory virus infection. These data have significant implications for vaccines that induce systemic rather than local immunity. The production procedure and examples of the usefulness of MHC Class I tetramer technology will be presented.

P20-M

Advantages of directionally immobilized 6xHis-tagged proteins for capture assays.

K. Steinert, F. Schäfer, S. Wahle, A. Hoffer, P. Söhnlein, J. Ribbe; QIAGEN GmbH, Max-Volmer-Strasse 4, Hilden, Nordrhein-Westfalen 40724, Germany

Assay signals and reproducibility can be greatly enhanced by immobilizing the assay components in a directed manner. For example, to set up an interaction assay it is essential that the immobilization of the capture molecule occurs in active conformation to allow binding of the interaction partner. Similarly, oriented binding in immunoassays results in optimal accessibility of antigenic regions and therefore increased signal strengths. Directed immobilization of assay components can easily be achieved by using 6xHis-tagged recombinant proteins and immobilizing them via their 6xHis tags. The use of this interaction is well established for efficient purification of 6xHis-tagged proteins, but is also advantageous for immobilization of recombinant proteins in highly sensitive and reproducible protein-based assays. In addition, relying on the 6xHis-Ni-NTA interaction for immobilization of assay components simplifies assay development, because immobilization conditions no longer have to be optimized for each individual protein, but rely on the same biochemical characteristics of the 6xHis-Ni-NTA interaction for each protein. As an alternative to Ni-NTA, monoclonal antibodies recognizing the 6xHis tag can be used for highly specific and oriented immobilization of 6xHis-tagged proteins. The effect of specifically oriented binding on assay results is demonstrated by comparison of ELISA experiments employing proteins passively adsorbed to polystyrene plates and proteins immobilized via their 6xHis tags, as well as by a study on the interaction of two chaperones and development of a protease assay.

P21-T

Versalinx chemical affinity tools for purification in protein analysis.

J.P. Wiley, M.D. Ferguson, A. Gall, K.A. Hughes, G. Li, K.P. Lund, M.L. Stolowitz; Prolinx, Inc., Bothell, WA, 22322 20th Ave. SE, Bothell, WA 98072-8341

Fractionating complex biological mixtures is one of the challenges in proteome analysis. MALDI-TOF (matrix assisted laser desorption ionization time of flight) mass spectrometry analysis, for instance, has a high mass accuracy but requires a relatively pure sample for positive structural identification. Prolinx^® Inc. has developed Versalinx™ Chemical Affinity Tools, and has demonstrated success using this technology to isolate and purify proteins of interest from complex mixtures. Versalinx Chemical Affinity Tools are based upon the interaction of phenyldiboronic acid (PDBA) and salicylhydroxamic acid (SHA) to form a complex that is reversible under certain conditions. Immobilization of PDBA-modified proteins can occur on a variety of SHA-modified surfaces. The system is stable over a wide pH range and is compatible with a number of denaturants, detergents and organic solvents. With Versalinx Chemical Affinity Tools, purification of biological molecules is rapid and reproducible.

P22-S

Fully automated 96-well protein purification and magnetic bead-based assay using 6xHis-Ni-NTA technology.

K. Steinert, H. Lubenow, S. Wahle, R. Stoll, J. Ribbe; QIAGEN GmbH, Max-Volmer-Strasse 4, Hilden, Nordrhein-Westfalen 40724 Germany

The purification of proteins from expression libraries requires a method that performs reliably regardless of the characteristics of the proteins to be purified. Using Ni-NTA for purification of 6xHis-tagged recombinant proteins provides a one-step purification method that is robust and meets the challenge presented by the need to purify thousands of proteins with differing structure and characteristics. To allow this method to be performed in high-throughput applications, automated protocols were developed which run on BioRobot systems.

Ready-to-run protocols covering a wide range of protein assay and purification applications are available. They rely on 6xHis-Ni-NTA technology based on the high affinity and specificity of nickel ions immobilized on matrices bearing nitrilotriacetic acid (NTA) for recombinant biomolecules with a tag of six consecutive histidine residues (6xHis tag).

Ni-NTA Magnetic Agarose Beads allow flexible interaction or diagnostic assays with structurally active immobilized 6xHis-tagged proteins as well as micro-scale purification of up to 15 µg of 6xHis-tagged protein. The purification and assay protocols based on Ni-NTA Magnetic Agarose Beads may even be carried out directly in series.

If much larger amounts of purified proteins are needed, the Ni-NTA Superflow 96 BioRobot Protocol provides a convenient high-throughput method for purification of approximately 100 µg of 6xHis-tagged protein per well in a 96-well format. Examples of purification and assay applications using both methods are shown and data on reproducibility and cross-contamination free processing are given.

P23-M

Detection of residual isopropyl-1-thio-beta-D-galactopyranoside in proteins by HPLC with pulsed amperometric detection.

X. Ji, L. Couch, A. Pennetti, K. Venkat, J. Mozdzanowski; SmithKline Beecham Pharmaceut., 709 Swedeland Road, King of Prussia, Pa 19406

Isopropyl-1-thio-beta-D-galactopyranoside (IPTG) is an inducer of the lac operon and may be used in fermentation processes for the production of protein biopharmaceuticals in E. coli. Use of this compound in the fermentation process creates the need for the determination of residual IPTG in the final product. Chemical and optical properties of IPTG (e.g. absorbance in the UV region) do not permit sensitive detection using traditional techniques. Hemiacetal group of galactose is substituted by a 2-thiopropyl group and thus eliminates an option for simple derivatization reactions leading to fluorescent derivatives. IPTG is a sulfur-containing galactoside, and therefore, properties of sulfur and sugar may be utilized for electrochemical detection. With this mode of detection, residual IPTG can be detected in the presence of high concentrations of protein. This required development of a modified HPLC separation method permitting the use of high concentrations of acetonitrile to remove protein from the column during a wash step. HPLC conditions developed for the separation are free from interference with protein and a so-called "oxygen dip" typically observed with the gold electrode. The waveform developed by William LaCourse for the Dionex electrochemical detector (ED-40) with the gold electrode can be used for the determination of IPTG at a concentration as low as 0.05 µM (12 ng/mL).

Reference:

William LaCourse, Determination of Isopropyl-beta-D-thiogalactoside (IPTG) by HPLC-Pulsed Electrochemical Detection. Pittcon-2000 presentation.

P24-T

Construction of a tagging system for subcellular localization of proteins encoded by newly discovered open reading frames.

W-J. Syu; Natl. Yang Ming Univ., 155 Sec 2, Li-Long St., Pai-Tao, Taipei, Taiwan 112, Taiwan

We have previously characterized a monoclonal antibody (SC1D7) that is directing to maltose-binding protein (MBP) of Escherichia coli and other closely related enteric bacteria. SC1D7 does not cross-react with proteins in eucaryotes and appears to be a highly specific tool in immunochemical analyses. To better map the epitope, we took advantage of an available plasmid pMAL-c2 that encodes E. coli MBP-coding sequence and constructed plasmids to express MBP fragments. A construct containing the N-terminal portion of MBP does not react with SC1D7 whereas a second construct expressing glutathione-S-transferase fused with the C-terminal half of MBP does react with SC1D7. To precisely define the epitope, random peptides displayed on M13 were used to react with SC1D7. Sequences of reactive peptides were aligned, and a consensus sequence of XDXRIPX was deduced. This sequence matches to MBP with an amino acid stretch of KDPRIAA. To consolidate the mapping result, a sequence encoding this epitope was inserted into an expression vector and the resulted recombinant protein did react with SC1D7. Thereafter, this epitope was incorporated into an eucaryotic expression plasmid that contains a previously defined hepatitis delta virus epitope for protein tagging. The so constructed two-epitope-tagging vector is useful in various molecular analyses. We demonstrated its usage in localization of a bacterial virulence factor in host cells. This vector should be applicable for high throughput characterization of new open reading frames found in genome sequencing.

P25-S

Truncated midkine (tMK): its molecular cloning and expression in Escherichia coli, and the detection of native tMK in Wilms' tumor by an anti-tMK monoclonal antibody.

S. Paul¹, W. Dansithong¹, T. Mitsumoto¹, Y. Asano¹, M. Kato², M. Kato², T. shinozawa¹; ¹Gunma Univ., Japan, 1-5-1 Tenjin-cho, Kiryu, Gunma 376-8515 Japan, ²Tottori Univ., Japan

Midkine (MK) is a heparin binding growth factor identified as a product of a retinoic acid-responsive gene; it is frequently expressed at high levels in many human carcinomas. Although the expression of the mRNA encoding truncated MK (tMK) in unique human cancer cells has been reported, the tMK polypeptide itself has not yet been identified. In order to clarify the biological role of tMK, recombinant tMK was expressed in Escherichia coli and purified. Purified recombinant tMK showed the same extent of proliferative activity for the Wilms' tumor (G401) cells as full length human MK. A mouse hybridoma cell line producing an IgG2b monoclonal antibody (mab) against this purified recombinant tMK was also established. This anti-tMK mab (MiStMK-V3) did not crossreact with synthetic full length (or c-half) human MK. A native tMK, showing the same apparent molecular mass as the recombinant tMK in SDS-PAGE, was identified in G401 cells using this mab. These results suggest that the structure of the recombinant tMK is same as that of native tMK. The expression of tMK in Wilms' tumor patient specimens was also detected in an immunohistochemical study using the anti-tMK mab. The usefulness of this mab (MiStMK-V3) for the detection of tMK in Wilms' tumor was demonstrated.

P26-M

Information management systems for molecular biology.

T.M. Smith, D. Campbell, R. Connelly, A. Leonard, S.G. Porter, J. Slagel; Geospiza, Inc., Seattle, 2442 NW Market St. #344, Seattle, WA 98107

Technological advances in molecular biology have made it possible for laboratories to generate unprecedented amounts of data. DNA sequencing, for example, has seen an increase in throughput of over 400-fold in recent years. The accelerated production of data has produced a corresponding need for better data management.

The increased need for data management is partly due to institutional changes. Many institutions have been able to economize by establishing core facilities which offer laboratory services to researchers on a per sample or project basis. Core facilities are cost-effective because they provide several researchers with access to expensive equipment and skilled personnel. Because these core facilities handle a variety of data types and a large number of samples from multiple sources, their information management needs provide a worthwhile model for laboratory information management systems (LIMS).

LIMS are collections of software, communication devices, and computers that acquire, store, analyze, and present data and information about laboratory samples and their processing. LIMS centralize data storage, automate data analysis, and provide quality assurance reports for process monitoring. The central component of a LIMS is a database with software interfaces for entering, viewing, and processing information.

The level of complexity needed in a LIMS depends on factors such as the mission of the laboratory, the number of samples to be processed, analysis requirements, and workflow complexity. LIMS development presents many challenges and may take years to complete. Most labs have no need for a custom LIMS and lack the wherewithal to create one, and thus will benefit from commercial LIMS

The general features of LIMS will be presented along with case studies of Geospiza's Finch™-Suite LIMS solution illustrating how a commercially available software system can be used to meet the information management needs of DNA sequencing facilities.

P27-T

Performance evaluation of a novel application for comparative DNA sequencing analysis and mutation detection.

C.A. Kosman, L. Johnston-Dow, H. Breu, G. Chappell, S. Kumar, B. Iasnopolski, B. Kshirsagar, P.A. Suri, R. Paul, G. Mason, D. Siu, E. Sword, M. Schoppe, V. Bawge; Applied Biosystems, 850 Lincoln Centre Drive, Foster City, CA 94404

Various software tools are available for the analysis and interpretation of de-novo DNA sequencing data. While such applications work well for sequencing for discovery many of the underlying assumptions in these tools are not valid when applied to comparative sequencing data.

We have developed a new practical application for comparative DNA sequencing analysis. The focus of this new application is the discovery or identification of sequence variants in a dataset of sequence derived from a single locus in various individuals. This tool allows for the rapid and accurate analysis and alignment of multiple sequence comparisons containing mixed base positions against a reference sequence. While NT-based, this software is compatible with data generated from all of Applied Biosystems sequencing platforms (ABI PRISM^® 373, 377, 310, 3100 and 3700).

This application uses procedures and algorithms better suited to the unique nature of comparative sequencing data. Specifically, the prior knowledge of a reference sequence available in many comparative sequencing projects is used to streamline and automate portions of the analysis. The software performs sequence analysis and aligns the analyzed sequence to a reference sequence. It also aligns imported sequences that contain variants. It will then analyze the compared sequences, provide protein translation and report the analysis in a convenient format.

Here, we present the results from the performance evaluation of this new software tool on several types of comparative sequencing datasets. For these analyses, this new application produced results concordant with those previously obtained using standard DNA analysis tools while providing substantial performance advantages over these tools. In all cases the analyses were completed in significantly less time and required less manual manipulation of the data. The innovative approaches employed in this tool result in less user intervention needed to achieve a better level of mutation detection.

P28-S

A Web interface for BioLIMS System--data and user management for sequencing core facility.

L. Liu, L. Roinishvili; Univ. of Illinois, Urbana, 330 ERML, 1201 W. Gregory Dr., Urbana, IL 61801

Sequencing core facility serves many customers. Samples in a sequencing core facility are very heterogeneous although the volume of samples from each customer is relatively small. Data and user management can be a challenge task. The BioLIMS database system from PE Bioinformatics provides a smooth data flow from sequencer to database. However, it really lacks of friendly user interface for both core facility and customers. To provide an easy management system for the core facility at the W. M. Keck Center for Comparative and Functional Genomics, we implemented a web interface for BioLIMS system. This interface serves two purposes in the core facility. One is for core facility staff to manage customers and their data. The other is for customers to view and retrieve their own data from the web. In customer management, we allow super user and sub user relationship. Super user will have access to all the data of his or her sub users. In data management, a customer can only view and retrieve his or her own data unless others give him or her privilege to view their data. The most convenient feature is web based. Customers can access their data from anywhere at anytime.

P29-M

Protein/DNA technology DSL version 2000 is a set of shareware protocols for sample and data management.

J. Medalle, M. Randesi, B. Imai; Rockefeller Univ., 1230 York Ave., Box 105, New York, NY 10021

Physical sample and data management are major concerns for a DNA sequencing core facility. Especially one with a small staff and limited space are affected the most. Samples can overtake the limited freezer space. Paper trail can overfill office and lab spaces. Both can cause a quick turnover of DNA sequencing staff. DNA Sequencing Lab's (DSL) version 2000 objectives are to alleviate repetitive computer tasks for DNA Sequencing staff and eliminate physical paper trail. DSL 2000 is a set of protocols for sample and data management. Its major components are comprised of online submittal forms at http://protein13-pc.rockefeller.edu/, scripts from Apple Script, and standardized forms and macros from Microsoft Excel 98, and Outlook Express 5. The DSL funnels the online sample submissions into sample statements for each investigator. This is a file containing sample IDs, names, account numbers, primer info, template info, etc. DSL also transforms the web submissions into logs for prep and for slab or capillary sequencing reactions. These logs direct the staff on how to process incoming samples. The set of protocols streamlines the sample flow from the investigator to the machine as well as the data flow from machine back to the investigator. DSL's protocols are "hands on" to help the staff create e-mails for investigator notifications and "a click of a button" to compress and to transfer investigator's sample statements or data files into their web account. DSL 2000 is a freeware application and can be integrated into other small core facilities with a limited budget.

P30-T

A laboratory information management system for a small DNA sequencing core facility.

M.J. Miller; NCI, NIH, Bldg 37, Rm 3C28, MSC 4255, Bethesda, MD 20817-4255

The DNA Sequencing MiniCore facility of the NCI's Division of Basic Sciences currently services over 300 investigators. The facility is designed to support small scale, short-term sequencing. In the past year we ran over 25,000 samples with an average turnover time of less than one day. While this represents a 60% increase in samples over the year before, it is still a relatively modest operation.

Keeping all these users and their data organized, as well as minimizing the amount of paperwork involved is a major concern. Although several Laboratory Information Management (LIM) System software packages exist, they are often too expensive or too inflexible for a small operation such as ours.

I describe here a LIM system built using components of the Microsoft Office software package. Users enter sample information data into an Internet form and this data is automatically entered into the database after inspection by the MiniCore staff. Programs built into the database determine which sample sets can be run together on the same gel, and what parameters (such as which virtual filter and which dye/primer set settings) should be used with each sample. Samples are assigned to a particular gel and the "sample sheet" for that gel is automatically generated by the database. By keeping track of when samples are submitted, how many samples there are, and how many basepairs of read the user requests, the database aids the facility's operators in determining sample priority. An email message is automatically sent when samples have been processed and data deposited in the user's data-destination directory. The database also keeps track of charges and automatically sends this data to the NIH's accounting system. The facility thus runs in a virtually "paperless" environment. There are no hardcopy forms to fill out. All information is maintained by the computer system.

P31-S

The analysis of complex tryptic peptide mixtures by multidimensional LC-MS/MS on a hybrid quadrupole orthogonal acceleration time-of flight (Q-TOF) mass spectrometer.

A. Millar¹, C. Hughes¹, T. Andresson², T. Hemesath², J.I. Langridge¹; ¹Micromass UK Ltd., Floats Road, Wythenshawe, Manchester M23 9LZ, United Kingdom, ²deCODE genetics Inc., Reykjavik

Advances in both HPLC and mass spectrometry instrumentation have allowed the analysis of protein complexes which have not been separated on a two dimensional gel. These experiments involve separation of the complex digest mixture by microcapillary liquid chromatography connected to an instrument capable of data directed switching between the MS and MS/MS modes. Protein identification is then achieved via databank searching of the ESI-MS/MS, providing qualitative information on the proteins that are present. Hundreds of MS/MS spectra can be acquired in a fully automated fashion, resulting in the identification of significant numbers of proteins, including low copy number proteins, from a single LC-MS/MS experiment¹.

If, however, a complex protein mixture is to be investigated then a fractionation step prior to separation of the peptides on the basis of their hydrophobicity is advantageous. We have, therefore, adopted a 2D LC-MS/MS approach using a capillary LC system (CapLC) operating at nanoliter per min flow rates coupled to a Q-Tof 2 mass spectrometer. By placing a strong cation exchange (SCX) cartridge followed by a C18 trap cartridge it is possible to pre-fractionate the peptides before separation on an analytical C18 column. After loading the sample, discreet fractions are sequentially eluted from the cation exchange cartridge using a salt step gradient; the eluted peptides are then retained on the trapping C18 cartridge whilst they are desalted. Finally the peptides are eluted from the C18 pre-column, at 200 nL/min, onto a 75 µM ID X 10 cm Waters Symmetry analytical column for separation and elution into the mass spectrometer.

This analytical approach will be discussed with examples where this methodology has been used for the analysis of standard protein mixtures and for the analysis of cell lysates and sub-cellular fractions.

1. Yates et al., Nature Biotechnology (1999);17, (7), 676-682.

P32-M

Identification of in vivo phosphorylation sites in Drosophila armadillo by tandem mass spectrometry.

C.S. Raska, R.M. Pope, D. Rubenstein; Univ. of North Carolina at Chapel Hill, 4 Casabelle Ct, Durham, NC 27713

Phosphorylation is one of the most important reversible modifications of eukaryotic proteins. Often, proteins which are associated with uncontrolled cell growth, and ultimately cancer show an anomaly in their phosphorylation/dephosphorylation pathways. In humans, a protein, beta-catenin, plays a central role in the development, organization, and regulation of epithelial tissues. Aberrant regulation of beta-catenin is associated with malignancies. For example, alterations in beta-catenin gene structure have been identified in colorectal and breast carcinoma, and in a large number of melanoma cell lines. Specifically, beta-catenin signaling function is constitutively active in many melanomas due to mutations that remove phosphoresidues at the amino terminus of the protein. Armadillo, a protein found in Drosophila melanogaster, is 71% identical to beta-catenin at the amino acid level. Alpha-catenin from Drosophila and from human keratinocytes binds to armadillo and beta-catenin, respectively, and this binding has been found to be influenced by phosphorylation. Thus, we have been using armadillo as a model system to study beta-catenin regulation. To further define the regulatory function of phosphorylation, we are using mass spectrometry to map post-translationally modified amino acids in armadillo. Armadillo was isolated by immuno-affinity and ion exchange chromatography. 1-D SDS-PAGE, and in-gel tryptic digestion were performed to produce a mass spectral fingerprint on a triple quadrupole instrument using nanoelectrospray. MS scans confirmed peptides matching both armadillo and alpha-catenin from one gel spot. Neutral loss scans identified potentially phosphorylated peptides. MS/MS generated enough sequence coverage to specifically identify phosphorylated residues. We note that phosphoresidues are located within the region where binding sites for cadherin, dTCF, and dAPC are located. We plan mutation studies to further elucidate the role of phosphorylation in these systems.

P33-T

A targeted approach for more sensitive and accurate protein identification.

D.K. Leung, M.J. Horn; BioMolecular Technol., Inc., 525F Del Rey Ave., Sunnyvale, CA 94086

Selective isolation of peptides containing low abundance amino acids such as tryptophan, offers a direct way to simplify protein digest mixtures prior to mass spectrometric analysis. This approach provides benefits for both in-gel digests and digests of protein mixtures.

To this end a solid phase reagent capable of selectively binding to tryptophan has been developed. This solid support material binds covalently to tryptophan or tryptophan containing peptides, allowing for their selective separation from mixtures of peptides and proteins. Bound peptides can be released from the solid-support by a reducing reagent to give analytical samples for further analysis (LC, MS, etc.). The specificity of the binding chemistry coupled with mass spectrometric analysis by MALDI affords improved accuracy in data base search and in protein identification.

References

1. Shechter, Y., Rubinstein, M. and Patchornik, A. (1976), Biochemical and Biophysical Research Comm. {4} (70), 1257.

2. Early, S. L., Magil, S. G., Novak, C. and Horn, M. J. (1989), Techniques in Protein Chemistry, Academic Press, Inc., 439.

3. McEldoon, W., and Horn, M.J. (2000), 14th Protein Society Symposium, {August}.

P34-S

Approaches to validating SEQUEST database search results.

R.E. Moore, M.K. Young, T.D. Lee; Beckman Res. Inst., City of Hope, 1450 E. Duarte Rd., Duarte, CA 91010

Tandem mass spectrometry and database searching is a very powerful approach to protein identification. Unfortunately, the wealth of data that makes the technique so powerful also requires extensive analyst time for data reduction and validation of search results. When analyzing complex samples, it can easily take more time to validate and reduce the data than it did to generate it. This time requirement is a serious problem because the technique is otherwise well suited to high throughput analysis.

Two aspects of result validation are analyzed: validation of search results for individual tandem mass spectra and validation of composite results for sets of individual spectral matches. Statistical analysis and experiments with falsified databases demonstrate that there is a significant chance of generating multiple incorrect matches to the same protein when examining large data sets. This makes it imperative to validate the individual spectral matches, as simple reliance on multiplicity of matches to a single protein does not guarantee a correct result.

Several approaches to validating individual spectral matches were tried and compared to the existing standard of manually comparing actual and predicted spectra. The most successful approach was to search each spectrum twice, once using tryptic specificity and once using no enzyme specificity, and correlate the results. When both searches generate the same result, that result is almost always validated by manual examination. When the non-specific search generated a higher cross-correlation score the result of the search using tryptic cleavage was almost never manually validated. Surprisingly, there were a large number of searches in which the search using tryptic cleavage resulted in a higher cross-correlation score than the search using non-specific cleavage. This third group yielded a moderate number of spectra that were manually validated.

P35-M

Guanylation of lysines as a means to enhance confidence of protein identification in proteomics.

S. Krishnan¹, M. Lin², K. Waddell¹; ¹Applied Biosystems, Foster City, CA, 500 Old Connecticut Path, Framingham, MA 01701, ²Applied Biosystems, Framingham, MA

The completion of several genomes, including that of the human, has shifted the emphasis to determining the proteome counterpart of biological systems. The current approach for such a proteomic analysis is the mass fingerprinting of tryptic digests of proteins (generated by ingel digest of 1 or 2-D gel spots or of chromatographic fractions) using MALDI-Tof mass spectrometry followed by database searching of the masses so obtained. It has recently been noticed that the arginine containing peptides ionize better under MALDI conditions compared to lysine containing peptides. We present here an approach to use this as an advantage in enhancing confidence of the database search matches by modifying the lysines to homoarginines. The increase in confidence is achieved due to increase in the number of peptides seen in the mass spectrum resulting from the guanylation of lysine e-amino group (addition of 42 Da) and also the fact that such a modification indicates the lysine containing peptides in the tryptic digest mix. Tryptic digests of BSA, Enolase, alpha lactalbumin, and cytochrome C were generated by routine procedures and mixed in various combinations. The digest mixtures were than analyzed by MALDI-Tof mass spectrometry followed by database searching of the peptide masses for identification of proteins. A portion of the digest mixture was then reacted with O-methyl isourea sulfate to convert the lysines to homoarginines. The modified digest mixtures were analyzed by MALDI-Tof mass spectrometry followed by database searching. The analysis before and after the modification helped identify the lysine containing peptides in the digest and hence a more comprehensive identification of the protein mixture.

P36-T

Automated analysis of single nucleotide polymorphism using mass spectrometry.

M.S. Minkoff¹, P. Ross¹, L. Hall¹, R. Jones², D. Ledman¹, L. Haff¹; ¹Applied Biosystems, 500 Old Connecticut Path, Framingham, MA 01701, ²Inst. for Child Hlth.

Single nucleotide polymorphisms (SNPs) are increasingly employed as genetic markers for associated studies of many inherited diseases and traits. It is apparent that approximately 3 million putative SNPs will be found in the Human Genome. In order to tackle the task of validating the significance of any of these putative SNPs the chemistry and analytical platforms must meet demanding throughput and cost requirements. The SNP assay developed to meet these criteria incorporates the use of simple, inexpensive primers and unlabeled nucleotides for single base extensions, thus keeping down the cost of analysis. Multiplexing the analysis of SNPs enhances affordability of genotyping studies as well as significantly improving throughput. The use of high throughput automated Mass Spectrometry with integrated data acquisition, genotyping, sample handling software package simplifies and speeds-up the task. In addition the top-level software application package tracks sample information throughout the workflow and stores all information in a central database. Data in support of multiplexed, automated sample handling, acquisition, analysis and reporting is presented.

P37-S

Data-directed real-time instrument feedback and control: creation of dynamic peptide include and exclude lists from on-the-fly databank searching.

A. Millar, J.B. Hoyes, R.A. Carruthers, C. Jones, A. Millar, C. Hughes, S. Leicester, J.I. Langridge; Micromass UK Ltd., Floats Road, Wythenshawe, Manchester M23 9LZ, United Kingdom

An alternative approach to 2D gel electrophoresis for the qualitative analysis of complex protein mixtures is the use of tryptic digestion followed by electrospray LC-MS/MS. This approach has been shown to increase the dynamic range of protein identification and identify low copy number proteins. However there is often a large degree of redundant sequence information acquired, as in theory one peptide MS/MS spectrum is sufficient to identify a protein from a sequence databank. If a protein identification is obtained from a databank search of an MS/MS spectrum, excluding the rest of the theoretical tryptic peptides during an LC-MS/MS experiment may allow deeper "mining" into the protein complex being studied.

We have introduced a new search engine capable of matching a tryptic peptide from the Swissprot/TrEMBL databank to an MS/MS spectrum in one second. Based upon these results, we are able to generate dynamic include or exclude lists, based upon the theoretical tryptic peptides from the identified protein. These can be passed to the acquisition software of our Q-Tof mass spectrometer in real time. Thus, we are able to automatically steer the Q-Tof during acquisition to select and switch to the MS/MS mode only on those peaks that meet the modified selection criteria. Therefore precursor ions that belong to a protein already identified during acquisition can be avoided. This exclusion list is based upon m/z, charge state and a user defined mass tolerance. Mass measurement on the Q-Tof 2 mass spectrometer is typically better than 10ppm and therefore a tight mass tolerance can be selected, making the exclude list extremely specific.

To illustrate this methodology we present examples where Q-Tof data acquisition has been directed based upon the results from a databank search. This data will be compared and contrasted to data acquired in the normal automated LC-MS/MS mode.

P38-M

Routine identification of large (>5 kDa) peptides derived from biological sources using an ESI-quadrupol-TOF mass spectrometer.

M. Kellmann, S. Neitz, M. Juergens; BioVisioN GmbH & Co KG, Hannover, Feodor-Lynen-Str. 5, Hannover, Lower Saxony 30625, Germany

Investigating the peptide portion of a proteom results in a multitude of peptides/small proteins in the mass range from 5 to 15 kDa [1]. We developed a routine procedure for direct identification exploiting the relatively high parent and daughter ion resolution (>7500) and mass accuracy (<50 ppm) of a hybrid ESI-quadrupol-time-of-flight mass spectrometer. Data processing was made with automated charge state deconvolution based on Zhang and Marshall´s Zscore algorithm [2] and a deisotoping algorithm included in the Voyager 5.0 software from Applied Biosystems. Database searching was performed by the MASCOT search engine [3].

We will illustrate that this procedure will lead to a fast and reliable identification of several large peptides in the mass range from 5 to 15 kDa derived from blood filtrate.

[1] Kellmann, M, et al., Proceedings of the 48th Conference on Mass Spectrometry and Allied Topics, 2000, Long Beach

[2] Zhang, Z and Marshall, AG. A universal algorithm for fast and automated charge state deconvolution of electrospray mass-to-charge ratio spectra. J Am Soc Mass Spectrom 9(3): 225-33 (1998).

[3] Perkins, DN, Pappin, DJ, Creasy, DM and Cottrell, JS, Probability-based protein identification by searching sequence databases using mass spectrometry data. Electrophoresis 20(18) 3551-67 (1999).

P39-T

Automated software for identification and relative quantification of differentially expressed proteins from isotope coded affinity tag LC-MS data.

A. Millar, P. Young, R. O'Malley, A. Millar, J.I. Langridge; Micromass UK Ltd., Floats Road, Wythenshawe, Manchester M23 9LZ, United Kingdom

Whilst LC-MS/MS has been utilised for the identification of proteins from complexes and cell lysates (qualitative proteomics), the quantitative study of gene expression using differential display has until recently been the preserve of a 2D gel based proteomic experiment. However, recently a great deal of interest has been generated on the use of isotope coded affinity tags (ICAT) 1 for the quantitative study of gene expression at the proteome level.

The technique is based upon chemically modifying the cysteine residues of proteins isolated from cells in two different states with light and heavy isotopically labeled reagents. The two cell states are then combined, digested with trypsin and the cysteine containing peptides preferentially selected by binding to an avidin column, prior to analysis by mass spectrometry. The eluent from this column is then analysed by capillary LC ESI-MS/MS. Interrogation of the eluting peptides by tandem mass spectrometry and databank searching results in the identification of the associated protein.

We describe how ICAT data analysis has been automated within a software environment. The MS and MSMS data acquired using the QTof instrument are processed and analysed using a new algorithm which recognises related isotope clusters and quantifies their relative intensities. Based on a user defined ratio threshold the software will automatically carry out an LC-MS/MS experiment and databank search in a client-server mode and provide a report of the identified proteins and their expression ratio in the two cell states.

1. Gygi et al. Nature Biotechnology (1999) 17, 994-999.

P40-S

New method for preparation of DNA for MALDI-MS analysis.

I.P. Smirnov, L.A. Haff, P.L. Ross, L.R. Hall, D.W. Ledman; Applied Biosystems, 500 Old Connecticut Path, Framingham, MA 01701

In recent years, matrix assisted laser desorption ionization mass spectrometry (MALDI MS) became one of the main analytical tools for short DNA fragments and synthetic oligonucleotides. The application of mass spectrometry for genetic analysis (such as SNP typing) offers, compared to other methods, higher sensitivity, higher accuracy and lower costs by removing the necessity for fluorescent- labeled primers. Still, the susceptibility of MALDI-MS to the presence of salts of alkali metals requires careful and often tedious desalting procedures which complicates and slows the throughput of the MS based methods. To overcome that obstacle we have developed a novel approach to sample preparation. The idea is to extract DNA out of the solution on a solid surface coated with an anion exchange polymer (e.g. polyethyleneimine). The observed binding is strong enough to remove and concentrate DNAs from buffer solutions and hold them on the plate through extensive washing procedures if necessary. The properties of anion exchange films make possible to perform purification, desalting and concentration in a single fast step. After the DNA is immobilized on the surface and supernatant solution is removed, subsequent addition of MALDI matrix releases material from the surface, which co-crystallizes with matrix. The MS analysis can then be performed directly from that support. Such surfaces may be obtained upon glass, plastic, or metal MALDI plates. Analysis of oligonucleotides and complex multiplexed SNP typing reactions (Sequazyme-PinPoint) performed by this method shows greatly improved sensitivity and excellent resolution for wide range of DNA lengths, together with high tolerance to various buffers components, such as alkali metals and surfactants. The importance and role of different factors have been studied, such as composition of ion exchange film and compatibility of the MALDI matrix in order to obtain the best performance.

P41-M

Review of gel- and column-based separation technologies for the identification of yeast proteins using MS/MS.

V.C. Wasinger, G.L. Corthals; Garvan Inst. of Med. Res., 384 Victoria St., Sydney, NSW 2010, Australia

Proteome analysis involves the identification and quantitation of expressed proteins by a given cell type, tissue or organism and most commonly involves 2-dimensional electrophoresis (2-DE) followed by mass spectrometry (MS) for protein separation and identification. While 2-DE is unequalled in its ability to separate and resolve thousands of proteins in complex solutions, we have recently shown that many proteins of biological significance lay beyond the detection limits of 2-DE. The application of 2-D gels becomes ineffective when proteins co-migrate to the same grid coordinates, merge to become one large spot or are not displayed because their concentration is simply too low for visualisation and subsequent biochemical MS analysis.

Two important biological aspects prevent us from fully exploiting the power of the 2-DE technology: 1) Firstly, protein expression exists for up to 12 orders of magnitude (in serum). This broad range of expression critically limits the current 2-DE approach, as at best only 5 orders of magnitude difference in expression can be displayed; 2) "functional" proteins operate in association with other proteins; such that research must be directed toward the analysis of multiply interacting proteins or functional modules that regulate complex biological networks and pathways.

Alternative fractionation methods such as chromatography exist and allow proteins/peptides to be separated based on similar (charge, size) or alternative (affinity, hydrophobicity) properties to 2-DE. Used in series with MS, these technologies have the potential to fractionate large numbers of proteins and peptides to enable the comprehensive, comparative and relational analysis of proteins. Yeast is an ideal choice for this study as both genome, proteome and transcriptome data is available for evaluation.

A comparison of the separation potential based on protein numbers, protein classes and properties are described using 2-D gel electrophoresis, size exclusion and cation exchange chromatography followed by protein identification using ESI-MS/MS.

P42-T

Quantitative analysis of tumor antigens by mass spectrometry.

A. Kishiyama, D. Arnott; Genentech, 1 DNA Way, South San Francisco, CA 94080

Both genomics and proteomics are useful for comparing gene expression patterns. The discovery of receptors overexpressed on tumor cells has lead to effective cancer therapies, for example. Techniques such as DNA arrays can be used to measure mRNA levels with great sensitivity and speed. More detailed information can be obtained through 2D gels, western blotting, and immunohistochemistry, but these protein-based approaches are time-consuming, prone to biases, or require generating antibodies. A method has therefore been developed, called the mass western experiment in analogy to the western blot, to better bridge genomics and proteomics.

In this experiment, a variation on the ICAT experiment described by Gygi et al. (Nature Biotech. 1999 v.17 p.994), specific proteins (such as those found to be of interest from DNA array experiments) are detected and compared between samples. Proteins extracted from two samples are labeled with a custom ICAT reagent. The samples are mixed, digested, and the labeled peptides collected. LC-MS/MS is performed on an ion trap instrument; anticipated tryptic peptides from the protein of interest are continuously subjected to CID. Heavy and light ICAT-labeled peptides are simultaneously trapped and fragmented. Both identification and quantitation is thus obtained in one experiment.

This approach has been validated by the comparison of cell lines expressing known tumor antigens in different amounts. Overexpression of the receptor Her-2 in breast cancer cell lines was shown, and by factors in agreement with other measurements. Other potential tumor antigens have likewise been detected.

P43-S

Characterization of differential protein expression between wild type and Rce1 knockout mouse embryonic fibroblasts using 2-D SDS PAGE and MALDI TOF mass spectrometry.

S.C. Hall¹, D. Smith-Beckerman², M. Lobo², S.G. Young³; ¹Applied Biosystems, 850 Lincoln Centre Drive, Foster City, CA 94404, ²San Francisco State Univ., ³UCSF

Ras proteins play an important role in transmitting growth signals from membrane receptors to the nucleus, triggering the transcription of genes involved in cellular proliferation. Ras proteins are membrane bound, guanine nucleotide-binding proteins with GTP-ase activity. Many human cancers contain mutationally activated Ras proteins that transmit growth signals in an uncontrolled manner, triggering neoplastic transformation and uncontrolled cellular proliferation. The proper intracellular location of the Ras proteins depends on a series of posttranslational modifications: isoprenylation of a C-terminal cysteine, endoproteolytic release of the carboxyl-terminal three amino acids, and methyl esterification of the carboxyl-terminal isoprenylcysteine. Each of these processing steps represents a potential target for cancer therapy. The endoproteolytic cleavage step is carried out by the product of the Rce1 gene. Recently, the Rce1 gene was inactivated in mice, completely blocking the endoproteolytic processing of the Ras proteins. We used 2D-PAGE to assess differences in protein expression in primary embryonic fibroblasts from wild-type and Rce1 knockout mice. First, we wanted to detect substrates for Rce1. Altered electrophoretic mobility of individual protein spots on the gel indicated the possibility of aberrant post-translational processing. Second, we wanted to detect proteins whose expression level was affected by the Rce1 knockout mutation. MALDI TOF mass spectrometry was used to generate peptide mass fingerprints to identify several proteins having identical molecular weight, pI, and relative abundance in both wild-type and Rce1 knockout mice. Additional confirmation of the identities of these proteins was obtained by performing post-source decay analysis on selected tryptic peptides. It was important to identify these "marker" proteins as they will be used as migration reference points when comparing future 2D-gel separations. Furthermore, they permitted optimization of protocols for the MS analysis of Ras proteins from wild-type and Rce1 knockout fibroblasts.

P44-M

Intelligent data acquisition and automated sample analysis via orthogonal MALDI- QqTOF, a new tool for protein identification.

C.M. Lock; MDS-Sciex, 71 Four Valley Drive, Concord, Ontario L4K 4V8, Canada

The application of a novel UV-MALDI ionisation source coupled to an Applied Biosystems/MDS-Sciex QSTAR Pulsar QqTof mass spectrometer for protein sequencing and identification is described. The coupling of these two devices enables collision induced dissociation spectra of singly charged MADLI ions to be generated, with all the associated QqTof benefits of high mass accuracy and resolution. The inherent pulsed nature of the o-MALDI source is converted into a pseudo continuous beam of ions by collisional cooling in the Q0 region. The o-MALDI source is completely decoupled from and has no influence on the orthogonal Tof analyser.

High mass accuracy and resolution is thus maintained simultaneously over the full mass range when switching between MS and MS/MS modes as opposed to conventional MALDI post source decay experiments.

The application of the technique to the analysis of low femtomole unseparated protein tryptic digests is demonstrated using an automated data acquisition approach. The software developed enables the intelligent acquisition of data from sample plates with minimal user intervention.

The high speed data acquisition capabilities of the o-MALDI source in combination with the high performance of the QqTof offers unique possibilities for rapid identification of proteins. Rapid analysis and identification of proteins via a peptide-mass fingerprinting approach and MS/MS sequence information will be shown.

P45-T

Identification of changes in protein expression in Deinoccocus radiodurans using isotope-coded affinity tags.

E.A. Panisko¹, T.P. Conrads¹, T. Veenstra¹, L. Pasa-tolic¹, G.A. Anderson¹, R. Aebersold², R.D. Smith¹; ¹Pacific Northwest Natl. Lab., P.O. Box 999, K8-98, Richland, WA 99352, ²Univ. of Washington

The isotope-coded affinity tag (ICAT™) strategy was used to identify the effects of irradiation on relative protein abundances in the highly radioresistant organism, Deinococcus radiodurans (D. radiodurans). The ICAT™ strategy uses two distinct isotopic versions of a cysteine-specific reagent; a "light" version and a "heavy" version in which eight hydrogen atoms are substituted for deuterium atoms. Proteome samples extracted from the organisms just prior to, as well as 30 min. and 3 hours after treatment with ionizing radiation were labeled with either the heavy or light isotopic version of the ICAT™ reagent. After mixing the samples in various combinations the cysteine containing polypeptides (Cys-polypeptides) were extracted using immobilized avidin chromatography. The Cys-polypeptides were initially identified by capillary reverse phase liquid chromatography (LC) using a conventional mass spectrometer (MS) operating in the tandem mass spectrometry (MS/MS) mode. The masses of the identified peptides were subsequently identified in a capillary LC separation coupled on-line with Fourier transform ion cyclotron resonance (FTICR) MS. Changes in relative protein expression were measured using the results obtained by LC/FTICR. The results show that most proteins undergo a decrease in expression when D. radiodurans is subjected to ionizing radiation.

P46-S

Quantification of neurosteroids using NCI GC/MS.

R.L. Fitzgerald; VA Med. Ctr., UCSD, VAMC-113, 3350 La Jolla Village Dr., San Diego, CA 92161

Electron capture negative chemical ionization (NCI) is one of the most sensitive ionization techniques available and is especially well suited for quantitative analysis of target compounds in biological extracts. For the analysis of small molecules, NCI is often several orders of magnitude more sensitive than traditional techniques such as electron ionization or positive chemical ionization. There are several important prerequisites for performing quantitative analysis using NCI, including an electronegative functional group, stable isotopic internal standards, and good method validation. We synthesized deuterium labeled analogs of neurosteroids and developed a NCI GC/MS method for quantification of neurosteroids in biological samples using isotope dilution.

Neurosteroids have distinct neurotransmitter mediated effects and consequently it is important to be able to identify and quantify individual compounds. Previously, the determination of neurosteroids in biological matrices involved complicated purification protocols or did not use appropriate internal standards. We added deuterium-labeled internal standards to brain (100 mg of cortex homogenate) or plasma (300 µL). Samples were homogenized in methanol, centrifuged and diluted to contain 5% methanol and then applied to C-18 columns. After washing the column with methanol/water (50/50), steroids were eluted with methanol. Following evaporation, steroids were converted to pentafluorobenzyl oxime/trimethylsilyl ether derivatives. The extracts were analyzed using SIM. The present method allows simultaneous quantification of pg amounts (100 pg in 300 µL of plasma and 250 pg in 100 mg of brain tissue) of neurosteroids and will be helpful in elucidating the role of neurosteroids in health and disease.

P47-M

Use of stable isotope amino acid labels to simplify MS/MS peptide analysis.

S.J. Berger, S-W. Lee, Y. Shen, G.A. Anderson, R.D. Smith; Pacific Northwest Natl. Lab., 902 Battelle Blvd. Box 999 MS: K8-98, Richland, WA 99352

Tandem mass spectrometry (MS/MS) is a primary analysis tool for the identification of peptides from proteolytic sample digests. When coupled with online separations, one or more peptides from a single 2D-gel spot digest can uniquely identify a given protein, or more complex digests can yield a global view of the proteome. The identification of individual fragment ions from a parent peptide is indirect, and the resulting peptide identification is strictly correlative. The ability to add sequence and composition constraints to assign peptide identity should significantly narrow the possible assignments of ions, and simplify interpretation of MS/MS spectra. Here we describe an application of stable isotope labeling in conjunction with MS/MS analysis of proteolytic digests that permit significant improvements assigning ions in the resultant MS/MS spectrum. Auxotrophic cells grown in parallel in the presence of a labeled or unlabeled amino acid are combined prior to cell disruption. Peptides derived from proteolytic digests generate pairs of peaks separated by a known mass difference during an initial LC/MS scan. These identified peptide pairs are isolated, and analyzed by LC/MS/MS in subsequent scans. When using a protease that cuts adjacent to the labeled amino acid, the paired/unpaired daughter ion pattern permits simplified identification of the parent peptide.

This work was supported by OBER (U.S. DOE) and PNNL Laboratory Directed R&D. Battelle Memorial Institute operates PNNL for the U.S. DOE under Contract DE-AC06-76RLO 1830.

P48-T

A comprehensive proteomic analysis of human cilia using nanoscale capillary LC/MS/MS.

M. Moyer¹, K. Blackburn¹, W. Burkhart¹, A. Moseley¹, L. Ostrowski², R. Boucher²; ¹Glaxo Wellcome, 5 Moore Drive, Research Triangle Park, NC 27709, ²Univ. of North Carolina

Ciliated cells play an integral role in the defense mechanisms of the respiratory system. By the coordinated beating of their cilia they provide the force necessary to clear potentially harmful material from the airways. In order to better understand the protein composition of cilia, human cilia were subjected to a comprehensive proteomic analysis. Cilia were isolated from cultures of airway epithelial cells and component proteins separated by 1D or 2D gel electrophoresis. Bands or spots were excised, subjected to in-gel proteolytic digestion, and component proteins identified by nanoscale capillary LC/MS/MS. Alternatively, proteolytic digests of intact cilia were analyzed directly by nanoscale capillary LC/MS/MS with or without the use of cysteine-specific affinity tags. Data will be presented on proteins identified as components of cilia as well as details of the analytical methodologies.

P49-S

4-(4-Hydroxystyryl)pyridine, a rationally designed negative ion matrix for MALDI-TOF mass spectrometry.

J.C. Lopez, P.A. Liddell, J.W. Lippert, D.C. Brune; Arizona State Univ., PO Box 871604, Tempe, AZ 85287-1604

To date, most matrices for MALDI-TOF mass spectrometry have been discovered by trial and error experiments, and through broad investigation of compounds chemically related to known matrix molecules. The most widely used matrices have been used to analyze proteins and other analytes in the positive ion mode. We chose to work with negative ion matrices because this area has received less attention, and because we occasionally analyze samples for which a good negative ion matrix is required. In this investigation, we synthesized and studied 4-(4-hydroxystyryl)pyridine. This compound resembles those derived from 4-hydroxycinnamic acid in having an electron-withdrawing moiety (in this case a pyridyl group) separated from a 4-hydroxyphenyl moiety by a vinyl group. In the electronically excited state, charge density delocalized across the vinyl group onto the pyridyl group will make the latter more strongly basic, favoring abstraction of protons from both the analyte and other matrix molecules. Preliminary experiments on Fmoc and other modified amino acids have shown that this matrix works as well as 3-aminoquinoline, one of the best negative ion matrices reported so far. Further comparisons involving structurally related matrix molecules are in progress and will be reported. A characteristic for negative ion matrices is a shift to longer wavelengths under acidic conditions.

P50-M

LC-MS of oligonucleotides using polystyrene supports and API-TOF mass spectrometry.

D.H. Hawke¹, J.S. Nelson², R. Vinayak², W. Xiao³, P.J. Oefner³; ¹Applied Biosystems, 850 Lincoln Ctr. Dr., Foster City, CA 94404, ²Applied Biosystems, ³Stanford Genome Res. Ctr.

Oligonucleotides may be analyzed by either Maldi or electrospray methods after suitable desalting of samples. Maldi methods are convenient and offer high-speed analysis, while electrospray methods can be directly coupled to HPLC separations. A classical application for these methods is oligonucleotide synthesis, in this case there is usually enough material available that sensitivity is not a major problem. LCMS (1) is advantagous if UV detection is required, and desalting is effectively performed on-line. A newer area is that of genotyping. In this application separation may be required prior to analysis and current ion-pairing methods have remarkable separating power for these molecules (2). We have coupled this separation technology to a high-resolution API-TOF MS system (Mariner) to give both high resolution separations and high mass accuracy measurements of oligonucleotides and labeled oligos.

1. Huber C.G., Krajete A., Anal. Chem. (1999), 71:3730-3739.

2. Underhill, P.A., L. Jin, A.A. Lin, S.Q. Mehdi, T. Jenkins, D. Vollrath, R.W. Davis, L.L. Cavalli-Sforza and P.J. Oefner. Genome Res. (1997), 7:996-1005.

P51-T

Advances in quantitation and identification of proteins using isotope-coded affinity tags and LC/MS/MS using information-dependent acquisition.

T.A. Settineri, C.L. Hunter, L. Nuwaysir, A. Patel, A. Tomaney, B. Halpern; Applied Biosystems, 850 Lincoln Ctre. Dr., Foster City, CA 94404

Recently, Aebersold et al. introduced a method in which proteins are labeled with an isotope-coded affinity tag (ICAT), digested, and the resulting labeled peptides separated and analyzed by mass spectrometry. The advantage to this technique is that the ICAT approach precludes the use of 2D gels while allowing complete automation of the entire process. In addition, the ICAT technique is generally applicable and can be applied to the global analysis of protein expression. In this study we investigate the ICAT approach using an API-QSTAR (hybrid quadrupole time-of-flight) mass spectrometer fitted with a nano-electrospray source, ICAT reagents synthesized in-house, and novel data reduction algorithms to both quantitate protein expression and identify proteins from MS/MS data acquired by using information dependent acquisition (IDA)

This approach will be demonstrated for the analysis of proteins arising from various cell lines including E. coli and Arabidopsis treated in different ways. After treatment with ICAT reagent and purification, samples are analyzed by LC/MS/MS using a QSTAR mass spectrometer using information dependent acquisition (IDA). Two new algorithms developed to identify and quantitate ICAT expression pairs are applied to the MS scans while a third new algorithm is applied to the MS/MS data to identify proteins. The first two algorithms cluster the data based on ICAT fragments, and optimally collapse adjacent spectra for maximum signal to noise. The third algorithm then identifies the protein using the MS/MS fragment ion data to perform a database search.

P52-S

High-throughput SNP genotyping by MALDI-TOF MS.

M. Kostrzewa¹, T. Fröhlich¹, T. Wenzel¹, C. Franke¹, W. Pusch², K-O. Kräuter², Y. Stalgies²; ¹Bruker Saxonia Analytik GmbH, Permoserstrasse 15, Leipzig, Saxonia D-04318 Germany, ²Bruker Daltonik GmbH, Bremen

One of the great challenges of the upcoming postgenomic era is the determination of sequence variations, in particular single nucleotide polymorphisms (SNPs). These polymorphisms are believed to have an enormous impact in diagnosis of diseases, improvement of drugs, and forensic analysis in future. In contrast to the increasing demand for SNP genotyping, there is still a lack of highly reliable high-throughput SNP typing techniques.

Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) is a powerful advanced technology capable of the accurate and cost-effective typing of many thousands of SNPs per day. The "GOOD assay", a novel sample preparation procedure for MALDI-TOF typing of SNPs was recently presented (Sauer et al., Nucleic Acids Research, methods online, 2000, 28, e13). This preparation uses sensitivity enhancing, chemical modifications of the allele specific products. Main advantage of the GOOD assay over other SNP genotyping preparation methods with MALDI detection is that this protocol does not require solid phase purification of the products prior MALDI analysis. SNP typing starts with a PCR encompassing a region containing a known SNP. Thereafter dNTPs are digested with shrimp alkaline phosphatase. In the following primer extension reaction using a charge tag containing extension primer, a conditioned set of ddNTPs and a DNA polymerase of allele specific products are generated. The 5' part of the extension primer is removed by an exonuclease. By alkylation the backbone phosphate groups are neutralised. This results in small singly charged product molecules which are transferred onto non-protonating matrix on the MALDI target.

Mass spectra of 384 samples were acquired in less than one hour. Allele calling was performed automatically by the novel genotools SNP manager software, online during spectra acquisition or offline. The resulting genotypes are stored in a table in the ASCII format together with a quality asse.

P53-M

Nano-LC ion-trap mass spectrometry for the investigation of low-fmol protein amounts.

U. Schweiger-Hufnagel¹, A. Schneider¹, A. Ingendoh¹, S. Liedtke², R. van Soest³; ¹Bruker Daltonik GmbH, Bremen, Fahrenheitstr. 4, Bremen, Bremen 28359 Germany, ²Bai GmbH, ³LC Packings NL

Ion trap mass spectrometry is a powerful tool for the proteomic research since it is predestined for the connection to a liquid chromatography system, which allows the simple fragmentation of all separated peptides in one experiment. However, the limiting factor for the protein analysis so far was the chromatography, which, due to the high inner column diameter and high flow rates reduced the detection limit of the mass spectrometer. To improve this, we connected a nano-LC system to the ion trap mass spectrometer to investigate the applicability of this setup for the proteomic research.

The peptides were separated on a 75 µm ID RP column at a flow rate of 160 nl/min. To achieve the LC-MS coupling fused-silica spray needles in combination with interfaces that support the use of peek based HPLC fittings (New Objective, USA) were used. The ion trap mass spectrometer was equipped with an online-nanospray device.

Using this setup, a detection limit in the low fmol range is achieved (about 1/100 of the detection limit for a capillary HPLC column). Thus, this powerful technique can be applied to those protein amounts which are provided by the proteome researcher (e.g. obtained from 2-D gels). Therefore, the nano-LC ion trap mass spectrometer meets the sensitivity requirements of proteomic research, and it enables the use of the ion trap for diverse analytical questions. Those include protein identification, the analysis of protein mixtures, the analysis of post-translational modifications and the sequence analysis.

P54-T

Characterization of the proteome of Oryza sativa.

P.A. Haynes, A. Koller, N. Andon, J. Wei, J.R. Yates, III; Novartis Agr. Discovery Inst., San Diego, 3115 Merryfield Row, San Diego, CA 92121

As part of our ongoing efforts to enhance our understanding of plant biology in cereal crops, we have undertaken the characterization of the proteome of the major tissues and subcellular fractions of Oryza sativa (Rice). The tissues analyzed include leaf, root and seed. These were chosen because, despite certain similarities, each tissue performs very different functions in the plant life cycle and interacts with unique environmental factors. Further fractionation of leaf tissue has allowed the analysis of organelles including nuclei, chloroplasts and mitochondria, and subcellular fractions including cytosol and membrane preparations. Further fractionation of seed has allowed the analysis of endosperm and embryo fractions. Protein extracts were prepared from each sample, and proteins were visualized by two dimensional electrophoresis using immobilized pH gradient-SDS-PAGE. Proteins were then excised from the gel and digested in-gel with trypsin. Peptide mixtures were separated and analyzed using liquid chromatography-tandem mass spectrometry, and peptides were identified using computer database searching of uninterpreted fragment ion mass spectra.

We have identified several hundred proteins by searching our data against publicly available sequence data from various plant species. These include proteins common to several tissues as well as those which are unique to a particular tissue. This data represents the most comprehensive picture to date of the proteome of tissues and organelles from cereals, and in particular rice plants. The knowledge of the identity and distribution of expressed proteins will add value to our genomic sequencing efforts, and will be used to provide a baseline for future studies of which proteins are up- or down-regulated in response to environmental stress.

P55-S

ESI-ion trap mass spectrometry for the analysis of PCR products.

S. Hahner, A. Schneider, A. Ingendoh; Bruker Daltonik GmbH, Bremen, Fahrenheitstr. 4, Bremen, HB 28359 Germany

The polymerase chain reaction (PCR) is an important tool used in many applications in molecular biology for the generation and amplification of specific DNA sequences from very small sample material. PCR products are commonly analyzed by means of gel electrophoresis. Major disadvantage of this technique is that the size of a PCR product is determined by comparison of the relative electrophoretic mobility in the gel matrix to an internal size standard. In addition, a labelling of the analyte is required for detection.

Mass spectrometry (MS) has been proven to be a useful technique for the determination of both the length and sequence variations of PCR products based on the accurate determination of molecular weights. So far, PCR products have been analyzed using either matrix-assisted laser desorption- (MALDI) or electrospray ionization- (ESI) MS. The performance of both techniques is less labour- and time consuming compared to gel electrophoretic techniques. However, the application of MS is mainly limited due to the tendency of the DNA molecules for the formation of multiple salt adducts which complicate the data.

The development of a purification approach based on the adsorption of PCR products on magnetic particles has been found to be efficient for purification of PCR products prior to MS. In addition, this approach is amenable to full automated handling. The application of the magnetic bead purification technique in combination with ESI ion trap MS yields mass spectra with a resolution and accuracy sufficient to assign single nucleotide sequence deviation in PCR products up to a size of 90 bp.

P56-M

Hydrophilic anchors enable high throughput protein identification in the subfemtomole range.

P. Hufnagel, M. Schürenberg, C. Köster, K-O. Kräuter; Bruker Daltonik GmbH, Bremen, Fahrenheitstr. 4, Bremen, Bremen D-28359, Germany

Research strategies in the Proteomics field rely on both high sensitivity and high throughput spectra acquisition. In addition, there is a strong need for convenient analysis tools that allow high throughput processing, database searches and the assessment of a high number of search results.

Hydrophobically coated targets equipped with small hydrophilic anchors have shown 10-100-fold higher sensitivity for MALDI TOF MS than conventional steel targets. A droplet (0.5-3 µl) containing matrix and analyte is applied onto a hydrophilic anchor and shrinks down to the anchor's size, thereby restricting the formation of matrix crystals to an exactly defined area. This drying process increases the concentration of the analyte by a factor that depends on the droplet's and on the anchor's size. The fact that the matrix crystals are restricted to the area of the anchor is a big advantage of anchor targets when it comes to automation.

Using an eight-channel pipetting robot for sample preparation (approx. 2 sec/sample) arrays of hydrophilic anchors (as well as standard metal targets) were loaded with protein digests of varying sample amounts. The combination of automatic spectra acquisition with batchwise database searching led to an overview of the sensitivity increase (a factor of approx. 100) achieved by the anchor technology. The database search results have been automatically assessed by a fuzzy-control-based algorithm.

Here, high throughput capabilities of preparation, acquisition and analysis have been used to compare the sensitivity of standard sample preparation with the hydrophilic anchor technology on a statistical basis. Furthermore, this analysis serves as a model for the application of these fundamental techniques for the fast achievement and analysis of search results coming from proteomic research tasks.

P57-T

Optimization of a procedure for labeling proteins with isotope-coded affinity tags.

K. Parker, S. Pillai, S. Daniels, W. Stanick, R. Lotti, B. Purkayastha, T. Nadler; Applied Biosystems, 500 Old Connecticut Path, Framingham, MA 01701

The use of the isotope coded affinity tag (ICAT™), developed by the laboratory of R. Aebersold (Nat. Biotechnol. 1999 Oct 17(10):994-9), allows one to compare the relative protein abundance between two samples by mass spectrometry. The reagent used is essentially cysteine reactive biotinylation reagent into which stable isotopes have been incorporated. One of the two samples is treated with the light reagent while the other sample is treated with the same reagent containing 8 deuterium atoms such that the samples are chemically equivalent but differ only in mass. In this way two peptides may be monitored in the mass spectrometer and quantified relative to one another by peak intensity.

A number of imporant questions concerning the use of this technique are addressed in this presentation. The reagent has been extensively characterized and the recommended protocol for its use has been optimized for ease of use and robustness. Experiments were carried out to determine the purity, stability, and reaction kinetics of the reagent. Additional experiments were performed to optimize the derivatization protocol, including measurements of the degree of alkylation obtained, determining the optimal reduction reagent, how to remove excess reagent, and measurements of stability, capacity, carryover, and recovery of biotinylated peptides from monomeric avidin columns, Finally, the reproducibility of quantitative measurements in the mass spectrometer was addressed and will be presented.

P58-S

Rapid genotyping using spectrometrically monitored selections.

J. Stoerker¹, J.D. Mayo², C.N. Tetzlaff³, D.A. Sarracino⁴, C. Richert³; ¹Bruker Daltonics, 15 Fortune Drive, Billerica, MA 01821, ²Tufts Univ., ³Univ. of Konstanz, ⁴Variagenics, Inc.

Presented here is an inexpensive and rapid method for genotyping single nucleotide polymorphisms (SNPs) on PCR products using spectrometrically monitored selection from probe libraries. We have previously reported nuclease selections of modified oligonucleotides with increased RNA or DNA targets. This methodology relies on the selection of short hybridization probes complementary to the target sequence with a single-strand specific nuclease that does not attack duplexes. Conditions are described where the matched duplex is profoundly better protected from enzymatic digestion than single base mismatches. Semi-quantitative MALDI-TOF MS is used to identify the selected probes after a 10-minute digestion reaction with snake venom phosphodiesterase. We have examined sequences from exon 10 of the cystic fibrosis transmembrane regulatory (CFTR) gene, using a probe library representing a number of alleles, including: native, I506S, 1651 benign polymorphism, I506S/1651 dual polymorphism, and dF508, and dI507. Probe libraries composed of 8-mers and 9-mers annealing to PCR products have been used in our experiments. DNA from blood or cheekbrush samples was used from 10 genomic sources. The method discriminates both alleles of a heterozygote pair in a single experiment, and can discriminate alleles from multiple loci in the same sample.

P59-M

Quantitative proteomic and mRNA expression analysis of S. cerevisiae via metabolic labeling and multidimensional protein identification technology.

M.P. Washburn¹, G. Oshiro², D. Wolters¹, D. Schieltz¹, C. Deciu¹, E. Winzeler², J.R. Yates, III¹; ¹Novartis Agr. Discovery Inst., 3115 Merryfield Row Suite 100, San Diego, CA 92121, ²Genomics Inst. of Novartis Fndn., San Diego

Proteomic technologies are being developed to determine large-scale changes in protein expression levels. Several quantitative proteomic methods have recently been published where the protein expression levels are compared between two different growth conditions by metabolic isotopic labeling of proteins or post-growth isotopic labeling of proteins in a sample. To fully understand how growth conditions effect biological systems, both protein and mRNA expression levels must be analyzed. We have applied metabolic labeling strategies to S. cerevisiae in order to determine the changes in protein and mRNA expression levels. S. cerevisiae was grown in ¹⁵N enriched minimal media and compared to S. cerevisiae grown in rich media. Each sample was analyzed via mRNA expression array and multidimensional protein identification technology. By combining these methods we were able to correlate the changes in protein and mRNA expression levels of several hundred gene products.

P60-T

A software algorithm for automated quantitation of ICAT labeled peptides analyzed by mass spectrometry is described.

J. Eng, J. Ranish, R. Aebersold; Inst. for Systems Biol., 4225 Roosevelt Way NE, #200, Seattle, WA 98105

We describe a software program which automatically generates quantitation ratios of Isotope Coded Affinity Tag (ICAT) labeled peptides analyzed in an LC-MS/MS experiment performed on an LCQ ion trap mass spectrometer. The peptides are initially identified via a sequence database search. Quantitation of each ICAT-labeled peptide pair is then performed by integrating over the peptide's light and heavy elution profiles. An interactive, graphical program displays the light and heavy elution traces and allows the researcher to adjust the area of integration as needed. This program should enable and facilitate large scale, quantitative proteomics studies.

P61-S

Genopure: a novel magnetic bead DNA purification system for MALDI-TOF MS analysis.

M. Kostrzewa¹, J. Bimmler², I. Thomas², T. Wenzel¹, E. Nordhoff³, H. Rauth³, T. Fröhlich¹; ¹Bruker Saxonia Analytik GmbH, Permoser-strasse 15, Leipzig, Saxonia D-04318 Germany, ²Bruker Saxonia Analytik GmbH, Leipzig, ³Max Planck Inst. of Molec. Genet., Berlin

MALDI-TOF mass spectrometry has a high potential for high throughput DNA analyses because of its accuracy, speed, automation capabilities, and cost-effectiveness. Unfortunately, molecular biological reactions are performed in the presence of high amounts of salts and detergents which form adducts with DNA or interfere with matrix crystallization, respectively. Therefore, effective DNA purification is indispensable for successful and valid MALDI-TOF MS analysis. We present a novel magnetic DNA purification system which gives excellent results in subsequent mass spectrometric measurement.

Small double stranded PCR products were precipitated on paramagnetic beads using an appropriate binding solution. The binding of DNA to the particles did not require any labeling of primers or dNTPs. After several washing steps with buffers containing either ethanol or isopropanol, DNA was eluted and measured with 3-HPA matrix in a MALDI-TOF mass spectrometer. Residual PCR primers were removed while PCR products down to about 50 bp in size were recovered with high yield. Quality of the acquired spectra was superior to that of alternative purification protocols.

Alternatively, eluted PCR-Products were used for subsequent reactions like primer extension or restriction enzyme digest. Primer extension products or very small restriction enzyme digest products (about 20 bp) were again purified by the magnetic bead system to MALDI quality with high recovery using a dedicated binding buffer. The combination of both purification procedures enabled the genotyping of SNPs by performing PCR, purification of the PCR product, primer extension or digest, and clean-up of the allele specific products in one reaction tube. High quality spectra could be acquired for several model systems allowing the unambiguous determination of genotypes. The bead handling can easily be automated on standard pipetting robots thereby enabling the processing of thousands of samples per day.

P62-M

Industrial scale high throughput proteomics.

J. Brown, D. Gostick, P. Young, J. Langridge; Micromass UK Ltd, Floats Road, Wythenshawe, Manchester M23 9LZ, United Kingdom

With the completion of the Human Genome sequence MALDI-TOF-MS is increasingly becoming an established method for identification of proteins separated by 2D gel electrophoresis. Mono-isotopic peptide mass fingerprinting (PMF) has been previously shown to be amenable to full automation encompassing the process of acquisition, data processing and databank searching under full software control.

Until now the throughput of MALDI-TOF-MS for proteomics has been limited to several hundred samples in a working day and this represents approximately 5-10% of the total proteins resolved by a large format 2D gel. To reduce the number of proteins to be identified the 2D gels are imaged and analysed to determine differences in expression levels within a set of gels. Although much of the image processing is semi-automated the comparison is labour intensive as manual pattern matching has a role in the gel alignments (land marking). Increased MS sample throughput allows the possibility of identifying every protein spot in a 2D gel within a day. This could eliminate the potentially erroneous step of human gel image alignment, whereby land marking could be achieved using the MS data.

Increased sample throughput requires greater capacity and robust unattended instrument operation. In this poster we describe an integrated robotic multiple plate loader that allows overnight unattended MS operation. Other improvements include an increased laser repetition rate that allows the data capture rate to increase four fold. Sample tracking, data archiving and data reporting are essential attributes of this new technology and these aspects are outlined in the present

P63-T

Bioinformatics assessment of mass spectrometric chemical derivatisation techniques for proteome database searching.

K.S. Sidhu¹, P.J. Lester¹, S.J. Gaskell¹, O.V. Wolkenhauer¹, S.G. Oliver², F. Brancia¹, A.G. Sullivan¹, P. Sangvanich¹, S.J. Hubbard¹; ¹UMIST, UK, PO Box 88, Manchester, Lancashire M60 1QD, United Kingdom, ²Manchester Univ.

Identification of proteins from the mass spectra of peptide fragments generated by proteolytic fission and subsequent database searching is one of the most powerful techniques in proteome science. Using computer simulation, we have studied how various chemical derivatisation techniques developed in the Michael Barber Centre for Mass Spectrometry at UMIST can improve the efficiency of protein identification from mass spectrometric data. The impact of these different derivatization strategies, which promote and stabilise certain fragmentation pathways yielding additional database search information has been assessed. For example, by reliably promoting fragmentation at aspartic acid residues after homoarginine derivatisation, 85% of yeast proteins can be unambiguously identified from a single peptide with a measured mass accuracy of 500 ppm by using the associated fragment ion data. We present results here for 3 alternative techniques, both for single proteins and for simple mixtures. Additional data have also been generated to compare these "partial sequencing" methods with high accuracy mass spectrometry where peptide mass accuracy can be achieved at 50 ppm and beyond. Interestingly, for protein mixture analysis, the inclusion of limited sequence information for the peptides can compensate and exceed the search efficiency available via high accuracy searches of around 20 ppm, suggesting the particular use of this experimental approach for the simple protein mixtures that are routinely obtained from 2D-gels.

P64-S

Automated high throughput protein identification on a hybrid quadrupole orthogonal acceleration time-of-flight mass spectrometer coupled with a MALDI ion source.

A. Millar, R. Tyldesley, J.I. Langridge, J.B. Hoyes, Y. Philip, R. O'Malley, R. Bateman; Micromass UK Ltd., Floats Road, Wythenshawe, Manchester M23 9LZ, United Kingdom

The mass spectrometry technique providing the highest throughput in terms of samples per hour, is currently MALDI-TOF-MS. This technique provides a peptide mass fingerprint of the protein digests and allows rapid and accurate identification of the parent protein by comparison to a databank. However, under some circumstances, for example if the number of peptides detected is small or if the sequence coverage is poor, it is advantageous to be able to include even a short piece of sequence information to provide added specificity. In a conventional MALDI-TOF-MS instrument post source decay (PSD) can be used to try and generate sequence information, however this approach is notoriously unreliable in producing good quality MS/MS data. One reason for this is that the peptide does not undergo fragmentation in a controlled environment such as a gas cell.

An alternative approach is to couple a MALDI ion source to a hybrid quadrupole orthogonal acceleration time-of-flight (Q-Tof) mass spectrometer, which results in predictable fragmentation. In contrast to a conventional MALDI-TOF-MS instrument the resolution and mass measurement accuracy of the data is comparable between the MS and MS/MS modes. This allows superior data acquisition in the MS-MS mode compared to other MALDI-TOF systems.

In this paper we demonstrate the application of the MALDI Q-TOF instrument for high throughput proteomics. A number of modifications have been made to optimise the system for high throughput proteomics, these include a MALDI source with a high-density target plate and software that has been developed for automated data acquisition in both the MS mode and the MS to MS/MS switching mode. Dedicated processing software has been developed to fully automate the post acquisition and databank searching. This software has been optimised to consider the unique nature of the data acquired from this configuration of instrument.

P65-M

Single scan PSD-MALDI-TOF analysis at high speed and sensitivity.

D. Suckau, A. La Rotta, A. Holle, P. Hufnagel, D. Suckau; Bruker Daltonik GmbH, Bremen, Fahrenheitstr. 4, Bremen 28359, Germany

The typical approach to Proteomics measurements involves the MALDI-TOF fingerprint analysis of protein digests, which allows the identification of a significant fraction of samples. PSD as it is available on a reflector TOF instrument is in our hands a very sensitive technique (1-10 fmol level) and is automated; therefore it principally is the ideal technique to continue MS/MS analysis from the same prepared sample. However, it suffers from the fact that each PSD spectrum is pasted together from a set of ca. 10 voltage segments, each consisting of 100-200 laser shots. Sample consumption is such that a maximum of 2-3 peptides can be submitted to MS/MS analysis and the process is time consuming. Here we report on the usability of single scan PSD spectra from digests for protein identification.

The acquisition of the entire fragment spectrum has been achieved for a MALDI reflector TOF mass spectrometer by introducing a "potential lift", a precursor selection device consisting of a pulsed HV cell with acceleration grid. An HV poten-tial is applied to the lift cell at the moment the precursor ion and its PSD fragment ions have entered it. By the subsequent acceleration, the energies of the fragment ions are compressed into the operating window of the reflector, which allows analysing them in a single spectrum.

With this device LIFT-PSD spectra from proteolytic digest mixture were obtained and evaluated with respect to sensitivity, mass accuracy, speed of analysis and number of PSD spectra per sample. It potential use in Proteomics is discussed.

P66-T

Quantitative proteomic analysis using a MALDI quadrupole time-of-flight mass spectrometer.

T.J. Griffin¹, S.P. Gygi², B. Rist³, H. Zhou¹, H. Lee¹, A. Loboda⁴, A. Jilkine⁵, W. Ens⁶, K.G. Standing⁶, R. Aebersold¹; ¹Inst. for Systems Biol., Seattle, 4225 Roosevelt Way N., Seattle, WA 98105, ²Harvard Med. Sch., ³BioVisioN GmbH and Co. KG, Hannover, ⁴MDS Sciex, Concord, Canada, ⁵Univ. of Manitoba, ⁶Univ. of Manitoba

We describe an approach to the quantitative analysis of complex protein mixtures using a MALDI quadrupole time-of-flight (MALDI QqTOF) mass spectrometer and isotope coded affinity tag (ICAT) reagents. Proteins in mixtures are first labeled on cysteinyl residues using an ICAT reagent, the proteins are enzymatically digested, and the labeled peptides are purified using a multi-dimensional separation procedure, with the last step being the elution of the labeled peptides from a microcapillary reverse-phase liquid chromatography column directly onto a MALDI sample target. After addition of matrix, the sample spots are analyzed using a MALDI QqTOF mass spectrometer, by first obtaining a mass spectrum of the peptides in each sample spot in order to quantify the ratio of abundance of pairs of isotopically tagged peptides, followed by tandem mass spectrometric analysis to ascertain the sequence of selected peptides for protein identification. The effectiveness of this approach is demonstrated in the quantification and identification of peptides from a control mixture of proteins of known relative concentrations, and also in the comparative analysis of protein expression in Saccharomyces cerevisiae grown on two different carbon sources. Additionally, we have demonstrated the utility of this approach for the quantitative analysis of isotopically labeled phosphorylated proteins, and we are currently working on automating the sample preparation procedures for high-throughput applications.

P67-S

Proteomics approach to investigate the toxicity of diclofenac.

L.J. Alward, O.V. Nemirovskiy, G.S. Cavey, J.E. Carlson, W.R. Mathews, J.A. Ware; Pharmacia, 301 Henrietta Street, Kalamazoo, MI 49007

Diclofenac, a nonselective inhibitor of cyclooxygenase, is a widely used non-steroidal anti-inflammatory drug. It has been shown to produce hemolytic anemia in sensitive individuals. A proteomics approach was used to determine the mechanism of diclofenac-induced hemolytic anemia in the mouse. A flexible, high performance proteomics laboratory was assembled using commercially available instrumentation. The system used was validated with protein standards and included robotic spot excision from 2D gels, automated in-gel digestion, robotic sample preparation, automated MALDI-Tof analysis, automated nanoLC-MS/MS analysis, and automated database searching. Diclofenac was administered to mice by gavage. Liver, plasma, and RBC proteins were separated using 1D and 2D gels, and diclofenac-protein adducts were visualized by either radiolabel or immunoblot. Proteins of interest were excised, reduced, alkylated, and trypsin digested. Successful protein identification was accomplished via MALDI-Tof MS peptide mass fingerprint analysis using 20% of digest sample. Nanoscale LC-MS/MS provided comprehensive identification using 50% of digest sample. The results obtained by nano-LC-MS/MS analyses were in a good agreement with those obtained by MALDI-Tof. Many proteins were identified including fibrinogen alpha and beta, hemoglobin beta and delta chains, and RBC anion exchanger (AE1). Abnormalities in AE1 have been shown to accelerate membrane destruction. Preliminary data from these experiments is promising and may help explain diclofenac toxicity.

P68-M

Selective separation of methionine-containing peptides combined with mass spectrometry: a high accuracy technique for identification of proteins.

W.L. McEldoon, M.J. Horn; BioMolecular Technol., Inc., 525F Del Rey Ave., Sunnyvale, CA 94085

Although MALDI-TOF MS is a widely accepted technique for identification of proteins and peptides in proteomic and genomic analyses, rapid and accurate identification of a large number of proteins is still a great challenge. Previously, we reported an easy method, which uses a novel water-insoluble reagent that selectively separates methionine-containing peptides from peptide mixtures. This method, combined with MALDI-TOF analysis, demonstrates a significantly improved accuracy in the identification of proteins. In the present study, we present a total methodology for the separation and identification of methionine-containing peptides generated from 2D gel digestion mixtures. The methodology provides a "one-pot" method for sample desalting and concentration after the isolation process, giving cleaned-up samples ready for multiple MALDI or other analytical analyses.

P69-T

Efficient detergent and Coomassie blue dye removal from peptide samples prior to MALDI-TOF MS.

W. Kopaciewicz, E. Kellard; Millipore, 17 Cherry Hill Drive, Danvers, MA 01923

Advancements in Mass Spectrometry (MS) instruments over the last 5 years have reduced the amount of sample required and expanded the molecular weight analysis range. As such, MS is an increasingly valuable tool for the analysis of biomolecules in areas of proteomics and functional genomics. However, the presence of salts, detergents and dyes can decrease macromolecular ionization and/or obscure mass peaks, thus decreasing instrument performance. As such, numerous methods and devices have been developed for sample preparation prior to MS analysis.

Millipore ZipTip sample preparation devices containing C18 and C4 media reversed phase media have been demonstrated to be effective for the concentration and desalting of microliter volumes of peptide, protein and oligonucleotide samples prior to MALDI-TOF MS. Although reversed phase chromatography meets the vast majority of the sample preparation needs, there are cases where it isn't totally effective. This is especially evident when the contaminants have similar hydrophobic character as the solutes and thus co-purify.

Alternatively, a subset of normal phase chromatography termed hydrophilic interaction chromatography has been shown to be an efficient sample preparation technique for certain biomolecules in those cases where reversed phase struggles (1,2,3). The method utilizes a highly hydrophilic adsorbent that binds sample out of a high organic mobile phase (e.g. 90% acetonitrile). After washing with this solvent, bound molecules are then eluted by decreasing the organic content (e.g. 50% acetonitrile). In this poster, we demonstrate the use of a ZipTip with hydrophilic interaction media (ZipTipHPL). The device was very effective for the removal of detergents and coomassie blue dye from sub-picomole quantities of peptides. The resulting mass spectra demonstrated good signal strength and fidelity of peptide capture.

1. Alpert, A.J., J. Chromatogr. 499 (1990) p.177

2. Zhu, B.Y., et al., J. Chromatogr. 548 (1991) p.13

3. Scherer, J.P., et al., Anal. Biochem. 215 (1993) p.292

P70-S

Improved proteome coverage through the use of affinity tags with nanoscale capillary LC/MS/MS.

K. Blackburn¹, W. Burkhart², R. Davis², M. Moyer², A. Moseley²; ¹Glaxo Wellcome, 5 Moore Drive, PO Box 13398, Research Triangle Park, NC 27709, ²Glaxo Wellcome

Because of the inherent difficulties associated with 2D gel electrophoresis (low throughput, bias against certain classes of proteins, poor peptide recovery from in-gel digests, etc.), numerous groups have explored ways to analyze complex protein mixtures directly by mass spectrometry, avoiding the gel separation altogether. In proteomic experiments where qualitative and/or quantitative data is required for complex biological samples such as protein complexes, organelles, or tissue samples, the number of proteins potentially encountered may range from tens to thousands. The number of proteolytic peptides generated from digests of such complex protein mixtures precludes "complete" proteome coverage by any 1D-LC/MS/MS analysis. Following on the work of Aebersold and Patterson, we have investigated the use of cysteine-modifying "affinity" tags with avidin affinity chromatography for the simplification of complex peptide mixtures prior to nanoscale capillary LC/MS/MS analysis. This approach reduces the number of peptides presented to the mass spectrometer to a more reasonable number, thus allowing improved proteome coverage. Data will be presented on improved sample handling procedures for labeling and digestion steps as well as qualitative and quantitative applications to complex biological samples.

P71-M

Factors affecting rapid sample processing for protein capture chips analyzed by MS.

J. Chan¹, P.T. Jedrzejewski¹, P. Wagner¹, S. Nock²; ¹Zyomyx, Inc, 3911 Trust Way, Hayward, CA 94545, ²Zyomyx, Inc.

Robust, sensitive, and comprehensive methodologies for systemic approaches to proteome analysis analogous to genomics methods (e.g., gene chips) are not available. Current methodologies (e.g., 2D-PAGE and multi-dimensional chromatographic methods) suffer from fundamental limitations. In order to overcome these limitations, we as well as others have focused on the development of alternative methods (e.g., protein capture chips) for proteomics.

Protein capture chips, consisting of arrayed capture molecules (e.g. protein, DNA, or small molecule), allow for rapid and comprehensive micropurification and analysis of proteins. Protein chips may be readily interrogated by mass spectrometric analysis. Data on the post-translational modifications (PTM), protein identification, and quantitation may be obtained with a single detection scheme without labeling of analytes.

We have been successful in microfabricating protein capture arrays, the accompanying microfluidics devices, and implementing MS analysis. We have investigated various parameters which have an effect on the utility and efficacy of the overall system for proteomics. In this presentation, we will demonstrate data on parameters affected by the type of capture molecules, digest efficiency (e.g., time), and sensitivity of analysis (e.g., LC-MS conditions).

P72-T

High throughput in-gel peptide digestion and microscale sample preparation for MALDI-MS analysis of the resulting peptide mass fingerprint.

M.G. Pluskal¹, A.M. Pitt²; ¹Proteome Systems, Inc., 14 Gill St., Woburn, MA 01801, ²Millipore Corp.

In-gel peptide digestion has become a widely used technique for characterizing proteins resolved by two dimensional gel electrophoresis. Peptides generated from gel pieces are frequently contaminated with detergent and salts. Prior to MALDI-MS analysis, these contaminants are removed using microscale C18 sample preparation columns. In this poster, data will be presented to demonstrate the application of a solvent resistant Multiscreen 96 well plate with an optimized low peptide binding membrane and ZipTip C18 micropipet based sample preparation. Recoveries of peptides (Mz range of 1000 to 5000 Dalton) derived from standard protein protease digests, were estimated at various stages of the analytical process. An optimized protocol has been established and all the reagents and consumables have been packaged in a ready to use commercial kit. Data will be presented to show application of this technology package to accelerate the throughput of protein characterization by protease fragmentation.

P73-S

Rapid and high throughput proteomics by protein capture chips combined with MS analysis.

J. Chan¹, P.T. Jedrzejewski¹, P. Zwahlen², P. Indermuhle², P. Wagner¹, S. Nock³; ¹Zyomyx, Inc, 3911 Trust Way, Hayward, CA 94545, ²Zyomyx, Inc., Hayward, CA, ³Zyomyx, Inc.

The study of the proteome is a critical step towards the function determination of each gene since proteins are the agents of life's work. However, robust, sensitive, and comprehensive methodologies for systemic approaches to proteome analysis similar to genomics methods (e.g., gene arrays) are not available. Current methodologies such as 2D-PAGE and multi-dimensional chromatographic methods (e.g., LC-LC, LC-CE) suffer from fundamental limitations. In order to overcome these limitations, we as well as others have focused on the development of alternative methods (e.g., protein capture chips) for proteomics.

Consisting of arrayed capture molecules, protein chips allow for rapid and comprehensive micropurification and analysis of proteins. Protein chips may be readily analyzed by optical methods; however, detailed data may be obtained when arrays are interrogated by MS. Data on the post-translational modifications (PTM), protein identification (especially important when complexes are isolated), and quantitation may be obtained with a single detection scheme without labeling of analytes. In this presentation, we will show our efforts in this area.

We have been successful in microfabricating protein capture arrays, the accompanying microfluidics devices, and implementing MS analysis. These mircrofluidic devices offer a number of advantages (e.g., high density elements, integration of operations). We have devised a microfluidic device capable of performing a number of sample preparation operations prior to MS analysis.

In this presentation, we will demonstrate the functionality and utility of these devices through the identification and characterization of proteins captured from mixtures. Various assay types will be demonstrated (e.g., protein-protein, small molecule-protein, protein-DNA). The overall system performance characteristics will be presented: low femtomole sensitivity, high dynamic range, and selectivity. The utility of rapid protein identification through database searching and PTM characterization will be highlighted in various applications.

P74-M

Reflector in-source-decay MALDI TOF MS: a powerful tool for N-terminal sequence characterization of proteins.

D. Suckau, A. Resemann, M. Witt; Bruker Daltonik GmbH, Bremen, Fahrenheitstr. 4, Bremen 28359, Germany

MS-characterization of proteins can simply be initiated by intact molecular weight determinations. If the experimental value doesn't agree with the anticipated structure, a closer investigation of the termini is a must--besides modifications. In particular N-terminal modifications impose a significant analytical problem, since Edman sequencing fails and MS/MS of digested peptides generates a tremendous amount of (useless) information to address this question and might even not contain the N-terminal peptide.

We introduce ISD measurements on a standard reflector MALDI-TOF instrument to obtain near-N-terminal sequence information (residues ~10-30). C-type MS/MS fragment ions generated from intact proteins of up to 60 kDa using this approach allow to read the sequence with a <0.04 Da mass accuracy. This sequence information can directly be used for sequence database searches. This information was sufficient to characterize the N-terminus, i.e., to identify the fusion protein system and the unexpected N-terminal Met-methylation, which was confirmed by additional PSD and fingerprint analysis.

ISD seem to be an ideal technique to characterize the N-terminus of proteins and in particual of N-terminally blocked proteins, provided they are available on the 10 pmol level and not contaminated by other proteins in a mixture.

P75-T

Rapid protein identification of blocked proteins using dilute acid cleavage and automated Edman sequencing.

S. Wong, A. Kishiyama, V. Pham, W.J. Henzel; Genentech

We have developed a method for rapidly cleaving and identifying proteins electroblotted onto polyvinylidene difluoride (PVDF) membranes. Cleavage is performed with 10% acetic acid in 7 M guanidine chloride at pH 2.5 for 1 hour at 90°C, resulting in fragmentation primarily at aspartyl-prolyl bonds. Peptides resulting from non Asp-Pro cleavage are N-terminally blocked by reaction with orthophthalaldehyde (OPA) prior to automated Edman degradation. Reaction with OPA after cleavage blocks all amino acids containing primary amino groups. Only peptides containing an N-terminal amino acid with a secondary amino group (proline) will be available for reaction with the Edman reagent. The sequences obtained are used for protein database searching. Using this approach, proteins that are found to be N-terminally blocked can be removed from the sequencer, cleaved with acetic acid, blocked with OPA and reapplied to the sequencer. The protein can then be identified from a database search using the sequence mixture obtained.

P76-S

Quantitation of gel separated proteins at the low picomole level.

S.W. Yuen, L.R. Zieske, K-L. Hsi, T.L. Schlabach; Applied Biosystems, 850 Lincoln Centre Drive, Foster City, CA 94404

Comparative proteomics depends on the ability to quantify changes in cellular protein levels between the control state and the abnormal or treated state. The most widely accepted method for protein quantitation is amino acid analysis, but this method performs poorly at low picomole levels and below, which are most common in proteomics studies.

Edman chemistry is a nearly stoichometric method for removing amino acid residues from a protein with typical repetitive yields of greater than 90%. The problem is the initial yield is not stoichometric and shows variability that is dependent on protein sequence.

Cleavage of an unknown protein into numerous fragments reduces the impact of variability in the initial yield of any one fragment and removes much of variability associated with sequence dependence. We examine both chemical and enzymatic digestion techniques for fragmentation.

We report in this presentation on the relative and absolute accuracy of protein sequencing for determining protein concentration. Both solution and gel isolated proteins will be studied and the results compared with sequence analysis of intact proteins.

P77-M

The utilities of N-terminal sequencing in the post-genomic era.

L.R. Zieske¹, S.W. Yuen¹, T.A. Settineri¹, D. Hawke¹, C. Bloch²; ¹Applied Biosystems, 850 Lincoln Centre Drive, Foster City, CA 94404, ²Embrapa-Cenargen, Brazil

Mass spectrometry continues to develop and improve methods and procedures to make it the primary tool to identify proteins for "proteome analysis". The high mass accuracy and sensitivity of instruments such as hyphenated quadrupole TOF mass spectrometers has made low level sequencing of peptides and proteins possible. Yet even with these advancements in MS the need for chemical sequencing still exists!

MS sequencing is very powerful, but exact location of the fragmentation patterns are not predictable. Thus creating difficulty in assigning the exact N-termini. The need for exact N-terminal sequence information is especially important in determining mRNA editing and/or determining the true open reading frames when EST database searching. Edman chemistry is the one sure way of determining this type of information. Complementary to this is the use of "multiple peptide sequencing" which simultaneously provides multiple amino acid information per residue cycle. Thus providing very high quality sequence information with which to interrogate the error prone databases now used for searching. In addition, chemical sequencing provides the necessary information to determine quality assurance of cloned products; again making sure no frame shifts occurred in the processing. Another complementary need for chemical sequencing is in assisting the assignment of the correct ion series patterns being generated even when relatively complete fragmentation data is recorded. Chemical sequencing also identifies difficult to discern or unusual amino acids easily such as I/L and/or hydroxyproline.

In this work we will show some examples of how N-terminal analysis was used in concert with mass spectrometry to unravel protein/peptide structural information. In addition, we will show demonstration of "multiple peptide sequencing" in inspecting the various databases available.

P78-T

Increasing the capacity of a commercial Edman sequencer: a protein auto sampler.

W. Shillinglaw, S. Wong, T. Moreno, W.J. Henzel; Genentech

We have designed an implemented a high throughput autosampler to increase the capacity and speed of protein sequencing. The autosampler attaches to a standard ABI Procise sequencer, enabling a single separate sample cartridge to now hold up to six separate samples. The autosampler is used in combination with faster Edman cycles and a rapid 12 minute PTH separation to significantly increase the speed of sample analysis. The reaction cartridges on the autosampler are composed of disposable Teflon tubing, allowing for a significantly cleaner background when compared to the reusable glass cartridge blocks, which are standard on the ABI sequencers. The lower background reduces the ambiguity of identifying the amino acids in the first few cycles, which is often a problem. A low cost program logic controller which is connected through an external relay to the protein sequencer controls the autosampler.

P79-S

Microsatellite analysis using fluorescent PCR primers synthesized in tandem.

J. Fernandez, M. Kirchner, J. Brito, T-J. Daley, G. Dolios, B. Imai; Rockefeller Univ., 1230 York Ave., New York, NY 11021

SDS-PAGE is typically the final purification step for isolation of small amounts of proteins for further chemical characterization. Proteins are digested in-gel with trypsin and the resultant peptides analyzed by 1) MALDI-TOF mass spectrometry for tentative protein identification using database search programs such as ProFound (http://prowl.rockefeller.edu) and/or 2) capillary HPLC followed by Edman sequence analysis to obtain definitive primary sequence information. While this strategy works well for Coomassie stained proteins it poses a problem for the more sensitive silver staining method. A technique has been published for protein identification using MALDI-TOF mass spectrometric analysis after reversal of the silver stain (Gharahdaghi et al., Electrophoresis 1999 20, 601-605). We have employed this technique for preparation of proteins for internal Edman sequence analysis. Because of the usually low amount of silver stained proteins, peptides need to be isolated by capillary HPLC and collection techniques need to minimize peptide loss. Silver stained gels, MALDI-TOF mass spectra, capillary HPLC, and Edman sequence data will be presented as well as guidelines for sample handling.

P80-M

Amino acid analysis 2000: a collaborative study from the ABRF AAA Research Group.

M.A. Alterman¹, P. Hunziker², K. West³, R. Harris⁴, D. Chin⁵; ¹Univ. of Kansas, 6038 Malott Hall, Lawrence, KS 66045, ²Univ. of Zurich, ³Cleveland Clin. Fndn., ⁴Genentech, ⁵Univ. of Missouri

The AAA Research Group of the ABRF periodically provides member laboratories with test samples in order to assist members in maintaining and improving the quality of AAA and assess the reliability of AAA. The latest study consists of two parts: Internet-based survey of equipment and technique currently used for AAA, and experimental data from this year's test sample. The experimental part of 2000 study focused on quantitation and identification of proteins. The participating laboratories received solutions of 4 pure proteins for the determination of their concentration by AAA as well as by a colorimetric method (BCA, Bradford, or similar). Data obtained will be used to examine the use of AAA for determining of the exact amount of protein in the sample and the use of compositional data to identify the protein. The relative precision and accuracy of the colorimetric assays will be compared with the AAA values. The comparison of the results of this study with the results from previous studies will show changes and improvements in the practice of AAA.

P81-T

How to obtain reproducible gradient capillary/nano LC for ultrahigh sensitivity MS detection of biological macromolecus with and without flow splitting.

Y-H. Jou, C. Wu, C. Wu, Y.W. Hong, F.J. Yang; Micro-Tech Scientific, Sunnyvale, CA, 140 South Wolfe Road, Sunnyvale, CA 94086

The applications of gradient capillary column LC for biological macromolecular analysis and drug discovery research have increased significantly in recent years. As a result of this increasing need for sub-picomole detection, capillary column LC with eletro-spray and nano-spray MS is increasingly important.

Flow stream splitting technique introduced in 1983 by van der Wal and Yang (1) has been utilized by many users for rapid evaluation and realization of the advantages of capillary LC and capillary LC-MS. However, the flow stream splitting technique has suffered from inherent poor retention time and gradient slope reproducibility when the following condition(s) occurred: high pressure, high split ratios, large sample amount of viscose sample solvent is injected, column inlet flow restriction change, or splitter flow restriction change.

Splitless flow capillary LC allows the same performance and ease of validation for routine work as the conventional 4.6 mm id. gradient HPLC. However, it requires each pump to deliver accurate/reproducible flow rates at sub-µl/min. It also requires high pressure mixer(s) with small volume for fast gradient generation and minimum solvent gradient delay time from the mixer to the column inlet.

This poster will discuss design concepts of a new reciprocating pump that delivers sub-µl/min flow rates for routine capillary/Nano LC applications. Performance of the system in terms of long-term retention time reproducibility for flow rates from 0.01 to 50 µl/min will be compared to a flow stream splitting system. Practical considerations in terms of gradient regeneration, system liquid end volume, mixer volume, mixing noise, and sample clean up, desalt, and concentrating will also be discussed.

1. Sj. van der Wal and F. J. Yang, J. Resolut. Chromatogr. Chromatogr. Commun. 6, 216 (1983).

P82-S

Proteomics: identification of low abundance proteins by MDLC/MALDI-TOF analysis.

M. Meys, S. Krishnan, K.C. Parker, M. Lin, M.D. Lynch, R. Carberry, K. Waddell; Applied Biosystems, Foster City, CA, 500 Old Connecticut Path, Framingham, MA 01701

The study of proteome components commonly involves the separation of a protein extract on a 2-D gel followed by the analysis of the protein spots. This analysis is currently being done by proteolytic digestion of the spots followed by mass spectrometric techniques. Although this approach has proven to be useful, it has several limitations. The main limitations are the inability to analyze low abundance proteins, proteins of extreme pI's and molecular weights. We present here a method that can overcome these limitations by eliminating the 2-D gel separation step. The process involves performing a multidimensional chromatographic separation of the protein extract followed by proteolytic digestion and MALDI-TOF mass spectrometric analysis of the digests. The removal of the gel step in the process offers accessibility to all classes of proteins, makes the process amenable for automation and enhances speed of analysis. The use of chromatography also improves peptide recovery. In addition, the use of the chromatographic separation enables the removal of the highly abundant proteins from the extract and aids in the identification of the low abundance proteins. In the present study Escherichia coli is used as a model to validate this approach. The crude extract was fractionated over a cation exchange column followed by further fractionation on a reverse phase column. The fractions from the reverse phase column were then digested with trypsin and subjected to MALDI-TOF mass spectrometric analysis. Using this approach we identified several low abundance proteins such as maltodextrin phosphorylase, leucine aminopeptidase, phosphate starvation inducible protein precursor. These proteins to our knowledge have not been located on 2-D gels of E coli indicating the validity and usefulness of the approach.

P83-M

An automated sequence assignment and multiple database searching.

T. Sasagawa, Y. Matsumoto, M. Kojima, Y. Mizuno; Toray Res. Ctr., 1111 Tebiro, Kamakura, Kanagawa 2488-8555, Japan

Mass spectrometry and Edman degradation are complementary to each other and are indispensable techniques in proteomics in which systematic analysis of a large number of expressed proteins is required. In order to facilitate the interpretation of these data, automated data analysis and multiple database search programs linked together are developed. PepMs is for de novo sequencing based on a new algorithm. The sequence information can be obtained even if a few gaps are present. Post-translationally modified amino acid and ambiguous Edman sequence data can be used to aid the sequence determination.

Seq is for the automated interpretation of Edman sequence data and for multiple database searching. By correcting raw data with extraction efficiency of PTH amino acids and with lag due to incomplete cleavage, unambiguous sequence assignment is possible. The assigned data are automatically subjected to database searching. A minor second sequence can be also determined. The program also has a routine to generate theoretically possible amino acid sequences based on observed PTH amino acids, when equal amount of multiple sequences are observed. Using the generated pairs of sequences, the correct pair of the sequence can be found by multiple database searching. The result is confirmed mass spectrometry and de novo sequence routine.

P84-T

Amino acid specific effects on C-terminal sequencing efficiency: comparison of activation chemistries.

D.R. Dupont¹, S.W. Yuen¹, R.L. Noble¹, K.S. Graham²; ¹Applied Biosystems, 850 Lincoln Centre Drive, Foster City, CA 94404, ²Beckman Res. Inst., City of Hope

Efficient initial activation of the C-terminus of a protein or peptide is essential to successful C-terminal sequencing. To facilitate the evaluation of activation chemistries, we have synthesized sets of peptides that contain each of the genetically coded amino acids (except proline) in three sequence positions, at the C-terminus, penultimate to the C-terminus and third from the C-terminus. The peptides are sequenced in groups of four or five to limit the number of sequencing runs necessary to evaluate the behavior of each of the amino acids in each sequence position. Included in each group are several amino acids with reactive side chains together with amino acids with unreactive side chains. The peptides are covalently attached to modified PVDF membrane to minimize the effect of sample loss due to washout.

In the ATH method, the C-terminus is reacted with acetic anhydride to form an oxazolone, which is then reacted with tetrabutylammonium thiocyanate in the presence of TFA vapor to form a C-terminal thiohydantoin. An alternative approach is the use of a reagent that can form thiohydantoin without first forming oxazolone, which should minimize the potential for side reactions and improve the sequencing initial yield. In the C-terminal sequencing chemistry developed at the Beckman Research Institute, one such reagent, diphenylphosphoroisothiocyanatidate (DPP-ITC), is used for activation and thiohydantoin formation. Several years ago we presented a preliminary comparison of these activation chemistries using several model proteins. Here we compare the activation chemistries for all the amino acids (except proline) using the peptide sets and outline the effects of incorporating DPP-ITC activation into the ATH chemistry. In addition, we compare the efficiency of the activation chemistries on a variety of proteins at different sample amounts.

P85-S

Rapid oligosaccharide mapping using fluorescent anthranilic acid detection.

S.T. Dhume, S.A. Batz; SmithKline Beecham, 709 Swedeland Rd, Mailcode UW2960, King of Prussia, PA 19406

Oligosaccharide mapping and characterization methods based on fluorescent Anthranilic acid (AA, 2-aminobenzoic acid) labeling affords high resolution and high sensitivity detection of glycans (1). The AA tagging method offers a significant improvement over other methods and is being rapidly adopted in the area of glycoprotein analysis.

The oligosaccharide mapping method starting with an N-linked glycoprotein requires an overnight enzymatic digestion, 1 hour reaction with AA, a purification step followed by a 2 hour chromatographic run. It is the intent of this work to allow rapid detection of oligosaccharides while retaining the quality and reproducibility of the original method.

The enzymatic release of oligosaccharides with the amounts of substrate and enzyme used is essentially complete in 3 hours. The glycan profiles obtained with PNGase F incubations between 30 min and 72 hours are similar except for the lower peak intensities at shorter (<2.5 hours) incubation times.

The purification step to remove excess AA may be omitted. The mapping is directly applicable to fetuin, a highly sialylated glycoprotein but needed gradient changes for neutral glycan species to retain resolution, tailing and quality of maps as those from the original method.

Oligosaccharide mapping was carried out on a short (2.1 mm X 15 cm) polymeric amine-bonded column allowing reduction in the time of analysis to about an hour. Initial results with reverse phase chromatography are promising and also allow 1 hour runs per sample.

The modifications may be used alone or in any combination. Without consideration of the time for enzyme incubation, the total time saved is more than 50%. In addition, 80% of HPLC solvent consumption is avoided. The reproducibility and universal applicability of this method for other oligosaccharides will be discussed in detail.

1. Anumula, K. R. and Dhume, S. T. (1998) Glycobiology, 8, 685-694.

P86-M

GlycoSuiteDB--a database of glycan structures.

C.A. Cooper, M.J. Harrison, M.R. Wilkins, N.H. Packer; Proteome Systems Ltd., North Ryde, Australia, 1/35-41 Waterloo Rd, North Ryde, NSW 2113, Australia

GlycoSuiteDB is a relational database that contains information from the scientific literature on glycoprotein derived glycan structures, their biological sources, the literature references used to obtain the information, and the methods used to determine each glycan structure. The main aims in the construction of GlycoSuiteDB are to present a consistent, up-to-date and reliable source of information. The database provides an essential resource for the glycobiologist and the protein chemist.

GlycoSuiteDB is available on the web at http://www.glycosuite.com. The web site allows the user to search the database using a combination of composition, monoisotopic or average mass, protein name, SWISS-PROT/TrEMBL accession number, species, biological system, tissue or cell type. There are at present no restrictions on the use of GlycoSuiteDB by non-profit organisations as long as its content is not modified in any way. Usage by and for commercial entities will require a licence after the initial free trial period on the web. Full conditions of use will be made available on the web site and through GeneBio (www.genebio.com), the exclusive worldwide distributor of GlycoSuiteDB.

P87-T

FMN is covalently attached to a specified threonine residue via phosphate group in the NqrB and NqrC subunits of Na⁺-translocating NADH-quinone reductase from Vibrio alginolyticus.

M. Maeda¹, M. Hayashi², Y. Nakayama², M. Yasui², K. Furuishi³, T. Unemoto²; ¹Applied Biosystems, Framingham, MA, ²Chiba Univ., ³Applied Biosystems, 4-5-4 Hatchobori, Chuo-ku, Tokyo 104-0032, Japan

Na⁺-translocating NADH-quinone reductase (NQR) from Vibrio alginolyticus is composed of six subunits (NqrA to NqrF). We previously demonstrated that both NqrB and NqrC subunits contain a flavin cofactor covalently attached to a threonine residue. Fluorescent peptide fragments derived from the NqrB and NqrC subunits were applied to a matrix-assisted laser desorption time of flight (MALDI-TOF) mass spectrometer and covalently attached flavin was identified to be FMN in both subunits. From post-source decay (PSD) fragmentation analysis, it was concluded that FMN is attached via phosphate group to Thr-235 in the NqrB and to Thr-223 in the NqrC subunits. The ester binding of FMN to the threonine residue reported here is a new type of flavin attachment to polypeptide.

P88-S

Disulfide bridge determination of SETI-IIa, a squash trypsin inhibitor.

V.M. Faca, L.J. Greene; FMRP-USP, Av. Bandeirantes, 3900, Ribeirao Preto, Sao Paulo 14049-900 Brazil

The squash trypsin inhibitor SETI-IIa, one of the smallest strong inhibitors described in the literature, contains 31 aminoacids residues of which 6 are cysteine residues. Its amino acid sequence is: EDRKCPKILMRCKRDSDCLAKC TCQESGYCG. It forms a compact tridimensional structure maintained by three disulfide bonds. Due to its small size, this inhibitor can be synthesized, using Fmoc chemistry. The reoxidation step which requires formation of the correct disulfide bonds is a critical step in obtaining the active inhibitor and knowledge of the correct disulfide pairing is required. We describe a procedure to determine the disulfide pairing of SETI-IIa using small amounts of protein. About 50 nmol (175 µg) of native SETI was submitted to digestion with thermolysin for 72 hours at 45°C in a 0.5 M MES buffer pH 6.5. The fragments obtained were submitted to RP-HPLC in a C₁₈ column (4.6 X 220 mm) in a TFA/Acetonitrile elution system. The major fragments were collected an identified by amino acid composition and sequencing by Edman degradation. The disulfide bridges obtained from native SETI-IIa and their yields were: Cys1-Cys4 (50%), Cys2-Cys5 (10%) and Cys3-Cys6 (38%). This procedure will permit us to characterize refolded synthetic analogs of SETI-IIA

Supported by FAPESP. Faça, V.M. has a FAPESP pre-doctoral fellowship.

P89-M

Identification of posttranslational modifications using MALDI-TOF-TOF.

J.P. DeGnore, J. Campbell, P. Juhasz, J. Peltier, M. Vestal; Applied Biosystems, 500 Old Connecticut Path, Framingham, MA 01701

To realize the full potential of proteomics, large numbers of proteins must be rapidly identified, usually separated by 2D gels. To investigate the function of these identified proteins it is important to determine the potential posttranslational modifications. It has been estimated that 80% of mammalian cells contain posttranslational modifications (1) and up to one-third of proteins expressed in a typical mammalian cell are thought to be phosphorylated (2). Identification of the site and type of modification in a protein provides insight into the activity and regulation of the posttranslationally modified protein. The location of posttranslational modifications within a peptide fragment can be determined from mass shifts of the fragment ions in tandem mass spectrum. Often the expression of posttranslationally modified proteins is low, requiring sensitive detection. Tandem mass spectrometry has become the method of choice for identification of posttranslational modifications due to it sensitivity and speed. The new MALDI-TOF-TOF mass spectrometer allows both rapid protein identification and the determination of the precise site of posttranslational modifications. For example, signature ions in the low mass range can serve as a tool to screen for phosphopeptides in a high throughput experiment. Typical fragmentation patterns will be presented for posttranslational modifications including phosphorylation. A number of processing tools developed to allow quality peak lists to be derived from the raw mass spectra will be presented.

(1) Kornfield, R.; Kornfield, S., Annu. Rev. Biochem. 1985, 54, 631-664.

(2) Hubbard, M.J.; Cohen, P. Trends Biochem. Sci. 1993, 18, 172-177.

P90-T

Immunoprecipitation as a method to enrich phosphoproteins destined for 2-D gel analysis.

T.R. Shephard¹, A. Ducret²; ¹Concordia Univ., Canada, 1455 de Maisonneuve West, Montreal, Quebec H3G 1M8, Canada, ²Merck Frosst

Our lab is interested in the construction of 2D phosphoprotein maps for determining the substrates of protein tyrosine phosphatase-1B (PTP-1B), a protein believed to play a role in the down regulation of the insulin receptor.

The construction of 2D phosphoprotein maps is made difficult by the fact that 1) phosphoproteins are intermixed with all cellular proteins, complicating gel analysis, and 2) phosphoproteins often show relatively low abundance, which can result in important proteins being overlooked.

To overcome the difficulties listed above, we attempted to enrich the phosphoprotein content of our cell lysates [obtained from control and insulin stimulated immortalized fibroblast cell lines derived from PTP-1B^+/+ and PTP-1B^-/- mice (Elchebly, M. et al. 1999 Science. 283, 1544-1548)] through immunoprecipitation procedures before proceeding with our 2D gel analysis. This was accomplished by pre-clearing total cell lysates and then immunoprecipitating under native conditions with a mix of i) monoclonal 4G10 antibody coupled to agarose beads and ii) biotin-conjugated RC20 antibody, as previously described by Lodish et al. (1999 Proc. Nat. Acad. Sci. USA 97, 179-184). This poster will illustrate the 1D and 2D gel analysis of tyrosine phosphorylated proteins immunoprecipitated under a variety of conditions from PTP-1B^+/+ and PTP-1B^-/- cell lines.

P91-S

The application of a novel precursor ion discovery method for the characterisation of phosphoproteins.

J.I. Langridge¹, J.B. Hoyes¹, R.H. Bateman¹, R.A. Carruthers¹, C. Jones¹, A. Millar¹, O.N. Jensen²; ¹Micromass UK Ltd., Floats Road, Wythenshawe, Manchester M23 9LZ, United Kingdom, ²Univ.of Southern Denmark, Odense Univ.

The huge increase in genomic sequence information available, combined with the increased sensitivity and selectivity provided by mass spectrometry, has allowed large-scale protein identification. However the analysis of the post translational modifications present on the identified proteins is a more challenging problem.

In this paper we describe a method that allows specific post translationally modified peptides to be identified and sequenced during the course of an HPLC experiment on the Q-Tof mass spectrometer. During the HPLC run the instrument is switched alternately at one-second intervals between low and high collision energy with argon in the collision cell. The quadrupole, MS1 operates in the rf only mode allowing the full mass range to be passed to the Tof. The first data set at low energy (4 eV) shows only the normal pseudo molecular ions. The second at higher energy contains all their fragments. Wherever a product ion of interest occurs in the high-energy data all its possible precursors are present in the corresponding low energy data. The mass spectrometer may then switch to MS/MS mode selecting the potential precursors in turn to reveal the true parent. In addition to creating a mass list of precursors from the low energy spectra the computer may also make a list of neutral losses from these precursors. The exact mass capability of orthogonal Tof's increase the specificity of the neutral loss particularly in the case of a mass deficient loss such as is observed with phosphopeptides. Appearance of the neutral loss causes the spectrometer to switch to MS/MS mode to acquire additional sequence information and confirmation of the neutral loss.

Examples of this methodology and its application to protein phosphorylation will be presented.

P92-M

Protein modifications in diabetics and individuals taking aspirin studied by mass spectrometric and functional assays of fibrinogen.

A.H. Henschen-Edman; Univ. of California, Irvine, Biological Sciences 2, Irvine, CA 92697-3900

It is a well-known fact that proteins are modified posttranslationally both intracellularly and extracellularly. Certain protein modifications may be specifically caused by diseases or drugs. In the case of the human blood plasma protein fibrinogen, a central protein in blood coagulation, both structural and functional properties are altered by glucose in diabetics and by aspirin in individuals on that kind of medication. For the structure analysis fibrinogen was S-pyridylethylated, cleaved with cyanogen bromide and the resulting 36 fragments partly separated by gel filtration chromatography; the pools were then analyzed by MALDI-TOF mass spectrometry. Glucose and aspirin modify amino groups by glycation and acetylation respectively, thereby adding the corresponding 162 and 42 amu to the original mass of the fragment. Under in vitro conditions extensive modifications and dose-dependent increases in the degrees of modification were observed. Additional evidence for the reaction with glucose was obtained by sequence analysis of tryptic fragments retained by a boronate affinity column; several predicted lysines were missing, as they were present in the glycated form. For the function analysis fibrinogen was clotted with thrombin and the proteolytic cleavage as well as the polymerization-dependent increase in absorbance monitored. The effect of glycation and acetylation on the proteolysis was marginal, but both caused an extensive and dose-dependent decrease in the polymerization reaction. These effects are expected to be of pathophysiological relevance.

P93-T

Analytical strategies for the identification of phosphorylation and O-linked glycosylation sites in proteins.

A. Furtos-Matei¹, J. Kihlberg², P. Thibault¹; ¹Natl. Res. Council, Canada, ²Umea Univ.

Glycosylation and phosphorylation are between the most common post-translational modifications in proteins. Over the past decade, tandem mass spectrometry has become the method of choice for mapping attachment sites. In contrast with N-linked glycans, oligosacharides attached to ser/thr residues tend to fragment at the glycosidic bond on MS-MS to leave an unmodified peptide backbone, thereby making assignment of modification sites more difficult. A similar situation is also encountered for phosphopeptides analysis where the interpretation of positive ion MS-MS can be more complicated in view of fragment showing a loss of H₃PO₄ moiety. In this work, synthetic glycopeptides and phosphopeptides were used to develop a method that allows simultaneous determination of phosphorylation and O-linked glycosylation sites at serine and threonine residues. The approach we propose is based on a ?-elimination reaction followed by a Michael type addition of a nucleophilic reagent, which is further used as a marker of the modification site. Among other reagents, the volatile triethylamine was selected to induce the elimination of both, phosphate groups and O-linked glycans. Several nucleophilic reagents have been tested and evaluated by mass spectrometry and capillary electrophoresis methods with respect to the chemical yield and the specificity toward the targeted functional groups. The relative reactivity of different attachment sites as a function of the amino acid nature and the type of modification is also discussed. This approach has been applied to enzymatic digestion mixtures of model glyco- and phosphoproteins.

P94-S

Comparative phosphopeptide mapping from gel-derived proteins using a multidimensional mass spectrometry approach.

R.S. Annan, M.J. Huddleston, F. Zappacosta, C.I. Susan, S.A. Carr; SmithKline Beecham Pharmaceut., Box 1539, King of Prussia, PA 19406

Within the last ten years mass spectrometry has emerged as a key technology in the difficult task of mapping protein phosphorylation sites. Previously we have described a comprehensive multidimensional phosphopeptide mapping strategy which uses mass spectrometry to isolate and identify phosphorylated peptides. We have now significantly improved the sensitivity of this approach by incorporating capillary HPLC chromatography (0.18 mm i.d.) and micro-ionspray mass spectrometer. The current configuration allows us to detect and identify less than 200 fmol of phosphopeptides from SDS-PAGE derived proteins. Because the first dimension of the analysis utilizes an LC separation coupled to on-line phosphopeptide detection, it generates a phosphopeptide profile. In this report we will show how such a profile can be used for comparative purposes, providing a semi quantitative evaluation of how the phosphorylation state of a protein changes under a given set of differing conditions. The comparative profile makes it easier to identify and concentrate analysis efforts on those phosphorylation sites which are relevant to the changing conditions and ignore those which are not.

P95-M

Structural properties of DsRed, a fluorescent protein from coral.

L.A. Gross¹, G.S. Baird¹, D.A. Zacharias¹, R.C. Hoffman¹, K.K. Baldridge², R.Y. Tsien¹; ¹UCSD, 9500 Gilman Dr., La Jolla, CA 92093-0647, ²San Diego Supercomputer Ctr.

With the cloning of a red fluorescent protein from the Discosoma species of marine corals (Matz et al. 1999), an entire spectrum of colors are now available as gene reporters in cell microscopy. Similar to green fluorescent protein (GFP) derived from the jellyfish Aequorea, with which it shares only 30% homology, DsRed develops its fluorescence without the need for exogenous enzymes or co-factors. Studies are on-going to optimize its properties, especially for applications involving fluorescence resonance energy transfer (FRET).

DsRed is an oligomer of four beta-barrels, as shown by analytical ultracentrifugation (Baird et al. 2000). Tandem mass spectrometry has shown that formation of the red chromophore involves a second step of oxidation, in comparison to the cyclization and oxidation which forms GFP (Gross et al. 2000). This extends the conjugation by adding an acylimine functional group at the 2-position of the imidazolidinone ring of the GFP chromophore. Initially upon expression, DsRed emits at green wavelengths. The green color diminishes as the protein matures to its red fluorescent form, in a process taking 3 to 4 days. However, the protein appears not to convert the total population of chromophores to the mature, red form. In the tetramer, FRET occurs between immature green (as donors of resonance energy) and mature red emitters (as acceptors), accounting for the observed fluorescence spectra.

Matz, M. V., Fradkov, A. F., Labas, Y. A., Savitsky, A. P., Zaraisky, A. G., Markelov, M. L. & Lukyanov, S. A. (1999) Nature Biotechnology 17, 969-973.

Baird, G. S., Zacharias, D. A. & Tsien, R. Y. (2000) Proc. Natl. Acad. Sci. USA 97, 11984-89.

Gross, L. A., Baird, G. S., Hoffman, R. C., Baldridge, K. K. & Tsien, R. Y. (2000) Proc. Natl. Acad. Sci. USA 97, 11990-95.

P96-T

Mass spectrometry analysis of the kinetics of in vivo rhodopsin phosphorylation.

K.A. Lee, G.A. Niemi, J.B. Hurley; Univ. of Washington, Howard Hughes Medical Institute, Box 357370 Health Sciences Building, Seattle, WA 98195

Upon light stimulation, rhodopsin, the light-sensing protein in the rod cells of the retina, is phosphorylated at several sites as the first step in its deactivation. We developed a mass spectrometry-based method to determine the kinetics of this phosphorylation in vivo by quantitating the extent of phosphorylation at each site. After exposing a freshly dissected mouse retina to a flash of light, we use a rapid-quench device is used to quickly terminate phosphorylation and dephosphorylation reactions by homogenizing the retina in 8 M urea. The samples are washed and proteolyzed to release the carboxyl-terminus of rhodopsin, which contains all known phosphorylation sites. The resultant peptides are analyzed by LCMS. Synthetic monophosphorylated and unphosphorylated peptide standards were used to show that the mass spectral response to the rhodopsin peptides is linear over a range from 10 fmol to 100 pmol. The relative sensitivities of the mass spectrometer to unphosphorylated, singly, doubly, and triply phosphorylated peptides were determined using synthetic peptide standards. The resultant correction factors were used to determine the relative quantities of the peptides in the actual retinal samples. LCMS/MS was utilized to identify three primary sites of phosphorylation, S334, S338, and S343. Peptides monophosphorylated on S334 were separable from those monophosphorylated on S338 and S343 by reverse phase HPLC with our standard C18 column chromatography conditions. However, the S338 and S343 monophosphorylated peptides coeluted. We demonstrated that the relative amounts of each species in the single peak could be determined by monitoring the ratio of specific daughter ions characteristic of each peptide.

P97-S

An improved procedure using immobilized metal-ion affinity chromatography for isolation and characterization of phosphopeptides from phosphoproteins.

K-L. Hsi, D.R. Dupont, S.W. Yuen, C.A. Settinery; Applied Biosystems, Foster City, CA, 850 Lincoln Center Drive, Foster City, CA 94404

It has become clear that phosphorylation is the most common and important reversible regulatory modification of proteins. Defining the sites of phosphorylation is necessarily important for the mechanism studies of cell regulation. However, methods to specifically isolate and characterize phosphopeptides remain challenged, especially for those proteins separated from 1D/2D gel at low concentration. An attempt is thus made to improve the procedure based upon immobilized metal-ion affinity chromatography (IMAC) for working with sub-microgram amount of phosphoproteins.

A protease digest of gel-separated phosphoprotein was treated with immobilized Fe⁺⁺⁺ on a chelating resin. The affinity bound phosphopeptides were then eluted from the resin with ammonium acetate buffer. We use a mini-size (250 µl) of cartridge to perform all the processes of isolation of phosphopeptides including washing of the resin, binding and elution of the phosphopeptides, instead of an open column as previously reported.

This improved procedure provides some advantages over the previously reported open column method: 1. Fast, i.e. 30 minutes or less. 2. High washing efficiency, thus high recovery of pure phosphopeptides. 3. Sensitive working level, i.e. sub-microgram of gel separated phosphoproteins.

Several phosphoproteins with different types of phosphorylation (Ser, Thr and Tyr) have been studied using the improved procedure. The effectiveness of the procedure was demonstrated by chemical sequencing and MS/MS analysis of MicroBlotter-collected phosphopeptides.

P98-M

Postsynthesis labeling of agiotensin II peptide with d-rhodamine for fluorescence monitoring of receptor binding.

S.P. Yadav, W-Z. Shen, Y. Luo, J. Zhang, S. Karnick; Lerner Res. Inst., Cleveland, Mail Code NC10, 9500 Euclid Avenue, Cleveland, OH 44195

Solid-phase peptide synthesis is now routinely used for drug discovery and immunological research and for studying the structure-function relationships of proteins. Protein-protein interactions play an important role in a wide variety of biochemical and physiological processes and generally involve large interfaces with many intermolecular contacts. In addition, protein-protein interaction may also occur through small surface binding epitopes such as in the case of human growth hormone-receptor binding and erythropietin-receptor complex formation. Therefore, the rational design of relatively small molecular size peptide activators/inhibitors for receptor surfaces has long been considered a formidable challenge. These findings convincingly open up the possibility that small peptides that mimic such small binding epitopes may help to block a large protein-protein interface. Usefulness of this hypothesis in drug design rationale remains to be further tested in different biological systems. Furthermore rationalization of general approach of inhibition of protein-protein interaction will of course have a tremendous impact on development of new therapeutic strategies for several human diseases. That is where fluorescent dye labeling of synthetic peptides becomes important. Studies on labeling of a peptide with fluorescent dyes in a manner without altering functional characteristics of the peptide are of considerable interest for studying the peptide-receptor interactions. Strategy for synthesis of an octa-mer angiotensin II by Fmoc chemistry and subsequent labeling of the cleaved peptide with Rhodamine dye are discussed in this report. The results of double labeling are presented here to show that the removal of excess free Rhodamine is critically important in confocal microscopy.

P99-T

Synthesis of lactam-bridged dipeptides mediated by aminoacylpyroglutamates.

I. Rodionov, A. Chulin, V.T. Ivanov; Br. of Shemyakin-Ovchinnikov Inst. of Bioorganic Chem., 8 Academy Avenue, Pushchino, Moscow Region 142290, Russian Federation

Aminoacyl incorporation reaction (AI) discoverd about 40 years ago by Shemyakin et al. has been evaluated as a general synthetic route to variously constrained lactam-bridged dipeptides. The AI is effectively a ring enlargement, which involves intramolecular acylation of amino group (or different nucleophilic functions like HS- and HO-) by the activated cyclic diacylamino moiety via intermediate formation of bicyclic azacyclols. The most straightforward models for studying AI are aminoacylated pyroglutamates derived from diaminoacids

R-Xaa-Glp-OR' (I), Xaa = Lys, Orn, Dab and Dpr,

a class of unusual peptides, which remains unexplored as yet. We have studied 3 different synthetic approaches to I:

direct acylation of sodium derivative of Glp-OR";

pyrrolidone ring closure in the plain dipeptides R-Xaa-Glu(OX)-OR" promoted by bases and/or by activation of side chain carboxyl of Glu;

selective oxidation of the related proline dipeptides Boc-Xaa-Pro-OR' by ruthenium tetroxide/sodium periodate.

Scope and limitation of the above approaches will be discussed and exemplified by the synthesis of 26 protected dipeptides I. As expected, deprotection of amino or hydroxy functions followed by exposure to potassium carbonate buffer, pH 9.5 (water-acetonitrile) resulted in smooth AI for the majority of the synthesized dipeptides I. No oligomerization products were detected. In this way a number of bridged dipeptides (9-12 membered cycles) were obtained in 35-70% yield and characterized by NMR and MS data. NMR data suggested that in 4 cases stable azacyclols have been isolated.

P100-S

The characterization of prostate specific antigen activating peptide.

M. Pakkala¹, P. Wu², J. Leinonen², U-H. Stenman², J. Vepsäläinen¹, A. Närvänen¹; ¹Univ. of Kuopio, P.O. Box 1627, Kuopio 70211, Finland, ²Univ. Central Hosp., Helsinki

Prostate specific antigen (PSA) is widely used as a marker of prostate cancer. PSA is a 30 kD serine protease and belongs to Kallikrein family. Peptides with specific binding properties to PSA were studied by using phage display peptide libraries. Four different PSA binding clones were isolated and characterized. One of the clones, containing 13 amino acids sequence with four cysteines, showed the highest affinity for PSA. This peptide (C4), as a fusion protein with glutathione-S-transferase (GST), increased the protease activity of PSA against a synthetic substrate (Wu et al. (2000), Eur. J. Biochem. 267, 1-10)

We have synthesized the active peptide C4 (CVAYCIEHHCWTC) and its modifications by using conventional solid phase peptide synthesis method (SPPS). Two cysteines (2. And 3.) were protected by using Acm protecting group, which remains uncleaved during the removal of the peptide from the resin. The final cyclization was made during the removal of the Acm-group from cysteines by Iodine treatment. The synthetic analog showed same activity to PSA than the original GST fusion protein where as the activity of the peptide with two Acm groups was less than 50% from the fully cyclized form. The acetylation of the ?-amino group did not affect to the activity. In addition we synthesized cyclic version with amide bond VAYCIEHHCWT instead of the cysteines 1. and 4. by using Allyl protected E(6) (Fmoc-L-Glu-OAll) as a starting amino acid. The final peptide was cyclic with one disulfide bond. This modification was totally inactive.

The results suggest that the correct secondary structure of the peptide C-4 plays an important role in increased activation of PSA. Structure analysis by NMR is under study. Based on the preliminary NMR results the prepared peptides are rather flexible and the structures are strongly dependent on the temperature.

P101-M

Implementing surface plasmon resonance biosensors in drug discovery.

D.G. Myszka; Univ. of Utah, 50 N. Medical Dr./School of Medicine Rm 4A417, Salt Lake City, UT 84132

Recent improvements in instrument hardware, experimental design, and data processing make it possible to utilize surface plasmon resonance (SPR) biosensor technology in the discovery and development of small-molecule drugs. The key features of SPR biosensors, real-time binding analysis and lack of labeling requirements, make this technology suitable for a wide range of applications. Current instruments have a throughput of ~100-400 assays per day, providing a complement to high-throughput screening. The ability to collect kinetic data on compounds binding to therapeutic targets yields new information for lead optimization. Small-molecule analysis and emerging applications in the areas of ADME and proteomics have SPR biosensors poised to play a significant role in the pharmaceutical industry.

P102-T

Study of the protein-DNA interaction responsible of the carbon catabolite repression in Lactobacillus casei.

C.D. Esteban¹, K. Mahr², G. Pérez-Martínez¹, W. Hillen², F. Titgemeyer²; ¹Agrochem. and Food Technol. Inst., Burjassot, Spain, Apartado de correos 73, Burjassot, Valencia 46100 Spain, ²Erlangen-Nürnberg Univ.

In the industrially relevant lactic acid bacterium Lactobacillus casei the preferential utilization of carbon sources is controlled by the mechanism of carbon catabolite repression (CCR). As in other low-G+C gram-positive bacteria, CCR takes place through the binding of the transcriptional repressor, CcpA, to an operator called cre (catabolite responsive element). CcpA binding to cre sequences is enhanced by its correpressor HPr-ser46-P.

It was our aim to characterize this protein-DNA interaction by Surface Plasmon Resonance (SPR). For this purpose CcpA was overexpressed and purified both with an N-terminal poly-histidine tag and without this tag. HPr was overexpressed, purified and in vitro phosphorylated. A synthetic biotinylated double stranded oligonucleotide containing the cre sequence present in the promoter of the lac operon of L. casei was immobilized on a streptavidin sensor chip. SPR experiments were performed flowing a range of CcpA (with and without his tag) concentrations over the chip in the presence and absence of saturating amounts of HPr-ser46-P. The experimental parameters for data acquisition were optimized and equilibrium and semiquantitative kinetic analysis allowed the calculation of KD and estimation of kinetic association and dissociation rate constants for protein-DNA interaction.

The experimental parameters for the study of this protein-DNA interaction by SPR have been established. Binding of unmodified CcpA to the cre sequence was characterized by equilibrium and kinetic rate constants. The presence of the his tag was shown not to interfere (results with and without his tag are comparable). In conclusion, we have developed a powerful system to accurately characterize the intermolecular relationships of histidine tagged CcpA that could allow efficiently characterizing mutations in CcpA in the near future.

P103-S

Mapping of protein-protein interactions using mass spectrometry.

D. Figeys, H. Duewel; MDS-Ocata, Toronto, 480 University Avenue Suite 401, Toronto, Ontario M5G 1V2, Canada

Functional proteomics is becoming the next generation of large-scale proteomic approaches. It is based on the concept that the function of a protein is defined by its interactions. Therefore, large-scale approaches for the mapping of protein-protein interactions at the cellular level will be essential to the comprehensive understanding of the interactome, i.e. the protein-protein interactions related to a proteome. Here we will present an approach for the large-scale screening of the interactome using mass spectrometry. This technology identifies proteins involved in specific protein-protein interactions using protein identification by mass spectrometry. We will also discuss the incorporation in the process of novel technology, such as the ICATtm technology, for the rapid and relative quantitation of proteins by mass spectrometry.

P104-M

Affinity chromatography and mass spectrometry in dissecting EGFr signaling interdicted by the quinazoline EGFr inhibitor OSI-774.

J.D. Haley¹, A. Thelemann¹, H. Pan¹, D. Fenyo²; ¹OSI Pharmaceut. Inc., 106 Charles Lindbergh Blvd., Uniondale, NY 11553, ²Proteometrics LLC

The blockade of EGFr signaling by the quinazoline tyrosine kinase inhibitor OSI-774 was investigated by direct affinity chromatography, reverse-phase chromatography and mass finger-printing. Phosphotyrosine complexes from OSI-774 and control treated HN5 squamous carcinoma cells were prepared by both Triton X-100 and RIPA lysis. Protein fractionation by SDS-PAGE was compared with capillary HPLC. Protein fractions were proteolytically digested with either trypsin, GluC or LysC, desalted by microC18 reverse phase tips and subjected to matrix assisted laser desorption- time of flight mass spectrometry. C4 chromatography greatly improved the signal strength and resolution of the peptide spectra obtained, when compared to direct MALDI MS of digested immunoaffinity fractions.

Eighty-eight spectra were evaluated from five HN5 phosphotyrosine protein complex chromatography separations. Mass analysis was performed using a PerSeptive DE-Pro mass spectrometer using a-cyano-4-hydroxycinnamic acid or dihydrobenzoic acid matrices. Data were analyzed using RADARS, a SQL compliant database search engine allowing comparison across experiments of proteins identified under different experimental conditions (e.g. trypsin vs. GluC). The predominant protein identified was the epidermal growth factor receptor (EGFr) which was found in the major C4 HPLC protein fraction. Phosphorylation on both P1 and P2 C-terminal tyrosines was readily observed by mass spectrometry. A large number of well known and less defined proteins which either (1) contain phosphotyrosine or (2) form stable complexes with targets proteins were identified from multiple experiments. The Triton X100 lysis and affinity capture methodology allows the identification of known and novel protein complexes not detected by gel-based techniques.

P105-T

Innovations in proteome analysis: biomolecular interaction analysis mass spectrometry.

R.W. Nelson, D. Nedelkov; Intrinsic Bioprobes Inc., 625 S. Smith Rd. Suite 22, Tempe, AZ 85281

The relative ease and fast pace by which the genomic data has been gathered stands out against the complexity of the proteome world and the involvedness required for protein characterization. Even though significant strides have been made lately in several protein characterization techniques, novel technologies and multiplexation of the existing ones are required to fully address the many sides of the proteome.

Biomolecular Interaction Analysis Mass Spectrometry (BIA/MS) is a two-dimensional chip-based analytical technique geared toward quantitative and qualitative detection and analysis of small volumes of biological samples. In the first (functional) dimension, BIA/MS takes a form of micro-scale planar-affinity chromatography performed on a sensor-active surface. Surface plasmon resonance (SPR) is used for detection of biorecognition events that occur at a sensor surface/solution interface between an immobilized biomolecule (the ligand) and its interacting partner (the analyte) present in the sample solution. For the second (structural) dimension, BIA/MS employs MALDI-TOF MS analysis. With minimal physical modifications and thorough application of MALDI matrix, SPR-active sensor surfaces are converted to amenable MALDI target. The ensuing mass spectrometry analysis serves the purpose of validating the SPR sensing data by providing the molecular mass of the retained analyte and it yields other qualitative information about the SPR-monitored interaction, such as identification of non-specific binding, binding of analyte variants/fragments and multi analyte binding. Presented here are results from the utilization of BIA/MS in detection of a number of biological markers found in complex biological fluids. Small volumes of human plasma and urine were analyzed for cystatin C, beta-2-microglobulin, urinary protein 1, retinol binding protein and transthyretin, exploring the effectiveness of BIA/MS in simultaneous detection of clinically related biomarkers. Issues such as limit of detection, recognition of protein complexes and delineation of non-specific binding were also explored.

P106-S

Dissecting structure-function relationships in RNA/protein interaction using biocore.

I.A. Laird-Offringa¹, P.S. Katsamba¹, D.G. Myszka²; ¹Univ. of Southern California, 1441 Eastlake Ave., Los Angeles, CA 90089-9176, ²Univ. of Utah

RNA-binding proteins play critical roles in gene expression and regulation at the post-transcriptional level. While much is known about the various naturally occurring RNA-binding motifs, and co-crystal structures of a number of RNA/protein complexes are available, very little is known about the dynamics of RNA/protein interactions. We have used the spliceosomal protein U1A and its RNA target in the U1 small nuclear RNA (U1hairpinII or U1hpII) as a model to study the kinetics of RNA/protein interaction. Using the previously solved structure of the U1A/U1hpII complex, we have engineered a series of mutants designed to probe the roles of electrostatics, hydrogen bonding, aromatic stacking, and RNA loop length, all of which have been implicated in formation of the U1A/U1hpII complex. The effects of these mutations on the binding dynamics were studied using BIACORE, which yielded high quality kinetic data about the interaction. We determined that neutralization of positive charges on the protein slows the association rate and reduces the deleterious effect of salt on complex formation. In contrast, removal of hydrogen-bonding or stacking interactions within the RNA/protein interface, or reducing the size of the RNA loop, increases the dissociation rate. Our data support a mechanism of binding consisting of a rapid initial association based on electrostatic interactions and a subsequent locking step based on the hydrogen bonding and stacking interactions that occur during the induced fit of RNA and protein. Our results demonstrate the power of BIACORE to dissect the functional differences between structural features of two interacting macromolecules.

P107-M

Miniature integrated surface plasmon resonance biosensor for characterization of protein-protein interactions.

M.L. Stolowitz, G. Li, K.P. Lund, J.P. Wiley; Prolinx, Inc., 22322 20th Avenue SE, Bothell, WA 98021

The utility of a miniature integrated surface plasmon resonance biosensor is described. The disposible device is approximately the size of a thumbnail and houses all of the optics and electronics needed to acquire surface plasmon resonance sensorgrams. The gold sensor surface has been modified so as to minimize nonspecific binding and utilizes Versalinx Chemical Affininty Tools to facilitate the immobilization of macromolecular targets for binding studies. The utility of the biosensor is demonstrated in conjunction with a prototype data acquisition interface and a simple orbital shaking device.

P108-T

The study of peptide-peptide interaction by ion-mobility MALDI.

A.S. Woods¹, J. Koomen², M.A. Huestis¹, K.J. Gillig², D.H. Russell², A.J. Schultz³, K. Fuhrer³, M. Gonin³; ¹NIDA, NIH, 5500 Nathan Shock Drive, Baltimore, MD 21224, ²Texas A&M Univ., ³Ionwerks Inc.

We showed in previous work that MALDI could be used to study peptide-peptide interactions. Matrices such as ATT [pH 5.4] do not disrupt non-covalent interactions, while more acidic matrices such as CHCA [pH 2.0] do. Our study found that one peptide had to have 2 adjacent Arg [RR] or an Arg-Lys-Arg [RKR] motif and the other had to have a minimum of two adjacent Glu [EE] or Asp [DD] in order to form a complex.

In this work we used MALDI/mobility/TOF mass spectrometry to further study the formation of these non-covalent complexes [NCX]. The instrument consists of a short drift tube (the ion-mobility cell) in which is applied an electric field which causes the MALDI ions to drift through a Helium carrier gas (2 torr) into an orthogonal TOF mass spectrometer. Ion-mobility [IM] separates gas phase ions on the basis of their collision cross section-to-charge ratio, when combined with mass spectrometry it can be a powerful instrument for structural studies to determine the conformations of biomolecules.

Dynorphin (a 17 a.a. peptide that contains an RR motif) and several other peptides containing two RR or RKR motifs formed NCX with acidic peptides (containing 2 or more adjacent Glu or Asp) and were detected by IM and TOF MS. We discuss the capabilities and limitations of this new instrument when applied to the gas phase conformational study of MALDI desorbed NCX, as well as the advantages of looking at peptide mixtures by IM in addition to MALDI-TOF MS

P109-S

Protein molecular communications involving two or more different ligands studied by microcalorimetry.

M.L. Doyle; Bristol-Myers Squibb Pharmaceut. Res. Inst., Mail Stop H13-07, 206 Provinceline Rd., Princeton, NJ 08543-4000

Biophysical methods are capable of characterizing the functional chemistry of protein-ligand interactions at a high, molecular level of resolution. Typically, biophysical studies have focused on interactions involving one protein with a single ligand. However, the biological roles of many proteins are likely to be regulated by interactions with multiple ligands. In this tutorial we describe how biophysical methods, particularly Isothermal Titration Calorimetry (ITC), can be used to begin to deconvolute the interdependency that different multiple ligands have on a protein's function. Two examples will be presented. One case involves comparing the binding interactions of a TNF cytokine with four different members of the TNF receptor superfamily in order to predict which receptor may be most biologically relevant (1). The other example describes a case where two different small molecules regulate the binding of one another to an enzyme (2).

References

1) Truneh, A., Sharma, S., Silverman, C., Khandekar, S., Reddy, M.P., Deen, K.C., Mclaughlin, M.M., Srinivasula, S.M., Livi, G.P., Marshall, L.A., Alnemri, E.S., Williams, W.V. and Doyle, M.L. (2000) J. Biol. Chem. 275, 23319-23325.

2) Du, W., Liu, W-S., Payne, D.J. and Doyle, M.L. (2000) Biochemistry 39, 6003-6011.

P110-M

Protein biochips: powerful new tools in proteomics.

K.L. Witte¹, F. Zaugg¹, P. Indermuhle¹, L. Ruiz-Taylor¹, N. Tolani¹, B. Muehlbauer¹, P. Wagner¹, S. Nock²; ¹Zyomyx, Inc., 3911 Trust Way, Hayward, CA 94545, ²Zyomyx, Inc.

With the imminent sequencing of the human genome there is a growing need for technologies, which can study the resulting wealth of gene products. While technological innovation has adapted the analysis of genetic material to a highly parallel and miniaturized format, the more delicate nature of protein structures has precluded the development of analogous devices. However, to achieve a fundamental understanding of biochemical pathways, high-throughput biology has to be expanded to protein analysis, proteomics and multi-target screening.

We have developed high-density protein microarrays for quantification of multiple proteins in complex mixtures. These microarrays contain 10.000 individually addressable features in a 1 cm X 1cm area. The implementation of new surface chemistries allows the immobilization of exactly defined quantities of proteins on each spot while retaining the activity of the protein. Using this platform we have developed a multiplexed, microchip-based immunoassay to analyze expression levels of serum proteins. We demonstrate the binding of proteins of interest with very low non-specific binding from non-cognate proteins using fluorescence as readout. Detection limits on this microassay are equal to or lower than commercial ELISA tests and reduce the sample volume by many orders of magnitude.

P111-T

Long-range fragment sizing on MegaBACE™ genotyping instrument.

M. Minarik, K. Pirkola, M. Mahtani; Molecular Dynamics, 928 E. Arques Ave., Sunnyvale, CA 94085

Over the past decade, DNA fragment analysis by capillary electrophoresis has become a powerful alternative to classic sizing by slab gel electrophoresis. The most important applications include microsatellite genotyping, AFLP fingerprinting, differential display and SNP typing. Due largely to slab gel limitations, most current protocols are based on fragment sizing only 300-500 bp. Extending this size range is desirable in order to accelerate throughput by increasing either the depth of microsatellite multiplexing, or the effectiveness of fingerprinting and differential display applications. Recently published results from long-read DNA sequencing suggest a possibility to separate DNA fragments over a large size range with high resolution (Zhou et al., Anal Chem 2000). The key factors are proper composition of separation matrix and optimal selection of experimental parameters defining injection and separation conditions.

In the present work, we have evaluated a sizing precision of large fragments up to several kilobases. Using optimized running conditions, we demonstrate routine sizing up to 1 kb with 2 base resolution in 120 minutes. For longer fragments, the sizing resolution was between 5 and 10 bases with analysis times under 3 hours. We present the influence of various experimental conditions on the peak resolution and its impact on the overall quality of sizing.

P112-S

High throughput SNP genotyping on MegaBACE™ using multiple, time-spaced injections.

A. Shuster¹, D. Shen¹, Y. Tsunoi¹, M. Minarik¹, K. Pirkola¹, K. Jones¹, T. Deldot¹, R. Belcinski¹, A. Mamone², K. Dains¹, M. Mahtani¹; ¹Molecular Dynamics, 928 E. Arques Ave., Sunnyvale, CA 94085, ²Molecular Dynamics

We have developed a method for high throughput SNP genotyping on the MegaBACE™ 1000 capillary electrophoresis platform. The system takes advantage of the instrument's flexibility to load 12 96-well sample plates over a period of about 30 minutes, using a proprietary method of repeated, time-spaced injections. Samples are prepared in a single-tube SNuPe™ (Single Nucleotide Primer Extension) reaction which contains the four different fluorescently-labeled dideoxynucleotides (terminators), Thermosequenase enzyme and optimized reaction buffer. After single base extension and cleanup, SNP products are loaded onto the MegaBACE by automated electrokinetic injection. Multiple, pulsed injections of up to 12 different SNP marker plates are then loaded into the same polymer matrix, spaced at two minute electrophoresis intervals. Separation provides the advantage of excellent signal-to-noise, predictable and characteristic marker genotype patterns, and good discrimination of negative controls over sample failures. 1300 SNP genotypes can be obtained with a turnaround time of less than 2 hours; we expect that further color and size multiplexing will greatly increase throughput. A characteristic injection marker is added to each sample prior to injection, acting both as internal control and also demarcating each injection from the next. A separate software package, SNP Profiler automatically processes the signal data and outputs the maximum likelihood SNP genotypes, and includes a user interface for editing and verification. The system can be completely switched between any three applications--high throughput sequencing, SNP typing or microsatellite genotyping--in less than five minutes.

P113-M

An automatable magnetic particle-based kit for the removal of dye terminators.

D.A. Spicer, K.A. Hughes, R.J. Kaiser, A.L. Springer; Prolinx, Inc., 22322 20th Avenue South East, Bothell, WA 98021

Prolinx^® Inc. has developed the RapXtract™ dye terminator removal system that is fully automatable for high throughput purification of cycle sequencing reactions. The kit is based on a novel magnetic particle format allowing for rapid purification of up to 384 samples at one time, depending on the robotic platform. The kit provides reproducible, high-quality results without the need for modified primers. The RapXtract kit is easily automated, as it does not require centrifugation, vacuum filtration, or multiple wash steps. The kit has been automated on several platforms including the TECAN Genesis, and the Tomtec Quadra 384 model 320.

P114-T

Applications of an electric field-assisted capillary LC technique and instrumentation for biological macromolecules analysis.

F.J. Yang, Y-H. Jou, C. Wu, C. Wu, Y.W. Hong; Micro-Tech Scientific, 140 South Wolfe Road, Sunnyvale, CA 94086

Recent rapid growth in the needs for biological macromolecules analysis and drug discovery research has increased significantly the utilization of capillary column separation techniques such as capillary column LC, capillary column electrophoresis, and capillary column electro-chromatography etc. The growth in the applications of capillary column separation techniques is also driven by the commercialization of electro-spray and nano-spray LC-MS interfacing techniques that allow mass spectra-analysis of biological macromolecules. Major advantages of capillary column separation techniques are:

1). Improvement of detection limit by more than 2000 times in comparison to that possible using conventional 4.6 mm id columns. The smaller the column diameter, the lower the detection limits. It allows femto-mole detection of trace amount of biological samples.

2). Direct interface to MS has greatly increased the applications of MS for biological sampler analysis.

3). It reduces solvent consumption by more than 2000 times.

4). 2D LC-MS can be easily configured for rapid sample clean up, de-salt, matrix elimination, and sample concentrating. It allows fast capillary LC-MS for high through put sample analysis.

5). Capillary columns are common to capillary LC, CE, and CEC, it facilitates an unified methodology and instrumentation for harness the advantages of capillary column separation utilizing both differential electrophoretic migration of charge molecules and the differential mobility of the sample molecular zones carried by either electro-osmosis flow or hydrodynamic flow.

This presentation will discuss the concept of an unified capillary column separation methodology and the design of the instrumentation that can perform conventional CE and CEC. It can also be utilized for both low pressure and high pressure gradient (pH, buffer strength, and organic composition gradient) CE and CEC applications. Examples for the applications of the unified capillary separation system in the electrical- field assisted high pressure capillary gradient LC applications for biological macromolecules analysis will also be presented.

P115-S

Systematic validation and optimization of capillary electrophoresis for high performance genotyping and fragment analysis.

D. Shen, M. Minarik, A. Shuster, K. Pirkola, S. Gopalan, K. Dains, M. Mahtani; Molecular Dynamics, 928 E. Arques Ave., Sunnyvale, CA 94085

Capillary electrophoresis systems have quickly replaced traditional fluorescent slab gels for DNA analysis in many laboratory environments. The technology transition has been abrupt and, on the up-side, has excitingly advanced the Human Genome Project. On the down-side, the transition for many labs has occurred without a clear understanding of the strengths and weaknesses of the new technology, methods for troubleshooting, and experience in setting benchmarks. We have spent the past year validating and optimizing the capillary electrophoresis platform, the MegaBACE for high throughput genotyping and fragment analysis. In the development of the system, we have devised a set of metrics and guidelines that predict the system performance.

We present data and benchmarks on variables including optical response, dynamic range, sizing precision and reproducibility, resolution, throughput, capillary and sample success rates, and other critical parameters of system performance. These metrics offer a better understanding of proper system function, and indices for perturbations of that function. We also provide data on the performance of the software and the effects of modifications to the workflow on the genotype outputs. In addition, we present a process control system (including software and chemistry) that can be used to evaluate the system performance or to test new methods and protocols before integrating them in a production environment. In collaboration with others, we will present data on the system's performance for fragment analysis applications including microsatellite genome scanning, AFLP analysis, SNP genotyping, and differential display.

P116-M

Long sequence readlengths with the MegaBACE 1000 DNA Analysis System.

J. Ellerbrock¹, M. Reagin¹, A. Mamone², J. Nelson¹, B. McArdle¹, C. Fuller¹; ¹AP Biotech, 800 Centennial Ave., Piscataway, NJ 08855, ²Molecular Dynamics

MegaBACE 1000 DNA Analysis System is a high-throughput, fluorescence-based DNA sequencer that utilizes capillary electrophoresis with 96 capillaries operating in parallel. The system automates gel matrix replacement, sample injection, DNA separation, and data analysis with integrating software. The sensitivity of DYEnamic ET energy transfer terminators and the superior resolving power of linear polyacrylamide (LPA) separation matrix allow detection over a broader range of template amounts. Likewise, the robust performance of Thermo Sequenase II DNA polymerase improves success rates.

The MegaBACE 1000 has the ability to sequence 96 samples in just 2 hours, perform up to 9 runs per day, and can be used in both high-throughput and core DNA analysis facilities. In high-throughput facilities it can produce as much as 500,000 raw sequence bases a day. We have developed protocols that allow reads of over 1000 bases ideally suited for final overnight runs (4-6 hours) when such sustained high-throughput is not required. We will present a variety of examples and protocols to obtain optimal results on the system. Data will be presented showing long DNA readlengths on a variety of different templates (BAC, plasmid, M13, high GC, or PCR product) using the DYEnamic ET terminator cycle sequencing kit and the MegaBACE 1000 system.

P117-T

New separation medium for the ABI Prism^® 3700 DNA Analyzer and ABI Prism^® 3100 Genetic Analyzer.

B.F. Johnson¹, K.O. Voss², J.N. Tian², J.A. Fisher²; ¹Applied Biosystems, 850 Lincoln Centre Drive, Foster City, CA 94404, ²Applied Biosystems, Foster City, CA

A new separation medium for the ABI PRISM^® 3700 DNA Analyzer and ABI PRISM^® 3100 Genetic Analyzer has been developed. The new polymer formulation has a viscosity similar POP-5™ and POP-6™ formulations previously released for DNA separations on our capillary instruments and consequently has similar capillary filling times and sheath flow characteristics. High quality separations can still be obtained with bare capillaries with a minimum of 100 separations before cleaning or replacement of the capillary array. Compared to POP-5™ separation medium using the default run conditions for POP-5™ on the ABI PRISM^® 3700 DNA Analyzer, the new separation medium results in a 20% decrease in electrophoretic separation time while increasing one-base peak resolution by 100 bases. On the ABI PRISM^® 3700 DNA Analyzer with a 50 cm array the new polymer formulation under these conditions results in an average length of read of 850 bases at 10 runs per day. On the ABI PRISM^® 3100 Genetic Analyzer the new polymer formulation results in an average length of read of 825 bases in a 50 cm array at 16 runs per day. Basecalling length of reads obtained from ABI PRISM^® Sequencing Analysis (1% error) and from TraceTuner 1.1 analysis software will be compared at different separation voltages, distances, and temperatures. Run parameters that yield the maximum length of read or maximum number of runs per day will be presented.

P118-S

Modifications to 96-well DNA preparation kits for the MegaBACE 1000.

R. Dhulipala¹, B. McArdle¹, A. Mamone²; ¹AP Biotech, 800 Centennial Ave., Piscataway, NJ 08855, ²Molecular Dynamics

With the introduction of high-throughput DNA sequencing instruments, methods for simultaneous preparation of template samples have come into common usage. Several groups have developed plasmid preparation methods enabling the user to easily decrease the time to prepare 96 DNA samples and thus increase sequencing throughput. The majority of these kits are based upon traditional alkaline lysis of bacterial culture yet differ in their method of purification of DNA. We have evaluated several of these methods for preparing templates for the MegaBACE 1000 capillary instrument. We find that results obtained with capillary sequencing instruments are dependent upon the quality and amount of the starting template. Hence, results might vary greatly due to the method of DNA preparation.

We have found that most of the preparation methods and kits are configured to optimize DNA yield, not well-to-well consistency. Fortunately, the DNA yields are greater than necessary and simple modifications to the standard protocols can be made that improve consistency without reducing yields significantly. Each purification method is evaluated for amount and consistency of DNA yield, and sequencing performance using DYEnamic ET terminators and the MegaBACE 1000. In addition, we make recommendations for each method to achieve optimal success rates with the system.

P119-M

Converting from slab gel to capillary electrophoresis: a user's guide for genotyping.

C.L. Brown, B.F. Johnson, C. Wike, K. Roy, A. Wheaton, Y. Wang, N. Caffo; Applied Biosystems, 850 Lincoln Centre Drive, MS 404-1, Foster City, CA 94404

Capillary electrophoresis has rapidly become the platform of choice for human disease research, clinical diagnostics and population genetics studies due to its ease of operation, automation and increased throughput. Historically, most protocols were developed on slab gel instruments due to their flexibility, high throughput and consistency, yet required considerable cost and labor. The ABI PRISM^® 3100 Genetic Analyzer, a 16-capillary electrophoresis instrument, was introduced to match the throughput and reproducibility of slab gels while offering lower run costs, less human intervention and greater sensitivity. Laboratories considering the conversion from slab gel systems should be mindful of a few distinctions between these two platforms. Herein, we will discuss modified sample preparation, fragment mobility differences and suggestions for multiplexing schemes. We will also highlight enhancements to data analysis, which reduce existing rate limiting steps.

P120-T

Sequencing reaction purification of BAC DNA using a combination of molecular weight cut-off membrane and a new grade of Sephadex™ G-50.

R. Dhulipala, A. Kumar; AP Biotech, 800 Centennial Ave., Piscataway, NJ 08855

The purification of DNA sequencing reactions by ethanol precipitation or gel filtration prior to the loading of automated fluorescent DNA sequencing instruments is a standard technique for the removal of unincorporated dye terminators. One standard gel filtration method that we have developed into a ready-to-use 96-well format with pre-swollen DNA Grade Sephadex™ G-50 is AutoSeq96 (please see our companion poster). AutoSeq96 purification is six times faster than ethanol precipitation, is easy to use, and minimizes handling because samples are purified directly into 96-well collection plates. However, one aspect of DNA sequencing reaction purification that has remained cumbersome for some types of samples containing large DNA constructs is the additional removal of template DNA.

We have developed a new type of gel filtration method for purifying DNA sequencing reactions using the basic format of the AutoSeq96 plate but with the added benefit of template removal. Elimination of template DNA prior to electrophoresis can both increase the overall success rate of the reactions and improve read lengths for large plasmids and BACs. This new gel filtration method, called PureSeq™96, has been optimized in 96-well format using a high molecular weight cut-off membrane (MWCOM) combined with pre-swollen low conductivity Capillary Electrophoresis Grade Sephadex G-50. The combination of the unique properties of the MWCOM and this new grade of G-50 are expected to improve sequencing success rates of both large construct plasmids and BACs using MegaBACE™ 1000. Described in this poster are the results of our studies and some of the development issues considered during the design of this gel filtration method for DNA sequencing reaction purification.

P121-S

DYEnamic ET terminator reagents and protocols for the ABI 3700 DNA sequencer.

C. Palaniappan¹, L. Hosta¹, M. Zahn¹, G. Kaput¹, J. Hockenberry¹, M. Campion¹, A. Brito¹, B. McArdle¹, A. Mamone²; ¹AP Biotech, 800 Centennial Ave., Piscataway, NJ 08855, ²Molecular Dynamics

DYEnamic ET terminator sequencing products have long been available for the MegaBACE 1000 sequencing instrument and the ABI 310, 373, 377 instruments. The method offers several significant advantages for routine DNA sequencing. DYEnamic ET terminators afford high sensitivity sequencing through energy transfer technology. Thermo Sequenase II DNA polymerase has unprecedented tolerance to salt and contaminants that may be present in template preparations. Rapid and efficient incorporation of dITP by this enzyme allows short cycle times and robust, compression-free sequencing of GC rich templates.

We have developed sequencing reagents incorporating ThermoSequenase II DNA polymerase and ET terminators for the ABI 3700 DNA sequencer. An optimized mobility file for DYEnamic ET terminator dyes, spectral matrix standards and protocols have been optimized for this instrument. Data will be presented that allow direct comparison of this chemistry on the ABI 3700 DNA sequencer with other available chemistries and instruments.

MegaBACE, DYEnamic and Sequenase are registered trademarks of Amersham Pharmacia Biotech.

P122-M

Development of a fifth-dye labeled short size standard for SNP interrogation.

S.S. Kuo, G. Ayanoglu, D. Wei, C-a. Chang, R. Brown, A.B. Tomaney, P. Dong; Applied Biosystems, 850 Lincoln Centre Drive, Foster City, CA 94404

We describe the development of a short size standard for analyzing small fragments on fluorescent detection systems. This standard contains 9 fragments ranging from 15 to 120 nucleotides, and is labeled with a fifth dye. During rigorous testing with various media on gel or capillary electrophoresis platforms, we are able to achieve excellent precision and curve-fitting cross-platform. The fifth dye is spectrally well resolved from other dyes.

This standard is designed in particular to enable automated data analysis in methods for single nucleotide polymorphism (SNP) detection such as single-base extension (SNE) assays. The presence of the size standard will minimize the need for manual data analysis especially in multiplexed electrophoresis assays. In our poster, we will demonstrate its utility in SNE assays. With the combination of different fragment lengths and four-color chemistry, the potential for multiplexing SNP loci exists for large-scale genotyping with minimal optimization.

P123-T

Whole mass determination of two-dimensional gel separated lens crystallins by electrospray ionization mass spectrometry.

Y. Ueda, L.L. David; Oregon Hlth. Sci. Univ., 611 SW Campus Dr., Portland, OR 97201

Crystallins, the major proteins of the eye lens, have long half-lifes, and the accumulation of post-translational modifications can cause insolubilization and cloudiness in the lens, a condition called cataract. The identity of modified crystallins on two-dimensional electrophoresis (2-DE) gels can be determined by peptide mass finger printing, but the incomplete recovery of peptide fragment sometimes limits the determination of protein modification. The purpose of this study was to develop a method to efficiently elute modified crystallins from 2-DE gels so that their masses could be measured by LC-ESI-MS.

Proteins from rat lens were separated by 2-DE and visualized by negative zinc stain. Excised spots were incubated in 0.1% SDS and then finely crushed to 20 mm particles by forcing them through a metal frit. Proteins were then allowed to diffuse into a non-detergent containing solution during sonication. The masses of eluted proteins were determined using a 1 mm C4 reverse-phase column (Vydac) to remove bound SDS and on-line ESI-MS. Using this method, we were able to determine the cleavage sites of either N-terminally or C-terminally truncated crystallins, and also identify phosphorylated species from single spots of moderately abundant proteins. The method was much simpler and resulted in higher yields compared to electroelution, and should be useful for analyzing a wide range of 2-DE separated proteins. We are currently testing whether µLC columns can be used to increase the sensitivity of the method.

P124-S

Protein separation optimization in large-format NuPAGE bis-tris gels.

R.A. Bogoev, J.W. Amshey; Invitrogen Corp., San Diego, 1600 Faraday Avenue, Carlsbad, CA 92008

Large format gels are frequently employed to obtain better resolution of proteins that have similar apparent molecular weights in SDS-PAGE. We have quantitatively determined the resolution of proteins using the distance of separation and the band width to establish a relationship between the run distance and the resolution for NuPAGE^® bis-tris gels. Resolution increased in a nonlinear fashion. Increasing the run distance three-fold led to less than a 2-fold increase in resolution. Increased run length did not improve the resolution of larger proteins with the 10% acrylamide concentration used. Increasing crosslinker concentration 3-fold led to an improvement in resolution for medium and small proteins but led to a decrease in resolution of larger proteins. Further increase did not improve the resolution for medium and small proteins and further reduced resolution for larger proteins. Over the range of 10-50°C, resolution was optimal at a temperature of 20 degrees. Decreasing the buffer concentration in the NuPAGE bis-tris gels provided a faster separation with no effect on resolution. For most applications, due to the relatively small increase in resolution with increasing run length, shorter gels may be preferable because they are more convenient for post-run processing.

P125-M

A high sensitivity silver stain without glutaraldehyde sensitization for protein gels used in mass spectrometry.

S.E. Whitney, A. Bautista, J.W. Amshey; Invitrogen Corp., San Diego, 1600 Faraday, San Diego, CA 92008

Conventional sensitive silver stains use glutaraldehyde as a sensitizer to enhance silver deposition in the protein zone. This sensitizer reacts with basic amino acids (i.e. Lysine) which are target sites for trypsin cleavage, and can result in reduced yields of peptides when the protein is processed for identification by mass spectrometry. A silver stain kit (SilverQuest™) has been developed which provides high sensitivity and good peptide recovery when stained gel bands are used for protein identification in mass spectrometry. Following staining and isolation of a protein band, the silver is removed by oxidation with potassium ferricyanide and complexation of Ag⁺ with sodium thiosulfate. At the 5 picomole level of BSA oxidation of metallic silver resulted in an increase in the number of identified peptides from 11 to 32, compared to 17 with a conventional glutaraldehyde-sensitized stain. Percent coverage rose from 22 to 41% of the sequence. For L-glutamate dehydrogenase we observed 24 peptides and 48% sequence coverage while 31 peptides and 34% coverage was found for Phosphorylase B. In addition, a rapid and sensitive procedure employing the use of a microwave has been developed which allows for sensitive staining in <30 min. The typical sensitivity of this new silver stain was found to be 0.2 ng for BSA, 0.4 ng for Ovalbumin, 0.3 ng for Soybean trypsin inhibitor and 0.4 ng for a-Lactalbumin.

P126-T

Proteomics of brown spider venom.

M.V. Sousa¹, C.A.O. Ricart¹, W. Fontes¹, K.C. Barbaro², L. Machado¹, R.B. Cunha¹; ¹Univ. of Brasilia, Campus Universitario, Brasilia, DF 70910-900, Brazil, ²Butatan Inst.

Brown spiders are found in several regions of the world. In Brazil, the brown spider Loxosceles gaucho is very common, and have caused severe accidents due to a potent dermonecrotic and lethal venom. Little is known about the composition of the venom proteome. Proteomic analysis was employed to identify some proteins in the venom in order to gain new insights on its mechanism of action.

Venon was separated by two-dimensional electrocphoresis using IPG gels in the first dimension and SDS-PAGE for the second dimension. The resulting gels were scanned and analysed via the ImageMaster software. Sixty protein spots were detected. Most of the spots were found in the 25-35 kDa range. Proteins such as insectotoxins, dermonecrotic toxin and annexin IV were identified by either mass spectrometry or chemical sequencing. The dermonecrotic toxin was purified by gel filtration, anion exchange FPLC and RP-HPLC. At least four isoforms of the toxin exist in the venom. The construction of a venom gland cDNA bank is under way.

P127-S

Identification of endosperm proteins by peptide mass mapping: patterns of accumulation during wheat grain development.

W.H. Vensel, W.J. Hurkman, C.K. Tanaka; USDA, 800 Buchanan Street, Albany, CA 94710

As part of a project with the goal of decreasing the impact of environmental conditions on wheat productivity and quality, we are using proteomics to identify key events and metabolic pathways in wheat grain development. Wheat (Triticum aestivum L. cv. Butte 86) plants were grown in a climate-controlled greenhouse that had an average daily maximum daytime temperature of 25°C and nighttime temperature of 17°C. Water and fertilizer were supplied by drip irrigation. Developing heads were tagged at anthesis. Grain was harvested at pre-selected time points and endosperm collected. Salt soluble proteins were extracted and separated by 2-D gel electrophoresis. Gel patterns for the different developmental stages were characterized using computer-based image analysis. As expected, a number of proteins increase, decrease, and were newly synthesized during grain development. Interestingly, basic proteins increased significantly throughout the developmental time course. Proteins were identified using Western blots, Edman degradation, and MALDI mass spectrometry. These proteins are involved in storage protein biosynthesis, protein folding, glycolysis, starch biosynthesis, and pest protection.

P128-M

Proteomics: facts, fictions and applications.

S.J. Cordwell; Macquarie Univ., Australia, Level 4, Building F7B, Macquarie University, New South Wales 2109, Australia

Proteomics as a science is dependent on finding molecules of significance from amongst highly complex biological mixtures. Technically, however, the protein 'pool' from which information can be gathered is dramatically lower than that predicted by genome sequencing. This is due to limiting factors defined by protein abundance, hydrophobicity, mass and pI, in association with two-dimensional gel electrophoresis. Furthermore, the variable nature of gene expression in vivo means that predicting the 'functional' proteome under a given set of conditions is dependent upon a thorough knowledge of biochemical pathways encoded by a genome and influenced by the surrounding environment. We have now begun to define the physical boundaries determined by these parameters to predict whether proteins will be visible using 2-D gel 'arrays' consisting of cellular pre-fractionation via relative protein solubility or cellular location, in combination with narrow-range immobilised pH gradients. The latter allows lower abundance proteins to be characterised utilising the sensitivity of mass spectrometry. Furthermore, we can now predict whether proteins are expressed via knowledge of physical characteristics within given detection parameters and upon the presence or absence of pathway components determined via nearest-neighbour predictions. Such predictions are based upon defining each translated gene within four theoretical sets. These interacting sets provide the real 'pool' from which 2-D electrophoresis can be expected to provide utility, while proteins outside the 'pool' must be annotated via alternative means. Such predictions of biologically significant molecules that remain outside the scope of 2-D electrophoresis then allow the researcher to target single proteins or sets of proteins utilising complementary technologies including peptide labelling or monoclonal antibodies. Here we present the advantages and disadvantages of 2-DGE-MS proteomics in comparison and in complement to alternative technologies and describe the application of high-throughput technologies for biological experimentation in a variety of microorganisms.

P129-T

Isolation and identification of rat liver proteins using ultracentrifugation with Nycodenz and 1D/2D-SDS-PAGE.

K. Murayama¹, T. Fujimura¹, M. Morita², N. Shindo¹; ¹Juntendo Univ. Sch. of Med., 2-1-1, Hongo, Bunkyo-ku, Tokyo, Tokyo 113-8421, Japan, ²Hitachi Koki Co., Ltd.

The use of 2D-SDS-PAGE as a clinical molecular scanner of various tissues and physiological fluid samples has proved useful. However, each organ contains more than 4,000 proteins and accordingly, it is almost impossible to study the functional role of these proteins unless separated. Our ultimate goal is to use 2D-SDS-PAGE as a clinical molecular scanner to define each organelle in various organs.

In this study, we report the isolation of rat liver organelles by density gradient centrifugation using Nycodenz. Nycodenz solution at 10, 20, or 30% concentration, containing 0.25 M sucrose (as an osmotic balancer), was added to each centrifuge tube and allowed to stand overnight at -20 to -80°. The solution was thawed at room temperature (>2 h), and analyzed to construct a density gradient curve. When used in a 5-ml tube, Nycodenz gradient densities from the top to the bottom without any centrifugation were 1.0334 to 1.2188 at 10%, 1.0506 to 1.2878 at 20% and 1.0856 to 1.3199 at 30%. Liver homogenate (0.4 ml, 4 mg) was loaded on the Nycodenz gradient solution and centrifuged at 28,000 rpm for 20 min using a Hitachi ultracentrifuge CP100alpha-RPS40T-2. The mixture was fractionated by a fractionator, DGF-U, its absorbance measured at 360 nm with a spectrophotometer and density with an Abbe refract meter. Next, 5 µl of each fraction was applied onto 10% gel for 1D-SDS-PAGE, electrophoresis, and the gel was stained by silver nitrate. Another 1D-SDS-PAGE was erector-blotted to a PVDF membrane and the presence of organelles was confirmed using antibody of the marker protein for each organelle.

P130-S

Rapid and simple single nanogram detection of glycoproteins in polyacrylamide gels and on electroblots.

W.F. Patton, T.H. Steinberg, K.N. Berggren, K. Pretty On Top, C. Kemper, Z. Diwu, R.P. Haugland; Molecular Probes Inc., 4849 Pitchford Avenue, Eugene, OR 97402

The fluorescent hydrazide, Pro-Q Emerald 300 dye, may be conjugated to glycoproteins by a periodic acid Schiff's (PAS) mechanism. The glycols present in glycoproteins are initially oxidized to aldehydes using periodic acid. The dye then reacts with the aldehydes to generate a highly fluorescent conjugate. Reduction with sodium metabisulfite or sodium borohydride is not required to stabilize the conjugate. Though glycoprotein detection may be performed on transfer membranes, direct detection in gels avoids electroblotting and glycoproteins may be visualized 2-3 hours after electrophoresis. This is substantially more rapid than PAS labeling with digoxigenin hydrazide followed by detection with an anti-digoxigenin antibody conjugate of alkaline phosphatase, or PAS labeling with biotin hydrazide followed by detection with horseradish peroxidase or alkaline phosphatase conjugates of streptavidin, which require more than eight hours to complete. Pro-Q Emerald 300 dye is spectrally compatible with SYPRO Ruby protein gel stain, allowing two-color detection of glycosylated and nonglycosylated proteins on the same gel or blot. Both fluorophores are excited with mid-range UV illumination. Pro-Q Emerald 300 dye maximally emits at 530 nm (green) while SYPRO Ruby dye maximally emits at 610 nm (red). As little as 300 pg of a1-acid glycoprotein (40% carbohydrate) and 1 ng of avidin (10% carbohydrate) or glucose oxidase (12% carbohydrate) are detectable in gels after staining with Pro-Q Emerald 300 dye. Besides detecting glycoproteins, as little as 2-8 ng of lipopolysaccharide is detectable in gels using Pro-Q Emerald 300 dye while 250-1000 ng is required for silver staining. Detection of glycoproteins may be achieved in 1-D or 2-D gels and on PVDF or nitrocellulose membranes.

P131-M

Selective depletion of major serum proteins and fractionation prior to 2-dimensional differential gel electrophoresis.

J.J. Cummings, E. Rohde, P.R. Griffin; Merck Res. Labs., RY800-B210, Rahway, NJ 07065

Two dimensional differential gel electrophoresis (2DIGE) is a powerful technique for the study of protein expression in physiological fluids such as serum. However, the presence of a few major proteins interferes with the separation and detection of many low abundant, yet physiologically important proteins. Albumin (52%), IgG (20%), IgA (2.5%), IgM (1.6%), transferrin (3.6%) and alpha1-antitrypsin (1.6%) are the major constituents of serum in many species.

Our objective was to selectively remove abundant proteins in a few sequential steps followed by the division of the depleted sera into multiple fraction based on their hydrophobicity (RP-HPLC) and/or charge (IEX-HPLC) prior to pre-electrophoresis fluorescent labeling.

The effective removal of albumin was accomplished using Cibachron blue dye spin columns. Immunoglobulins (IgG, IgM) were removed by gel filtration over protein A and G columns. Affinity resins specific to transferrin and alpha1 antitrypsin were prepared in house and used for the depletion of the respective proteins. Chromatographic fractionation was carried out on large-bore columns (4.6 X 100 mm). During all depletion and fractionation steps emphasis was placed on maximizing protein recovery. The protein concentration was monitored spectrophotometrically and effectiveness of depletion was assessed by SDS-PAGE.

The separation of the depleted and fractionated sera by 2DIGE resulted in a significant increase in the dynamic range of the separation. Utilizing this approach proteins were detected in areas previously obscured by major serum constituents. Furthermore an increased number of proteins was observed. Combined with the power of differential protein mapping using fluorescent dyes the procedure has shown great utility in the search for differentially expressed proteins. We present examples of coupling 2DIGE with µLC-MS/MS and database searching for the identification of surrogate markers in sera.

P132-T

Rice tissue proteomics: towards a functional analysis of the rice genome.

S. Komatsu, S. Shen, Z. Li, G. Yang, H. Konishi, M. Yoshikawa, R. Rakwal; Natl. Inst. of Agrobiol. Resources, 2-1-2 Kannondai, Tsukuba, Ibaraki 305-8602, Japan

The technique of proteome analysis with two-dimensional polyacrylamide gel electrophoresis (2D-PAGE) has the power to monitor global changes that occur in the protein expression of a tissue, an organism, and/or under stresses. In this study, proteins extracted from endosperm, embryo, root, callus, green shoot, etiolated shoot, leaf sheath and panicle of rice were separated by 2D-PAGE. The separated proteins were electroblotted onto a polyvinylidene difluoride membrane. The N-terminal amino-acid sequences of 117 out of 377 proteins were determined in this manner. N-terminal regions of the remaining proteins could not be sequenced and they were inferred to have a blocking group at N-terminus. Internal amino-acid sequences of 260 proteins were determined using the protein sequencer or matrix-assisted laser desorption/ionization time-of-flight mass spectrometry after enzyme digestion of proteins. Finally, a data-file of rice proteins was constructed, which included information on amino-acid sequence and sequence homology. Using this experimental approach, we could identify the major proteins involved in growth and development of rice. Some of these proteins, including a calcium-binding protein, which turned out to be carleticulin in rice, have functions in signal transduction pathway. The information thus obtained from amino-acid sequence of these proteins will be helpful in predicting the function of the proteins and for their molecular cloning in future experiments.

This work was supported in part by a grant of Rice Genome Project PR-1201, MAFF, Japan.

P133-S

Application of human cDNA microarray for the assessment of gene expression in mouse livers.

G.S. Huang, M-Y. Hong, P-C. Yang; Pig Res. Inst., Taiwan, PO Box 23, Chunan, Miaoli, Taiwan, ROC 350, Taiwan

A genomic survey for differentially expressed genes was performed to livers of ApoE deficient mice using human cDNA microarray containing approximately 9,000 human cDNA clones. Due to the homology between mouse and human, hybridization was performed at lower stringency condition, 10 degrees below the regular hybridizing temperature. Gene expression profiles of livers corresponding to high levels of blood cholesterol were generated at genomic scale. Thirty-seven genes were randomly selected from a pool of differentially expressed genes and subjected to semi-quantitative RT-PCR, further confirmed the result from microarray hybridization. These included genes associated with atherosclerosis, and novel genes that implied novel pathways correlated to high levels of blood cholesterol. It is promising using human cDNA microarray, the most complete collection among all species, to study other mammalian systems with satisfying speed and accuracy.

P134-M

Planar waveguide microarrays for increased sensitivity in gene expression monitoring.

W. Price¹, A.P. Abel², N.G. Scharer-Hernandez², A.M. Zrolka¹, A. Schafer¹, H. Schwarz¹; ¹QIAGEN GmbH, 28159 Avenue Stanford, Valencia, CA 91355, ²Zeptosens AG

The successful use of current microarray technology is impacted by its limited sensitivity, low signal to noise ratio and problems in reproducibility. In many bioanalytical applications--especially in toxicology, drug metabolism, and disease diagnostics--the amount of biological material that is available for an assay is limited. Reliable and robust methods for the quantitative detection of less than 1 µg of total RNA (equivalent to about 1 mg of tissue or 100,000 cells) are required in order to obtain valid experimental data. To fulfill these requirements currently available microarray detection approaches apply enzymatic target or signal amplification schemes, which bear among others the risk of biased amplification and non-linearity of the result. By combining Zeptosens' planar wave guide (PWG) gene chip technology with QIAGEN's expertise in sample preparation, extremely high levels of sensitivity and enhanced signal to noise ratios can be achieved. With this integrated approach, mRNA derived from the equivalent of 100 HeLa cells could be monitored without signal and/or target amplification.

P135-T

Meeting the demands for high throughput custom microarray production with the GeneMachines^® OmniGrid™.

K.S. Small, D. Henninger, G. Barragan, J. Dabrowiak, J. Fayet-Faber; GeneMachines, 935 Washington St., San Carlos, CA 94070

As researchers in both independent labs and core facilities continue to approach the developing field of genomics with custom microarray production, the need for microarrayer instruments that maximize flexibility and throughput without compromising cost has become critical.

The OmniGrid microarrayer offers a fully integrated and automated solution for creating custom DNA arrays. The OmniGrid is capable of printing 20,000 spots/slide onto 100 slides using 48 pins in under 7 hours. Key components of the instrument's flexibility include software features that allow random access of unique samples within a microtiter plate and inter-plate access among the 74 microtiter plates for production of fully customized microarrays. Data attesting to the instrument's throughput, precision, and accuracy are discussed. Solutions to ensure microarray consistency during extended print runs through minimizing the effect of environmental variables are also explored.

P136-S

An investigation on reproducibility of GenChip expression methodology in UVa BMRF.

B.P. Dragulev, P. Gallagher, Y. Bao, J.F. Fox; Univ. of Virginia, Jordan Hall, Box 441, Charlottesville, VA 22908

A new service for microarray gene expression analysis based on the Affymetrix Gene Chip System is being developed in the Biomolecular Research Facility at University of Virginia. Concerns from investigators over the reproducibility of this high cost procedure have been raised that need to be addressed. In order to gain certain insight of the systematic performance of this system, we conducted a study focusing on reproducibility of the expression profiles obtained from repetition of one experiment under the same conditions. We used as a model VMM5 human melanoma cell line as the source of RNA. Five total RNA samples were extracted from five flasks of VMM5 cells grown under the same conditions. Labeled cRNAs were synthesized following Affymetrix procedure and hybridized to Affymetrix test chips. Statistical analyses were performed on the data based on differential intensity on both normal and log scales. The reliability of the data sets was assessed by SPSS Reliability Analysis software and other software. Another goal of this study was to compare the results produced from two different types of GeneChips, i.e., HuGene FL arrays and Human Genome U95A arrays in order to access the feasibility of cross-type comparison.

P137-M

Dissecting structure-function relationships in RNA/protein interaction.

P.S. Katsamba¹, I.A. Laird-Offringa¹, D.G. Myszka²; ¹Univ. of Southern California, Norris Cancer Center/1441 Eastlake Ave., Los Angeles, CA 90089-9176, ²Univ. of Utah, 50 N. Medical Dr./School of Medicine Rm 4A417, Salt Lake City, UT 84132

RNA-binding proteins play critical roles in gene expression and regulation at the post-transcriptional level. While much is known about the various naturally occurring RNA-binding motifs, and co-crystal structures of a number of RNA/protein complexes are available, very little is known about the dynamics of RNA/protein interactions. We have used the spliceosomal protein U1A and its RNA target in the U1 small nuclear RNA (U1hairpinII or U1hpII) as a model to study the kinetics of RNA/protein interaction. Using the previously solved structure of the U1A/U1hpII complex, we have engineered a series of mutants designed to probe the roles of electrostatics, hydrogen bonding, aromatic stacking, and RNA loop length, all of which have been implicated in formation of the U1A/U1hpII complex. The effects of these mutations on the binding dynamics were studied using BIACORE, which yielded high quality kinetic data about the interaction. We determined that neutralization of positive charges on the protein slows the association rate and reduces the deleterious effect of salt on complex formation. In contrast, removal of hydrogen-bonding or stacking interactions within the RNA/protein interface, or reducing the size of the RNA loop, increases the dissociation rate. Our data support a mechanism of binding consisting of a rapid initial association based on electrostatic interactions and a subsequent locking step based on the hydrogen bonding and stacking interactions that occur during the induced fit of RNA and protein. Our results demonstrate the power of kinetics to dissect the functional differences between structural features of two interacting macromolecules.

P138-T

A rapid and efficient BAC DNA purification method for high throughput applications.

C.M. Smith, A. Pryor; Princeton Separations, 920 HW 33, Building 7, Suite 6, Freehold, NJ 08724

We have developed a method for the rapid and efficient purification of BAC DNA. Typical yields range from 0.5 to 1.0 µg from an overnight culture with Absorbance (OD600) = 5.0. BAC DNA was purified from an overnight culture of E. coli in 5 mL Terrific Broth (Cm 20 µg/mL). The culture was subject to gentle alkaline lysis, mixed with one volume of binding buffer, bound to a solid phase membrane, then eluted in a low salt buffer. Purified BAC DNA contains little or no chromosomal DNA or RNA visible after agarose gel electrophoresis and is suitable for downstream applications including restriction enzyme digestion and sequencing. A discussion of the protocol and accompanying data will be presented. The simplicity of the method makes it suitable for tailoring to a 96 well format providing a reliable method for high throughput applications

P139-S

Automation for reliable RNA purification using silica-gel-membrane technology.

F. Siegman, C. Schade, A. Wehren; QIAGEN Inc., 28159 Avenue Stanford, Valencia, CA 91355

An automated system has been specially developed for laboratories requiring purification of high-quality RNA from animal and human cells for high-throughput gene-expression analysis, including microarray and real-time RT-PCR. This robotic workstation is designed to maximize reproducibility for all sample preparation steps (e.g., cell lysis, RNA purification, reaction setup), ensuring reliable results in subsequent analysis. CVs of less than 3% are consistently observed for quantitative TaqMan analysis when using this system for sample preparation. RNA purification using silica-gel-membrane technology allows recoveries of more than 90%, even with limiting starting material (e.g., as few as 10 cells). Purified RNA is free from enzyme-inhibiting impurities for excellent performance in the most sensitive quantitative applications. Optimized, ready-to-run protocols allow up to 192 samples to be purified in 90 minutes. The accurate liquid-handling system provides cross-contamination-free pipetting and small-volume liquid transfer, making it suitable for reaction setup and other liquid-handling tasks such as sample rearray. The easy-to-use operating system automatically tracks samples and records all process data generated in each run for complete documentation. Data is stored in standard formats and can be exchanged with other instruments, such as thermal cyclers or gridding robots, via network or disk.

P140-M

Cotton biotechnology in China.

B-H. Zhang; Cotton Res. Inst., Chinese Acad. of Agr. Sci., Baibi, Anyang, Henan 455112, People's Republic of China

During the past two decades, China has made great progress in cotton biotechnology and genetic engineering. Obtaining firstly regenerative plants from cotton anther and protoplast culture, and also obtaining regenerative plants from many domestic elite cotton varieties. After transgenic cotton carrying the insect-resistant (Bacillus thuringiensis: B.t.) gene was commercialized in 1996, at least ten Bt-cotton varieties were planted in China. In 1998, there were over 100,000 hectares of Bt-cotton were planted. Two kinds of bivalent insect-resistant transgenic cotton have been obtained. These new bivalent insect-resistant transgenic cotton carried two insecticidal genes, B.t. gene and CpTI gene, or pea lectin (P-Lec) gene and soybean Kunitz trypsin inhibitor (SKTI) gene respectively, and will be commercialized in 2000. Herbicide-resistant varieties for 2,4-D and Bromoxynil are under development and are expected to reach the market by 2001 or 2002. Disease-resistant transgenic cotton is under development and testing in lab and fields, and is expected to reach the market by 2000 or 2001. Fiber improvements, stress resistance, and male sterility and fertility for hybrid cotton are the next targets for cotton biotechnology. Several genes for fiber improvement and hybrid cotton are being tested in various laboratories. New genes for insect, herbicide and disease resistance are being sought.

P141-T

Somatic embryony patterns and plant regeneration in cotton.

B-H. Zhang; Cotton Res. Inst., Chinese Acad. of Agr. Sci., Baibi, Anyang, Henan 455112, People's Republic of China

Globular, heart, torpedo and cotyledon stages of development were observed in cotton somatic embryos, and only late torpedo and cotyledon stage somatic embryos could germinate into plantlets, the others could only produce advanentive embryos or adventive roots. Various developmental stage somatic embryos existed in cotton tissue cultue, and the number of the globular and heart-shaped stage was more than that of cotyledon embryos. The germination of cotton somatic embryos could be classified into three kinds and were affected by many factors such as hormons, basis medium et al. Resting phenomenon esists in cotton somatic embryos.

P142-S

The novel protein patterning method and its application for protein-protein interactions.

B-G. Kim, C-S. Lee; Sch. of Chem. Engin., Seoul Natl. Univ., Seoul National University, Kwanak-Gu, Seoul, Seoul 151-742, Republic of Korea

This research describes the new protein patterning method for the formation of highly defined two-dimensional protein arrays onto very hydrophobic thin film coated silicon substrate expressing protein functional activity at its patterned surface.

In protein patterning process, Because very critical problem is non-specific protein binding onto undesirable region, We had tried a spin coated FluoroCarbon thin film, very hydrophobic surface property, which was expected to protect non-specific protein binding in our patterned silicon chip which was made from the lift-off process. From this simple method, we had obtained very high ordered protein patterns (180 µm, 300 µm, 500 µm diameter circle and 100 µm by 200 µm rectangular square) and also discarded nonspecific protein binding and blocking proteins which were generally used for protecting nonspecific binding such as bovine serum albumin(BSA) and skim milk proteins et al.

Patterned micron-scale arrays were shown by fluorescence tagging proteins and had successfully obtained reliable quantitative data with confocal microscopy. And We also could create a versatile patterned protein surfaces for protein-protein interactions such as sandwich immuno-assay and indirect immuno-assay.

From the these results, We have developed miniaturized assay systems showing the rapid, low sample volumes and high sensitivity.

P143-M

Relationship between sexual maturation, serum leptin and sex hormone levels in obese and normal children.

S.C. Hao, Y.C. Yuan; Harbin Med. Univ., China, 199 Dazhi Street,Harbin, Harbin, Heilongjiang 150001, People's Republic of China

In order to study the relationship between sexual maturation, serum leptin and sex hormone levels, the sexual maturation of 300 obese children and 300 normal children not taking any medication or having evidence of endocrine or metabolic disease were questioned and their serum leptin and sex hormone levels were measured. Serum leptin levels were significantly increased (P < 0.01) in obese children compared with controls, while the sexual maturation of obese children was much earlier than that of controls (P < 0.01). Serum leptin levels were significantly increased (P < 0.01) in female compared with male, while the sexual maturation of female was much earlier than that of male (P < 0.01). There was significant negative correlation between testosterone and leptin in male (obese group: r = -0.83, P < 0.01; control group: r = 0.93, P < 0.01), while the estradiol was positive correlated with leptin in female (obese group: r = 0.84, P < 0.01; control group: r = 0.95, P < 0.01). This study shows that leptin can promote sexual development in children and may be regulated by sex hormone. There are gender differences in the correlations between leptin and sex hormone, which may cause the differences of sexual maturation in male and female. This study also suggests that leptin play a more important role than sex hormones in the sexual development.

P144-T

Experiences in implementing single nucleotide extension in a high throughput genotyping core facility.

R. Scholl, L.W. Ballard; Univ. of Utah, 50 North Medical Dr. 4A430A, School of Medicine, Salt Lake City, UT 84132

This poster will discuss some of our problems and successes in implementing single nucleotide extension (SNE) in a high throughput genotyping core facility as a method for DNA fragment analysis of SNP's. We will discuss primer selection, PCR protocols, cleanup methods, quality control issues, and loading methods used. Product contamination and multiplexing of the samples will also be addressed.

P145-S

Effects of irradiation on the structure of proteins.

K. Balamurugan, D-H. Chin; Natl. Changhua Univ. of Educ., Post Doctor, c/o Prof. Der-Hang Chin, Department of Chemistry, National Changhua University of Education, Changhua 50058, Taiwan

In this study, we investigate the effects of irradiation on protein samples in optical spectrometers. The protein samples exposed to light in a spectrofluorometer are found to undergo in situ photoreactions. Continuous irradiation of proteins resulted in progressive decrease in the emission intensity. The light-induced changes are irreversible and appear to be oxygen-dependent. Continuous exposure to light renders the protein susceptible to cleavage. The secondary structural contents in proteins show appreciable decrease upon irradiation. In addition, the temperature-induced unfolding of proteins appear to be significantly influenced by the photoreactions caused due to irradiation.

P146-M

Improving productivity in core facilities using the BioRobot 3000.

F. Siegman, H. Kijenski, C. Schade, A. Wehren; QIAGEN Inc., 28159 Avenue Stanford, Valencia, CA 91355

Core facilities providing molecular biology services often require automated systems to perform routine tasks. To meet the various demands faced by a core facility, these workstations should provide the flexibility to automate a wide variety of applications as well as the ability to exchange data with other laboratory instruments and data management systems. The BioRobot 3000 is a series of custom-designed workstations for automating routine tasks such as nucleic acid purification, protein purification and assay, and other liquid-handling tasks including reaction setup and sample rearray. Each workstation is tailor-made to meet individual application requirements. Different BioRobot 3000 configurations designed for different applications will be described. A high-throughput sample purification system with 4 integrated vacuum manifolds provides rapid processing of 384 samples in a single run. Up to 6 µg DNA (plasmids, cosmids, BACs, PACs, or P1 constructs) are purified per well in about 1 hour and 15 minutes using optimized protocols which minimize operator interaction and runtimes. Areas outside the BioRobot 3000 worktable can be accessed using configurations with an extended arm. This arm allows the robotic handling system to move labware to external instruments, such as spectrophotometers and thermal cyclers. The operating system uses standard data formats to facilitate data exchange with these instruments as well as data management systems, allowing samples to be tracked as they move through various processing and analysis steps.

RESEARCH GROUP ABSTRACTS

R1-M

Fragment Analysis Research Group Study 2001: enhancing automated allele calling by tailing dinucleotide repeat markers to eliminate nontemplate driven nucleotide addition seen with Taq polymerase.

D.A. Bintzler¹, P.S. Adams², L.W. Ballard³, Y. Bao⁴, D. Bartley⁵, L. Kasch⁶, L.O. Petukhova⁷, C. Rosato⁸, C.E. Terrell¹; ¹Univ. of Cincinnati, 231 Bethesda Avenue, Cincinnati, OH 45267-0524, ²Trudeau Inst., 100 Algonquin Avenue, Saranac Lake, NY 12983, ³Univ. of Utah, ⁴Univ. of Virginia, ⁵Johns Hopkins Univ., ⁶John Hopkins Univ., ⁷Rockefeller Univ., ⁸Oregon State Univ.

Fragment analysis is a growing field in the genetic sciences. Some basics in methodology may provide useful tools to the many service facilities expanding to include fragment analysis. This tutorial will be an instruction based discussion covering basic methodology and challenges associated with fragment analysis using slab gel and capillary electrophoresis. Topics will include methods for producing clean PCR products through PCR optimization for single markers and multiplexing markers before and after PCR. Instrument related factors that may affect the quality of the data would be discussed for both slab gel and capillary electrophoresis platforms. Finally, a section covering analysis of the results will discuss improving accuracy through selecting correct size standard peaks and developing the best standard curve. Methods of calculation will be compared. Means to properly identify the correct fragment peak from the multiple peaks commonly seen during the fragment analysis run will be demonstrated. Participants in the tutorial will come away with methods designed to improve their capabilities to start, or continue Fragment Analysis in their laboratory.

R2-M

PSRG 2001 Study: synthesis of difficult sequences.

N.P. Ambulos, Jr.¹, L. Bonewald², S. Kates³, G. Osapay⁴, H. Remmer⁵, A. Somogyi⁶, G. Tsaprailis⁶, S. Vigil-Cruz⁷; ¹Univ. of Maryland Sch. of Med., 655 W. Baltimore St. BRB13-009, Baltimore, MD 21201, ²Univ. of Texas Hlth. Sci. Ctr., San Antonio, ³Consensus Pharmaceut., Inc., ⁴Univ. of California, Irvine, ⁵Univ. of Illinois, ⁶Univ. of Arizona, ⁷Univ. of Connecticut

Particular peptide sequences are inherently a challenge to solid-phase synthetic methodologies as a result of concomitant side-reactions and/or conformational issues that may result in the failure to generate the desired product. Development of improved synthetic methodologies, reagents and amino acid derivatives are offering new solutions to the old problem of difficult sequences. This year's ABRF Peptide Synthesis Research Group designed a study to evaluate how participating laboratories solve the challenge of correctly synthesizing these troublesome peptides. The laboratories were asked to prepare the following peptide that has several potential synthetic obstacles.

H-Gln-Thr-Ser-Ser-Gly-Thr-Thr-Ser-Trp-Val-Thr-Ser-Arg-Arg-Asp-Gly-Ala-Gly-Ala-Gly-Pro-OH

The peptide was synthesized and both crude and purified material was analyzed and characterized by the members of the Peptide Synthesis Research Group by HPLC, AAA, MALDI-MS and ESI-MS. Both crude and purified samples were requested from participating laboratories and were also characterized by HPLC, AAA, MALDI-MS and ESI-MS. Analysis of the samples also included a comparison of yield of correct product during purification. The results of these analyses will be presented.

R3-M

A current profile of microarray laboratories: results of the 2000-2001 ABRF Microarray Research Group Survey of Laboratories using microarray technologies.

G.S. Grills¹, C. Griffin², A. Massimi¹, K. Lilley³, K. Knudtson⁴, J. VanEe⁵; ¹Albert Einstein Col. of Med., 1300 Morris Park Avenue, Bronx, NY 10461-1602, ²UCSF, ³Cambridge Univ., ⁴Univ. of Iowa, ⁵Cornell Univ.

The goal of this survey study was to build a current picture of the microarray analysis world. Microarray analysis is a fast developing field. New methods and instrumentation are being constantly introduced and the number of investigators using the technology is rapidly expanding. We focused on Affymetrix GeneChip and on cDNA, oligonucleotide and protein microarray spotting technologies. Data was requested from laboratories using these technologies by posting instructions for participation on microarray related electronic discussion groups. A web based survey form was used to collect information such as instrumentation, protocols, staffing, funding, and throughput. The survey was geared toward gathering information from academic, pharmaceutical, and commercial laboratories that offer microarray technologies as a shared resource. Individual laboratories that have these technologies were also invited to participate. This study is part of a continuing survey. The data from this survey was analyzed to build a current profile of microarray analysis laboratories.

R4-T

NARG 2000-2001 DNA Synthesis Study: Part I. Surveys of trends and strategies of DNA synthesis core facilities. Part II. Evaluation of oligonucleotide synthesis in core facilities utilizing homopolymers of A(20), G(20), C(20), T(20) and one heteropolymer.

M.E. Gunthorpe¹, J.F. Fox², B.P. Holloway³, K.S. Lilley⁴, M. Lively⁵, K. Mills⁶, S.A. Scaringe⁷, T. Thannhauser⁸, A. Yeung⁹; ¹Howard Hughes Med. Inst., UCSF, 533 Parnassus Ave. at 3rd Ave. Rm U436, San Francisco, CA 94143-0793, ²Univ. of Virginia, ³Ctrs. for Dis. Control and Prevent., ⁴Univ. of Cambridge, ⁵Wake Forest Univ. Sch. of Med., ⁶Millennium Pharmaceut. Inc., ⁷Dharmacon Res. Inc., 3200 Valmont Road #5, Boulder, CO 80301, ⁸Cornell Univ., ⁹Fox Chase Cancer Ctr.

The Nucleic Acids Research Group (NARG) of the Association of Biomolecular Resource Facilities (ABRF) invited subscribers to the ABRF electronic discussion group to participate in our 2000-2001 study, which was accessed online at the web address listed below. Part I was a general survey regarding various aspects and issues pertaining to DNA synthesis core facilities, such as quality control, work philosophy and the future. Part II was an evaluation of oligonucleotide synthesis from core facilities by using a novel approach: Isolation of synthesis problems by analysis of homopolymer oligonucleotides. In this approach, the synthesis problems are magnified 19X-20X depending on the chemical properties of the individual bases. Oligonulceotides submitted by participating labs will be analyzed by a range of analytical methods including HPLC, capillary electrophoresis, MALDI-TOF and ESI mass spectrometry to assess the quality of each product. The results of the evaluation of oligonucleotides from participants utilizing different instrumentation and protocols will be presented.

Web address: http://www.abrf.org/ and choose "Nucleic Acids Research Group (NARG) 2000-2001 Study" under the category "Open Research Group Studies".

R5-S

Preliminary results from the DNA Sequencing Research Group 2001 Study: factors that affect the sequencing and detection of mixed base sequences in PCR products.

M.A. Robertson¹, L. Hall², J. Hawes³, T. Hunter⁴, E. Jackson-Machelski⁵, K. Knudtson⁶, D. Leviten⁷; ¹Univ. of Utah, 4A 438, School of Medicine, 50 N. Medical Drive, Salt Lake City, UT 84132, ²Albert Einstein Col. of Med., ³Indiana Univ. Sch. of Med., ⁴Univ. of Vermont, ⁵Washington Univ. Sch. of Med., ⁶Univ. of Iowa, ⁷ICOS Corp.

With the imminent release of the draft Human Genome Sequence, many laboratories are seeing increased demand for comparative sequencing. However, detection of mixed bases in a DNA population is a very demanding technique for many DNA Sequencing Core Laboratories. The DSRG will report on a study to test the ability of the ABRF DNA sequencing community to sequence PCR products containing several different kinds of base pair mutations, insertions and/or deletions. To limit variables, PCR products and sequencing primers were supplied along with a wild type reference sequence as a text file. Members were asked to sequence the PCR products using their choice of chemistry for mutation/mixed base detection. The set of PCR products contained mutations of varying difficulty. The PCR products were amplified from primers that contained universal forward and reverse primers to allow testing of both dye terminator and dye primer sequencing chemistries. Results were submitted as ABI chromatogram files with a short survey to gather relevant information. Analysis of the study will address:

a) The ability of participants to be able to correctly identify the mutations.

b) The accuracy of dye primer versus dye terminator sequencing for correct mutation detection.

c) The ability to detect mutations from only one primer sequence.

d) The differences in instrumentation in the abilty to detect mutations, e.g., capillary vs slab gel instruments.

e) The software/sequence alignment tools being used for mutation analysis.

R6-S

Dissecting structure-function relationships in RNA/protein interaction.

I.A. Laird-Offringa¹, D.G. Myszka², P.S. Katsamba¹; ¹Univ. of Southern California, Norris Cancer Center, Room NOR 6420/1441 Eastlake Ave., Los Angeles, CA 90089-9176, ²Univ. of Utah, 50 N. Medical Dr./School of Medicine Rm 4A417, Salt Lake City, Utah 84132

RNA-binding proteins play critical roles in gene expression and regulation at the post-transcriptional level. While much is known about the various naturally occurring RNA-binding motifs, and co-crystal structures of a number of RNA/protein complexes are available, very little is known about the dynamics of RNA/protein interactions. We have used the spliceosomal protein U1A and its RNA target in the U1 small nuclear RNA (U1hairpinII or U1hpII) as a model to study the kinetics of RNA/protein interaction. Using the previously solved structure of the U1A/U1hpII complex, we have engineered a series of mutants designed to probe the roles of electrostatics, hydrogen bonding, aromatic stacking, and RNA loop length, all of which have been implicated in formation of the U1A/U1hpII complex. The effects of these mutations on the binding dynamics were studied using BIACORE, which yielded high quality kinetic data about the interaction. We determined that neutralization of positive charges on the protein slows the association rate and reduces the deleterious effect of salt on complex formation. In contrast, removal of hydrogen-bonding or stacking interactions within the RNA/protein interface, or reducing the size of the RNA loop, increases the dissociation rate. Our data support a mechanism of binding consisting of a rapid initial association based on electrostatic interactions and a subsequent locking step based on the hydrogen bonding and stacking interactions that occur during the induced fit of RNA and protein. Our results demonstrate the power of kinetics to dissect the functional differences between structural features of two interacting macromolecules.

R7-S

On the development of standards for quantitative molecular interaction analysis: the interaction of barnase with its inhibitor barstar.

E. Eisenstein; NIST, 9600 Gudelsky Drive, Rockville, MD 20850

Virtually all biological processes rely on specific molecular interactions. Consequently, there is a growing need not only for useful tools for the analysis of molecular interactions, but also for standards to evaluate results with new experimental systems, and to assure comparability among different measurement techniques. Recently, the Molecular Interactions Research Group (MIRG) of ABRF focused on the interaction of barnase with barstar as a possible standard for reversible molecular interactions. Preliminary analyses were undertaken to clone, express, purify and characterize two barnase and barstar variants, and to measure the equilibrium constant for their association using three common experimental approaches: analytical ultracentrifugation, isothermal titration calorimetry and surface plasmon resonance using a BIAcore biosensor. Excellent agreement of the results obtained from the three techniques, as well as other experimental attributes of the proteins, suggests that the barnase-barstar system may be a useful standard for quantitative molecular interaction analysis.

SPEAKER ABSTRACTS

S1

Mass spectrometry of neurosteroids.

R.L. Fitzgerald¹, A. Alomary², J.D. Rivera¹, R.H. Purdy³, G.F. Koob⁴, M. Vallee⁴; ¹VA Med. Ctr., UCSD, VAMC-113, 3350 La Jolla Village Dr., San Diego, CA 92161, ²VA Med. Ctr., UCSD, Scripps Res. Inst., ³VA Med. Ctr., Scripps Res. Inst., ⁴Scripps Res. Inst.

Electron capture negative chemical ionization (NCI) is one of the most sensitive ionization techniques available and is especially well suited for quantitative analysis of target compounds in biological extracts. For the analysis of small molecules, NCI is often several orders of magnitude more sensitive than traditional techniques such as electron ionization or positive chemical ionization. There are several important prerequisites for performing quantitative analysis using NCI, including an electronegative functional group, stable isotopic internal standards, and good method validation. We synthesized deuterium labeled analogs of neurosteroids and developed a NCI GC/MS method for quantification of neurosteroids in biological samples using isotope dilution.

Neurosteroids have distinct neurotransmitter mediated effects and consequently it is important to be able to identify and quantify individual compounds. Previously, the determination of neurosteroids in biological matrices involved complicated purification protocols or did not use appropriate internal standards. We added deuterium-labeled internal standards to brain (100 mg of cortex homogenate) or plasma (300 uL). Samples were homogenized in methanol, centrifuged and diluted to contain 5% methanol and then applied to C-18 columns. After washing the column with methanol/water (50/50), steroids were eluted with methanol. Following evaporation, steroids were converted to pentafluorobenzyl oxime/trimethylsilyl ether derivatives. The extracts were analyzed using SIM. The present method allows simultaneous quantification of pg amounts (100 pg in 300 µL of plasma and 250 pg in 100 mg of brain tissue) of neurosteroids and will be helpful in elucidating the role of neurosteroids in health and disease.

S2

High throughput small molecule analysis using LC/MS and SFC/MS.

M.J. Greig, W. Farrell, M. Ventura, J. Robinson, K.E. Milgram, K. Tivel, C. Aurigemma, X. Xiong, C. Pham, R. Lopez, A. Yanovsky; Pfizer--La Jolla, 3550 General Atomics Ct., San Diego, CA 92121

In our high-throughput discovery research, we use both commercial and custom mass spectrometry techniques to analyze tens of thousands of different compounds per month. Atmospheric pressure ionization sources have allowed the routine coupling of HPLC and recently supercritical fluid chromatography (SFC) to mass spectrometers for the analysis of a wide variety of small compounds. Electrospray ionization (ESI) is used not only for large biomolecules, but also for peptides, oligonucleotides, and a wide range of small polar molecules. Atmospheric pressure chemical ionization (APCI) extends the scope of mass spectrometry applications to include many small (<1000 daltons) polar and neutral molecules. The differences in these sources and their applicability to both LC/MS and SFC/MS will be described. The addition of quantitative parallel detectors such as nitrogen chemiluminescence detectors (NCD) and evaporative light scattering detectors (ELSD) will also be discussed. Finally, the benefits of SFC/MS in a high throughput environment will be discussed as well as strategies for high throughput data analysis.

S3

Algorithms and systems for high throughput structural biology.

B.D. Randall¹, T.A. Neubert², H. Erdjument-Bromage³, S.A. Hefta⁴, R.S. Johnson⁵, J.T. Stults⁶; ¹Dartmouth, 6211 Sudikoff Lab., Rm. 113, Dartmouth, Hanover, NH 03755-3510, ²New York Univ., 540 First Avenue, Lab 5-18, New York, NY 10016, ³Mem. Sloan-Kettering Cancer Ctr., 1275 York Avenue, New York, NY 10021, ⁴Bristol-Meyers Squibb, ⁵Immunex Corp., ⁶Genentech, Inc.

In the post-genomic era, key problems in molecular biology center on the determination and exploitation of three-dimensional protein structure and function. For example, modern drug design techniques use protein structure to understand how a drug can bind to an enzyme and inhibit its function. Large-scale structural and functional genomics will require high-throughput experimental techniques, coupled with sophisticated computer algorithms for data analysis and experiment planning. This talk will introduce techniques my lab is developing in two key areas: (1) data-directed computational protocols for high-throughput protein structure determination with nuclear magnetic resonance spectroscopy, and (2) experiment planning and data interpretation algorithms for reducing mass degeneracy in mass spectrometry for protein complex binding mode identification. These techniques promise to lead to fast, automated tools to aid the pursuit of deep structural and functional understanding of biopolymer interactions in systems of significant biochemical and pharmacological interest.

S4

A crack in the egg: protein-protein interactions in failing hearts.

A.R. Marks; Columbia Univ., 630 West 168th Street, New York, NY 10032

Calcium (Ca²⁺) ions are second messengers in numerous signaling pathways in all cell types. In the heart Ca²⁺ regulates muscle contraction, electrical signals that determine the cardiac rhythm, and probably plays a role in controlling cell growth. In the past decade elucidation of the molecular structure of the intracellular Ca²⁺ release channels on the sarcoplasmic and endoplasmic reticulum (SR/ER) has lead to an understanding of how these molecules regulate Ca²⁺ homeostasis in the heart. Consequently the role of these channels (ryanodine receptors, RyR, and inositol 1,4,5-trisphosphate receptors, IP3R) in cardiac pathophysiology is beginning to be understood. Intracellular Ca²⁺ release channels form a unique class of ion channels distinguished on the basis of structure size, and function. RyRs and IP3Rs have large cytoplasmic domains that are involved in the regulation of the channel pore located in the carboxy terminal 10% of the channel sequence. These enormous cytoplasmic domains serve as scaffolds for targeting proteins that bind kinases and phosphatases to the channels. Protein kinase A (PKA) phosphorylation of RyR2 dissociates the channel regulatory protein FKBP12.6 and regulates the channel open probability (Po). We have defined a macromolecular complex comprised of RyR2, FKBP12.6, PKA, the protein phosphatases PP1 and PP2A, and an anchoring protein mAKAP. In failing human hearts RyR2 is PKA hyperphosphorylated resulting in defective channel function due to increased sensitivity to Ca²⁺-induced activation resulting in impaired excitation-contraction (EC) coupling in failing hearts.

S5

Distinguishing agonist and antagonist binding to GPCRs using coupled plasmon-waveguide resonance analysis.

G. Tollin, Z. Salamon, S. Cowell, V. Hruby; Univ. of Arizona, Biosciences West, Tucson, AZ 85721

Structural changes accompanying the binding of ligands to the cloned human delta-opioid receptor immobilized in a solid-supported lipid bilayer have been investigated using coupled plasmon-waveguide resonance spectroscopy. This highly sensitive new technique directly monitors mass density, conformation, and molecular orientation changes occurring within a single bilayer, and also allows direct determination of binding constants without structural modification of materials. Although both agonist binding and antagonist binding to the receptor cause increases in molecular ordering within the proteolipid membrane, only agonist binding induces an increase in thickness and molecular packing density of the membrane. This is a consequence of mass movements perpendicular to the plane of the bilayer occurring within the lipid and receptor components. These results are consistent with models of receptor function that involve changes in the orientation of transmembrane helices.

S6

VLiPs: a technique to study ligand binding to GPCRs in their natural environment.

T.H. Jessen, N. Hunt; EVOTEC BioSystems AG, Hamburg, Schnackenburgallee 114, Hamburg, Hamburg 22525, Germany

Virus Like Particles (VLiPs) are generated by insect cells or mammalian cells through an engineering process of the retroviral export mechanism and the respective G-protein coupled receptor (GPCR). The retroviral gag protein interacts non-covalently but tightly and specifically with the C-terminus of the receptor of interest. The interaction leads to an enrichment of the receptor in its natural environment, the cell membrane, before budding of the VLiPs into the supernatant occurs.

VLiPs carry up to 100 receptor molecules and turn out to be ideally suited for screening due to their robustness, long lived stability and physiological binding properties. Data obtained in binding and competition experiments are identical to data from literature. Currently, functional receptor screens are under investigation. The method allows to explore a number of other protein classes as well.

S7

A knockout for every gene and a chip for every purpose.

M. Sussman; Univ. of Wisconsin, 425 Henry Mall, Madison, WI 53706

We will report on progress in the development of two genomic technologies-saturation reverse genetics using a collection of insertionally mutagenized 'knockout' Arabidopsis plants, and a maskless array synthesizer (MAS) for producing high density DNA oligonucleotide arrays 'on the fly'. Our reverse genetic approach involves the rapid screening of several hundred thousand insertionally mutagenized Arabidopsis lines, for the isolation of knockout plants for any gene of interest. The bias of T-DNA for insertion at particular regions of the genome will be described and compared to published accounts of transposon bias. We have previously reported on a mathematical treatment required to isolate a knockout in every gene (Krysan et al., 1996), and we extend this analysis by taking into account our observations on the small but significant bias for the distributiion of T-DNA sites. Progress towards the establishment of a computer database for T-DNA insertions, using TAIL PCR to generate flanking sequences, will also be described. As a general tool for genotyping and for discerning the effects of a particular mutation on global genome expression, we have been testing existing and new technologies based on high density oligonucleotide arrays. Recent experiments utilizing a maskless array synthesizer (Singh-Gasson et al., 1999) which is capable of generating a half million different oligonucleotides on a 2 cm squared glass surface, will be described. The MAS uses a digital micromirror device developed by Texas Instruments, to generate virtual masks for use in photolithography and is a powerful tool for bringing combinatorial chemistry to the benchtop in research laboratories. Krysan, P.J., Young, J.C., Tax, F. and Sussman, M.R. 1996 Identification of T-DNA insertions within Arabidopsis genes involved in signal transduction and ion transport. Proc. Natl. Acad. Sci. 93:8145-8150. Singh-Gasson, S., Green, R.D,, Yue, Y., Nelson, C., Blattner, F., Sussman, M.R. and Cerrina, F. 1999. Maskless fabrication of light-directed oligonucleotide microarrays usig a digital micromirror array. Nature Biotechnology 17:974-978.

S8

Assaying single cells to single organelles using mass spectrometry.

J.V. Sweedler; Univ. of Illinois, Urbana, 600 S. Mathews Ave. 63-5, Urbana, IL 61801

Understanding the interactions of relatively simple networks of neurons is hampered by a lack of knowledge of the full complement of neuropeptides involved in most neuronal systems. Using matrix-assisted laser desorption/ionization (MALDI) time-of-flight mass spectrometry, neuropeptides can be identified in single cells and even in individual neuronal processes. Mass spectrometric imaging methods are described that can provide spatial "maps" of the neuropeptides found in simple invertebrate networks, as well as identify new neuropeptides. Using these techniques, multiple novel neuropeptides have been discovered in the common neuronal model Aplysia californica. Approaches for direct de novo sequencing of peptides in single cells are demonstrated. A unique sampling protocol has been developed that allows the peptides in single attoliter to femtoliter volume vesicles to be measured using mass spectrometry. Using the atrial gland of Aplysia as a model, more than ten bioactive peptides are found in individual vesicles indicating the complexity of such hormonal signaling. Methods which combine capillary electrophoresis, fluorescence detection and mass spectrometry on the same sample are described.

S9

Profiling signal transduction networks in mammalian cells.

N.L. Allbritton, C.E. Sims, G. Meredith; Univ. of California, Irvine, Medical Sciences I, Rm D380, Irvine, CA 92697-4560

A central goal of genomics and proteomics is to catalog the biological molecules present in different organisms and cell types under various conditions. A greater challenge for accurate and comprehensive characterization, however, lies in determining the activities and functional relationships of the biological molecules, particularly the enzymes, as they occur within the complex cellular networks that comprise biological systems. To accomplish this task, new technologies must be developed to measure multiple chemical species within intact intracellular networks. We have demonstrated a new method, the laser micropipet system, for the simultaneous measurement of the activation of key regulatory enzymes in small groups of cells, a single cell, or portions of a cell. This assay strategy should be broadly applicable to measurements of a broad range of enzymes, including kinases, phosphatases, proteases, and nucleases.

S10

Chemical sensors for monitoring secretory and metabolic dynamics at single cells: application to pancreatic beta cells.

R.T. Kennedy, W-j. Qian, M.G. Roper, L.S. Kauri, G.D. Dalgren; Univ. of Florida, PO Box 117200, Gainesville, FL 32611-7200

Microscale electrochemical sensors that allow detection of insulin, glucose, and oxygen at single cells or small groups of cells have been developed. The insulin electrodes have been used to detect insulin secretion at the level of single exocytosis events while the glucose and oxygen electrodes have been used to monitor the dynamics of glucose and oxygen consumption. Such measurements, in combination with pharmacological probes and gene knock-outs, have been used to characterize secretory pathways and the interaction of metabolism with secretion. In one study, it was demonstrated that glyoclytic and respiratory oscillations occur in single islets of Langerhans and these oscillations require Ca²⁺ entry into the cell for proper feedback. Such oscillations provide a mechanism for oscillatory insulin secretion seen in vivo. In another study, using the single cell approach, we have shown that activation of insulin receptors leads to insulin secretion suggesting that positive feedback contributes to the mechanism for the first phase of insulin secretion. The signaling pathway by which insulin stimulates insulin secretion has been further studied revealing critical roles for IRS-1 and PI3-K in mediating insulin-stimulated insulin secretion. These studies have demonstrated that temporally resolved measurements at single cells or cell clusters are useful in evaluating mechanisms of signal transduction. The temporal measurements, especially of multiple key analytes, allow sequences of events to be evaluated and new mechansims to be uncovered.

S11

Genetic analysis by mass spectrometry.

L.M. Smith; Univ. of Wisconsin-Madison, 1101 University Ave., Madison, WI 53706-1396

In the last decade two powerful new tools for the mass spectrometric analysis of biomolecules have been developed, Matrix-Assisted Laser Desorption Mass Spectrometry (MALDI-MS), and Electrospray Ionization Mass Spectrometry (ESI-MS). The power of these methods lies in their ability to produce and mass analyze intact gas phase ions from very large molecules such as proteins and nucleic acids. The speed, accuracy, and sensitivity of the technologies make them well-suited to address a number of problems in genetic analysis, including the analysis of DNA sequence, genetic variations, and gene expression. Results in these areas will be presented, including recent work in which single nucleotide polymorphisms (SNPs) in genomic DNA may be analyzed without need for a prior PCR amplification step.

TUTORIAL ABSTRACTS

T1

Oligonucleotide synthesis chemistry.

R.T. Pon; Univ. of Calgary, 3350 Hospital Drive NW, Calgary, AB T2N 4N1, Canada

Solid-phase oligonucleotide synthesis remains one of the most important technology platforms in the life sciences. It has been twenty years since the first automated DNA synthesizers were commercialized and phosphoramidite synthesis chemistry is well established. However, demand for increasing numbers and quantities of oligonucleotides continues to drive improvements to this technology. This presentation will provide an overview of DNA and RNA synthesis chemistry as it relates to a core facility operation. Alternative reagents for various synthetic steps will also be reviewed. This will include the Q-Linker and Linker Phosphoramidites reagents developed at the University of Calgary. These reagents have been used to improve synthesis productivity, recycle solid-phase supports, and perform tandem oligonucleotide synthesis. Such reagents may find future application in large scale or high throughput synthesis facilities.

T2

Oligonucleotide synthesis quality control.

A.T. Yeung, C.G. Miller, R.R. Muhlhauser; Fox Chase Cancer Ctr., 7701 Burholme Ave., Philadelphia, PA 19111

Solid-phase oligonucleotide synthesis using automated DNA synthesizers and phosphoramidite synthesis chemistry produces high quality oligonucleotides most of the time. However, a variety of situations can compromise the quality of the DNA products. These situations include synthesizer malfunctions, reagent and column problems, incomplete reagent delivery and inadequate wash steps. The purpose of quality control is to prevent products of inferior quality from becoming variables in the users' experiments. This presentation will discuss the wide range of oligonucleotide quality control methods in use in various laboratories. The semi-automated procedure in our facility, using Mono Q column anion-exchange on a FPLC unit, will be shown. A modification of the DNA synthesis procedure on the ABI model 394 synthesizer to improve product quality will be discussed.

T3

Practical aspects of DNA synthesizer operation.

T.J. Demmitt; Biolytic Lab Performance, Inc., 39120 Argonaut Way, #229, Fremont, CA 94538

Solid-phase oligonucleotide synthesis is typically performed via automated instrumentation. The quality of the product produced is related to many things one of which is the ability of the synthesis instrument to perform the chemistry consistently with quality. Maintenance, troubleshooting, repair and validation is key to keeping any instrumentation operating at peek efficiency. Maintaining these workhorse instruments has typically been entrusted to highly trained factory service engineers. The high level of expertise required to perform maintenance is defined at the instrument design stage, however it is entirely possible to perform many maintenance tasks with minimal electro-mechanical expertise. The focuses of this presentation is to show that a typical lab can perform much of its own service and thus minimize the need for high level maintenance engineering services.

T4

Tutorial/practical aspects of DNA synthesizer operation: a look at high throughput DNA synthesizers.

M.E. Gunthorpe; Howard Hughes Med. Inst., UCSF, 533 Parnassus Ave. at 3rd Ave., Rm U436, San Francisco, CA 94143-0793

Tough competition from commercial companies require DNA synthesis core facilities to continuously provide excellent products and services. One possible solution to ideally meet the challenge is to purchase high-throughput instrumentation. In doing so, the facility would easily increase their capacity without additional personnel and would be able to reduce the base cost so that prices can be lowered to match those of the commercial market. The current choices for high-throughput DNA synthesizers provide a variety of features. A review of these instruments would help to delineate which choice would adequately match the facility's unique needs. The features, advantages/disadvantages and viewpoints regarding these instruments will be discussed in an open format.

T5

Technology resource grant opportunities.

M.T. Marron¹, M.A. Tingle¹, N. Pearson², M.J. Saunders³; ¹NCRR, NIH, 6705 Rockledge Drive, Bethesda, MD 20892-7965, ²CSR, NIH, ³Natl. Sci. Fndn.

Federal funding is critical for equipping university-based Biomolecular Resource Facilities with state-of-the-art technologies. In this session, representatives from two Federal agencies, NSF and NIH, with day-to-day responsibility for administration of instrumentation grants, will discuss the review and funding of their various competitive grant programs.

The tutorial session will be split into two parts. The first half of the session will provide an overview of the various instrumentation programs at NSF and NIH of interest to ABRF members. The second half of the session will focus on the preview processes at both agencies. Discussion will focus on practical tips on how to prepare a successful NSF and NIH proposal. Formal presentations will be followed by a question and answer period.

T6

Advances in DNA sequencing.

M.A. Robertson; Univ. of Utah, 4A 438 School of Medicine, 50 N. Medical Drive, Salt Lake City, UT 84132

The Human Genome Project has been a major driving force for the development of new instrumentation and methodology for high throughput sequencing. During the last five years we have seen major advances both in instrumentation and sequencing chemistries. This tutorial will present user insight into one of the newer capillary instruments that promises to fit well into a core laboratory setting. The tutorial will also look at the changes in sequencing chemistry and present some solutions for difficult templates using these new chemistries. This tutorial is designed to be interactive. There will be a 30 minute question/answer session after the two main speakers, with a panel comprising the speakers and members of the DNA Sequencing Research Group. Please come armed with your most difficult sequencing problems and help spark a lively discussion session.

T7

Advances in DNA sequencing.

L.S. Hall¹, E. Thomas¹, K. Lilley², G. Grills¹; ¹Albert Einstein Col. of Med., 1695 Poplar Street, NewYork, NY 10464, ²Cambridge Univ.

This tutorial will review past advances in DNA sequencing and present new techniques for sequencing difficult templates. Significant advances have taken place in DNA sequencing in the past four years by virtue of improved instrumentation, advances in the sequencing chemistries and alternative reaction formats for sequencing difficult clones. In particular, the replacement of Rhodamine terminators with Big Dye terminators increased the range of signal strengths compatible with good data and eliminated many of the peak height errors associated with rhodamine terminators; such as base calling ambiguities linked to the pattern of a small 'G' following an 'A'.

The recent introduction of dGTP terminators has improved the quality of data from templates with high GC content or regions of secondary structure. Data will be presented demonstrating the effectiveness of dGTP compared to Big Dye. In addition, the efficacy of betaine (NNN Trimethylglycine) versus DMSO will be discussed. New data will be presented demonstrating that a combination of Big dye and dGTP can sequence through regions of high G-C content without incurring compressions, a disadvantage of dGTP terminators by themselves.

Finally, studies are in progress to evaluate the ability of different techniques to sequence through various types of difficult motifs by addition of commercial additives, such as the new GIBCO Rx Enhancer buffers. In general difficult motifs, such as repeats, induce stalling of Taq and loss of signal beyond the offending region. Depending upon the type of template, these buffers can restore the processivity of the enzyme and resolve regions of secondary structure. Results will be presented from studies that identify combinations of additives that work well with particular motifs.

T8

Incorporating a 16 capillary electrophoresis DNA sequencer into a core facility.

P.T. Morrison; Dana-Farber Cancer Inst., 44 Binney St., JFB216, Boston, MA 02115

In the past few years the Human Genome Project (HGP) has driven innovation in DNA sequencing chemistry and instrumentation. For the most part this has been beneficial for all types of DNA sequencing projects both large and small. Recently manufacturers are now filling a void for instrumentation that is better suited for facilities that sequence less than 100,000 templates per year.

The incorporation of an Applied Biosystems Model 3100 Genetic Analyzer into a core facility environment will be discussed.

T9

More than one way to capture a phosphopeptide.

H. Erdjument-Bromage; Mem. Sloan-Kettering Cancer Ctr., 1275 York Avenue, New York, NY 10021

Despite the advances made during the past few years, mapping the phosphorylation sites on a protein often proves to be quite challenging. In this tutorial, several issues concerning sample preparation will be discussed, namely the choice of proteases and chromatographic media (both reversed-phase and metal affinity). To illustrate the importance of sample preparation, "failed" mapping experiments will be used to highlight possible steps where the capturing may go wrong. The optimal approach to affinity capture is a combination of (1) one or more different proteases with careful planning of the "pre-affinity" buffer exchange steps, and (2) specific, micro-preparative capture of the phosphorylated fragments on immobilized metal ions (e.g. Ga³⁺), followed by a desalting/concentration step. Critical examination of all sequential steps will be considered in turn.

T10

Immunocapture of proteins--hints and tricks.

R.M. Goetz; Mem. Sloan-Kettering Cancer Ctr., 1275 York Avenue, Box 137, New York, NY 10021

Immunoprecipitation is a classical technique for capturing proteins from cell and tissue samples. Its sensitivity and specificity may be greater than those offered by any other technique, but critically depend on the experimental conditions applied. In this tutorial, various aspects of protein immunocapture will be discussed including the choice of sample buffer (detergents, chelators, enzyme inhibitors, reducing agents), the choice of antibody (monoclonal versus polyclonal; bivalent versus monovalent) and the choice of capture of formed immune complexes.

T11

Affinity capture of DNA-binding proteins.

M. Yaneva; Mem. Sloan-Kettering Cancer Ctr., 1275 York Ave. Box 137, New York, NY 10021

Regulatory nuclear proteins and their complexes bind to cis-regulatory elements depending on the gene activities. These proteins can be isolated from cell nuclei and captured in vitro using specific DNA oligonucleotides (or their multimers) immobilized on solid supports (e.g. beads). The successful capture of these proteins depends on: 1) abundance of the protein in the cell; 2) efficiency of protein extraction; 3) stability of the protein in extract; 4) maintenance of post-translational modification during extraction; 5) conditions used for in vitro DNA binding; 6) affinity of the protein for the isolated control element. The conditions for sample preparation and characterization as well as the optimization of the affinity capture on DNA beads will be discussed. Finally, the captured proteins are displayed on SDS PAGE and stained proteins bands are subjected to identification by MS.

T12

Preparing and delivering an effective presentation.

L.A. Steinke¹, A. Smith²; ¹Univ. of Nebraska Med. Ctr., Omaha, NE 68198-4525, ²Stanford Univ. Med Ctr.

As scientists, we are often expected to present our data orally, accompanied by slides or a computer presentation. Many of us receive little training in public speaking, and see a microphone only at national meetings. Too often, important results are obscured by a poorly planned or poorly executed presentation. Sometimes, as speakers, we do not even know how to appropriately use the audio visual equipment that is available to us. Al Smith, Stanford University will present "How to get the message across--do's, dont's, and maybe's". The audiovisual department of the Town and Country will then present a short overview of the equipment found at most meeting places, and demonstrate its proper use. A professional speech coach will then cover "Organizing and Preparing a Scientific Presentation. Topics to be covered in this session will include overview of a presentation, organization and analysis of data, and the importance of a conclusion and a take home message.

T13

Fragment Analysis Research Group Tutorial 2001: a practical discussion of fragment analysis: sample preparation, instrumentation and analysis of the results.

D.A. Bintzler¹, L.O. Petukhova², C. Rosato³; ¹Univ. of Cincinnati, 271 Bethesda Avenue, Cincinnati, OH 45267-0524, ²Rockefeller Univ., ³Oregon State Univ.

Fragment analysis is a growing field in the genetic sciences. Some basics in methodology may provide useful tools to the many service facilities expanding to include fragment analysis. This tutorial will be an instruction based discussion covering basic methodology and challenges associated with fragment analysis using slab gel and capillary electrophoresis. Topics will include methods for producing clean PCR products through PCR optimization for single markers and multiplexing markers before and after PCR. Instrument related factors that may affect the quality of the data would be discussed for both slab gel and capillary electrophoresis platforms. Finally, a section covering analysis of the results will discuss improving accuracy through selecting correct size standard peaks and developing the best standard curve. Methods of calculation will be compared. Means to properly identify the correct fragment peak from the multiple peaks commonly seen during the fragment analysis run will be demonstrated. Participants in the tutorial will come away with methods designed to improve their capabilities to start, or continue Fragment Analysis in their laboratory.

AUTHOR INDEX

Abel, A.P., P134-M

Adams, P.S., P19-S, R1

Aebersold, R., P45-T, P60-T, P66-T

Allbritton, N.L., S9

Alomary, A., S1

Alterman, M.A., P80-M

Alward, L.J., P67-S

Ambulos, Jr., N.P., R2

Amparo, G., P3-T

Amshey, J.W., P124-S, P125-M

Anderson, G.A., P45-T, P47-M

Andon, N., P54-T

Andresson, T., P31-S

Annan, R.S., P94-S

Arnott, D., P42-T

Asano, Y., P25-S

Aurigemma, C., S2

Ayanoglu, G., P122-M

B

Baird, G.S., P95-M

Balamurugan, K., P145-S

Baldridge, K.K., P95-M

Ballard, L.W., P144-T, R1

Bao, Y., P136-S, R1

Barbaro, K.C., P126-T

Barragan, G., P135-T

Bartley, D., R1

Bateman, R., P64-S

Bateman, R.H., P91-S

Batz, S.A., P85-S

Bautista, A., P125-M

Bawge, V., P27-T

Baybayan, P.A., P8-M

Belcinski, R., P112-S

Berger, S.J., P47-M

Berggren, K.N., P130-S

Bimmler, J., P61-S

Bintzler, D.A., P9-T, R1, T13

Blackburn, K., P48-T, P70-S

Bloch, C., P77-M

Bogoev, R.A., P124-S

Bonewald, L., R2

Booth, L.R., P1-S

Boucher, R., P48-T

Brancia, F., P63-T

Breu, H., P27-T

Brito, A., P121-S

Brito, J., P79-S

Brown, C.L., P119-M

Brown, J., P62-M

Brown, R., P122-M

Brune, D.C., P49-S

Burkhart, W., P48-T, P70-S

C

Caffo, N., P119-M

Campbell, D., P26-M

Campbell, J., P89-M

Campion, M., P121-S

Carberry, R., P82-S

Carlson, J.E., P67-S

Carr, S.A., P94-S

Carruthers, R.A., P37-S, P91-S

Cavey, G.S., P67-S

Chan, J., P71-M, P73-S

Chang, C-A., P122-M

Chappell, G., P27-T

Chen, S-M., P3-T

Chin, D., P80-M

Chin, D-H., P145-S

Chulin, A., P99-T

Connelly, R., P26-M

Conrads, T.P., P45-T

Cooper, C.A., P86-M

Cordwell, S.J., P128-M

Corthals, G.L., P41-M

Couch, L., P23-M

Cowell, S., S5

Cummings, J.J., P131-M

Cunha, R.B., P126-T

D

Dabrowiak, J., P135-T

Dains, K., P112-S, P115-S

Daley, T-J., P79-S

Dalgren, G.D., S10

Daniels, S., P57-T

Dansithong, W., P25-S

Das, R., P12-T

David, L.L., P123-T

Davis, R., P70-S

Deciu, C., P59-M

DeGnore, J.P., P89-M

Deldot, T., P112-S

Demmitt, T.J., T3

Dhulipala, R., P118-S, P120-T

Dhume, S.T., P85-S

Diwu, Z., P130-S

Dobrzanski, M.J., P19-S

Dolios, G., P79-S

Dong, P., P122-M

Doyle, M.L., P109-S

Dragulev, B.P., P136-S

Ducret, A., P90-T

Duewel, H., P103-S

Dupont, D.R., P84-T, P97-S

E

Eisenstein, E., R7

Ellerbrock, J., P116-M

Eng, J., P60-T

Ens, W., P66-T

Erdjument-Bromage, H., S3, T9

Esteban, C.D., P102-T

F

Faca, V.M., P88-S

Farrell, W., S2

Fayet-Faber, J., P135-T

Fenyo, D., P104-M

Ferguson, M.D., P21-T

Fernandez, J., P79-S

Figeys, D., P103-S

Fisher, J.A., P117-T

Fitzgerald, R.L., P46-S, S1

Fontes, W., P126-T

Fox, J.F., P136-S, R4

Franke, C., P52-S

Fröhlich, T., P52-S, P61-S

Fuhrer, K., P108-T

Fujimura, T., P129-T

Fuller, C., P116-M

Furtos-Matei, A., P93-T

Furuishi, K., P87-T

G

Gall, A., P21-T

Gallagher, P., P136-S

Gaskell, S.J., P63-T

Gillig, K.J., P108-T

Goetz, R.M., T10

Gonin, M., P108-T

Gopalan, S., P115-S

Gostick, D., P62-M

Graham, K.S., P84-T

Greene, L.J., P88-S

Greig, M.J., S2

Griffin, C., R3

Griffin, P.R., P131-M

Griffin, T.J., P66-T

Grills, G., R3, T7

Gross, L.A., P95-M

Guettler, R., P18-T

Gunning, K.M., P2-M

Gunthorpe, M.E., R4, T4

Gygi, S.P., P66-T

H

Haff, L.A., P36-T, P40-S

Hager, K.M., P11-M

Hahner, S., P55-S

Haley, J.D., P104-M

Hall, L.R., P36-T, P40-S

Hall, L.S., P6-T, R5, T7

Hall, S., P18-T

Hall, S.C., P43-S

Halpern, B., P51-T

Hao, S.C., P143-M

Harris, R., P80-M

Harrison, M.J., P86-M

Haugland, R.P., P130-S

Hawes, J., P6-T, R5

Hawke, D.H., P77-M, P50-M

Hayashi, M., P87-T

Haynes, P.A., P54-T

Hefta, S.A., S3

Hemesath, T., P31-S

Henninger, D., P135-T

Henschen-Edman, A.H., P92-M

Henzel, W.J., P75-T, P78-T

Hillen, W., P102-T

Hobbs, J.B., P14-M

Hockenberry, J., P121-S

Hoffer, A., P20-M

Hoffman, R.C., P95-M

Hogan, R.J., P19-S

Holle, A., P65-M

Holloway, B.P., R4

Hong, M-Y., P133-S

Hong, Y.W., P114-T, P81-T

Horn, M.J., P33-T, P68-M

Hosta, L., P121-S

Hoyes, J.B., P37-S, P64-S, P91-S

Hruby, V., S5

Hsi, K-L., P76-S, P97-S

Huang, G.S., P133-S

Hubbard, S.J., P63-T

Huddleston, M.J., P94-S

Huestis, M.A., P108-T

Hufnagel, P., P56-M, P65-M

Hughes, C., P31-S, P37-S

Hughes, K.A., P1-S, P21-T, P113-M

Hunt, N., S6

Hunter, C.L., P51-T

Hunter, T., P6-T, R5

Hunziker, P., P80-M

Hurkman, W.J., P127-S

Hurley, J.B., P96-T

I

Iasnopolski, B., P27-T

Imai, B., P29-M, P5-M, P79-S

Indermuhle, P., P73-S, P110-M

Ingendoh, A., P53-M, P55-S

Ivanetich, K.M., P7-S

Ivanov, V.T., P99-T

J

Jackson-Machelski, E., P6-T, R5

Jedrzejewski, P.T., P71-M, P73-S

Jensen, O.N., P91-S

Jessen, T.H., S6

Jett, M., P12-T

Ji, X., P23-M

Jilkine, A., P66-T

Johnson, B., P8-M

Johnson, B.F., P8-M, P117-T, P119-M

Johnson, R.S., S3

Johnston-Dow, L., P27-T

Jones, C., P37-S, P91-S

Jones, K., P112-S

Jones, R., P36-T

Jou, Y-H., P114-T, P81-T

Juergens, M., P38-M

Juhasz, P., P89-M

K

Kaiser, R.J., P113-M, P1-S

Kaput, G., P121-S

Karnick, S., P98-M

Kasch, L., R1

Kates, S., R2

Kato, M., P25-S

Kato, S., P25-S

Katsamba, P.S., P106-S, P137-M, R6

Kauri, L.S., S10

Kellard, E., P69-T

Kellmann, M., P38-M

Kemper, C., P130-S

Kennedy, R.T., S10

Kihlberg, J., P93-T

Kijenski, H., P146-M

Kim, B-G., P142-S

Kirchner, M., P79-S

Kishiyama, A., P42-T, P75-T

Kitchen, D., P16-S

Knudtson, K., P6-T, R3, R5

Koh, J., P18-T

Kojima, M., P83-M

Koller, A., P54-T

Komatsu, S., P132-T

Konishi, H., P132-T

Koob, G.F., S1

Koomen, J., P108-T

Kopaciewicz, W., P69-T

Kosman, C.A., P27-T

Köster, C., P56-M

Kostrzewa, M., P52-S, P61-S

Kräuter, K-O., P52-S, P56-M

Krishnan, S., P35-M, P82-S

Kshirsagar, B., P27-T

Kumar, A., P120-T

Kumar, S., P27-T

Kuo, S.S., P122-M

L

La Rotta, A., P65-M

Lach, F., P5-M

Laird-Offringa, I.A., P106-S, P137-M, R6

Langridge, J., P62-M

Langridge, J.I., P31-S, P37-S, P39-T, P62-M, P64-S, P91-S

Ledman, D.W., P36-T, P40-S

Lee, C-S., P142-S

Lee, H., P66-T

Lee, K.A., P96-T

Lee, S-W., P47-M

Lee, T.D., P34-S

Leicester, S., P37-S

Leinonen, J., P100-S

Leonard, A., P26-M

Lester, P.J., P63-T

Leung, D.K., P33-T

Leviten, D., P6-T, R5

Li, G., P107-M, P21-T

Li, Z., P132-T

Liddell, P.A., P49-S

Liedtke, S., P53-M

Lilley, K.S., R3, R4, T7

Lin, M., P35-M, P82-S

Lippert, J.W., P49-S

Liu, L., P28-S

Lively, M., R4

Liwei, Q., P8-M

Lobo, M., P43-S

Loboda, A., P66-T

Lock, C.M., P44-M

Lopez, J.C., P49-S

Lopez, R., S2

Lotti, R., P57-T

Lubenow, H., P22-S

Luk, D., P18-T

Lund, K.P., P107-M, P21-T

Luo, Y., P98-M

Lynch, M.D., P82-S

M

Machado, L., P126-T

Maeda, M., P87-T

Mahr, K., P102-T

Mahtani, M., P111-T, P112-S, P115-S

Mamone, A., P112-S, P116-M, P118-S, P121-S

Marks, A.R., S4

Marron, M.T., T5

Mason, G., P27-T

Massimi, A., R3

Mathews, W.R., P67-S

Matsumoto, Y., P83-M

Mayo, J.D., P58-S

McArdle, B., P116-M, P118-S, P121-S

McEldoon, W.L., P68-M

Medalle, J., P29-M, P5-M

Mendis, C.A., P12-T

Meredith, G., S9

Meys, M., P82-S

Milgram, K.E., S2

Millar, A., P31-S, P37-S, P37-S, P39-T, P39-T, P64-S, P91-S

Miller, C.G., P15-T, T2

Miller, M.J., P30-T

Miller, T.B., P19-S

Mills, K., R4

Minarik, M., P111-T, P112-S, P115-S

Minkoff, M.S., P36-T

Mitsumoto, T., P25-S

Mizuno, Y., P83-M

Moore, R.E., P34-S

Moreno, T., P78-T

Morita, M., P129-T

Morrison, P.T., T8

Moseley, A., P48-T, P70-S

Moyer, M., P48-T, P70-S

Mozdzanowski, J., P23-M

Muehlbauer, B., P110-M

Muhlhauser, R.R., P15-T, T2

Murayama, K., P129-T

Myszka, D.G., P101-M, P106-S, P137-M, R6

N

Nadler, T., P57-T

Nakayama, Y., P87-T

Närvänen, A., P100-S

Nedelkov, D., P105-T

Neitz, S., P38-M

Nelson, J., P116-M

Nelson, J.S., P50-M

Nelson, R.W., P105-T

Nemirovskiy, O.V., P67-S

Neubert, T.A., S3

Niemi, G.A., P96-T

Noble, R.L., P84-T

Nock, S., P110-M, P71-M, P73-S

Nordhoff, E., P61-S

Nutter, B., P2-M

Nuwaysir, L., P51-T

O

Oefner, P.J., P50-M

Oliver, S.G., P63-T

O'Malley, R., P39-T, P64-S

Osapay, G., R2

Oshiro, G., P59-M

Ostrowski, L., P48-T

P

Packer, N.H., P86-M

Pakkala, M., P100-S

Palaniappan, C., P121-S

Pan, H., P104-M

Panisko, E.A., P45-T

Parker, K.C., P57-T, P82-S

Pasa-tolic, L., P45-T

Patel, A., P51-T

Patton, W.F., P130-S

Paul, R., P27-T

Paul, S., P25-S

Pearson, N., T5

Peltier, J., P89-M

Pennetti, A., P23-M

Pérez-Martínez, G., P102-T

Petukhova, L.O., R1, T13

Pham, C., S2

Pham, V., P75-T

Philip, Y., P64-S

Pillai, S., P57-T

Pingue, R., P8-M

Pirkola, K., P111-T, P112-S, P115-S

Pitt, A.M., P72-T

Pluskal, M.G., P72-T

Pon, R.T., P13-S, P17-M, T1

Pope, R.M., P32-M

Porter, S.G., P26-M

Pretty On Top, K., P130-S

Price, W., P134-M

Pryor, A., P138-T

Purdy, R.H., S1

Purkayastha, B., P57-T

Pusch, W., P52-S

Q

Qian, W-j., S10

Qui, J., P16-S

R

Rakwal, R., P132-T

Randall, B.D., S3

Randesi, M., P29-M, P5-M

Ranish, J., P60-T

Raska, C.S., P32-M

Rauth, H., P61-S

Reagin, M., P116-M

Remmer, H., R2

Resemann, A., P74-M

Ribbe, J., P20-M, P22-S

Ricart, C.A.O., P126-T

Richert, C., P58-S

Rist, B., P66-T

Rivera, J.D., S1

Robertson, M.A., P6-T, R5, T6

Robinson, J., S2

Rodionov, I., P99-T

Rohde, E., P131-M

Roinishvili, L., P28-S

Roper, M.G., S10

Rosato, C., R1, T13

Ross, P., P36-T

Ross, P.L., P40-S

Roy, K., P119-M, P8-M

Rubenstein, D., P32-M

Ruiz-Taylor, L., P110-M

Russell, D.H., P108-T

S

Salamon, Z., S5

Sanchez, C., P12-T

Sandifer, F., P7-S

Sanghvi, Y.S., P17-M

Sangvanich, P., P63-T

Sarracino, D.A., P58-S

Sasagawa, T., P83-M

Saunders, M.J., T5

Scaringe, S.A., P16-S, R4

Schade, C., P139-S, P146-M

Schafer, A., P134-M

Schäfer, F., P20-M

Scharer-Hernandez, N.G., P134-M

Schieltz, D., P59-M

Schlabach, T.L., P76-S

Schneider, A., P53-M, P55-S

Scholl, R., P144-T

Schoppe, M., P27-T

Schultz, A.J., P108-T

Schürenberg, M., P56-M

Schwarz, H., P134-M

Schweiger-Hufnagel, U., P53-M

Settineri, T.A., P51-T, P77-M

Settinery, C.A., P97-S

Shen, D., P112-S, P115-S

Shen, S., P132-T

Shen, W-Z., P98-M

Shen, Y., P47-M

Shephard, T.R., P90-T

Shillinglaw, W., P78-T

Shindo, N., P129-T

Shinozawa, T., P25-S

Shuster, A., P112-S, P115-S

Sidhu, K.S., P63-T

Siegman, F., P139-S, P139-S, P146-M

Sims, C.E., S9

Siu, D., P27-T

Slagel, J., P26-M

Small, K.S., P135-T

Smirnov, I.P., P40-S

Smith, A., T12

Smith, C.M., P138-T

Smith, L.M., S11

Smith, R.D., P45-T, P47-M

Smith, T.M., P26-M

Smith-Beckerman, D., P43-S

Söhnlein, P., P20-M

Somogyi, A., R2

Song, Y., P9-T

Sousa, M.V., P126-T

Souto, L.M., P10-S

Spicer, D.A., P113-M, P1-S

Springer, A.L., P113-M, P1-S

Spurgeon, S., P3-T

Stalgies, Y., P52-S

Standing, K.G., P66-T

Stanick, W., P57-T

Steinberg, T.H., P130-S

Steinert, K., P20-M, P22-S, P22-S

Steinke, L.A., T12

Stenman, U-H., P100-S

Stoerker, J., P58-S

Stoll, R., P22-S

Stolowitz, M.L., P21-T, P107-M

Stults, J.T., S3

Suckau, D., P65-M, P65-M, P74-M

Sullivan, A.G., P63-T

Suri, P.A., P27-T

Susan, C.I., P94-S

Sussman, M., S7

Sweedler, J.V., S8

Swei, A., P2-M

Sword, E., P27-T

Syu, W-J., P24-T

T

Tanaka, C.K., P127-S

Taylor, R., P7-S

Terrell, C.E., R1

Tetzlaff, C.N., P58-S

Thannhauser, T., P6-T, R4

Thelemann, A., P104-M

Thibault, P., P93-T

Thomas, E., T7

Thomas, I., P61-S

Tian, J.N., P117-T

Tingle, M.A., T5

Titgemeyer, F., P102-T

Tivel, K., S2

Tolani, N., P110-M

Tollin, G., S5

Tomaney, A., P51-T

Tomaney, A.B., P122-M

Tsaprailis, G., R2

Tsien, R.Y., P95-M

Tsunoi, Y., P112-S

Tuyet-Doan, A., P18-T

Tyldesley, R., P64-S

U

Ueda, Y., P123-T

Unemoto, T., P87-T

Vallee, M., S1

V

van Soest, R., P53-M

VanEe, J., R3

Veenstra, T., P45-T

Venkat, K., P23-M

Vensel, W.H., P127-S

Ventura, M., S2

Vepsäläinen, J., P100-S

Vestal, M., P89-M

Vigil-Cruz, S., R2

Vinayak, R., P50-M

Voss, K.O., P117-T

W

Waddell, K., P35-M, P82-S

Wagner, P., P71-M, P73-S, P110-M

Wahle, S., P20-M, P22-S

Wang, Y., P119-M

Ware, J.A., P67-S

Washburn, M.P., P59-M

Wasinger, V.C., P41-M

Wegener, S.L., P13-S

Wehren, A., P139-S, P146-M

Wei, D., P122-M

Wei, J., P54-T

Wen, L., P4-S

Wenzel, T., P52-S, P61-S

West, K., P80-M

Wheaton, A., P119-M

Whitney, S.E., P125-M

Wiebe, G.J., P13-S

Wike, C., P119-M

Wiley, J.P., P21-T, P107-M

Wilkins, M.R., P86-M

Winzeler, E., P59-M

Witt, M., P74-M

Witte, K.L., P110-M

Wolber, D., P7-S

Wolkenhauer, O.V., P63-T

Wolters, D., P59-M

Wong, S., P75-T, P78-T

Woodland, D.L., P19-S

Woods, A.S., P108-T

Wu, C., P81-T, P114-T

Wu, D., P81-T, P114-T

Wu, P., P100-S

X

Xiao, W., P50-M

Xiong, X., S2

Y

Yadav, S.P., P98-M

Yan, W., P7-S

Yaneva, M., T11

Yang, F.J., P81-T, P114-T

Yang, G., P132-T

Yang, P-C., P133-S

Yanovsky, A., S2

Yasui, M., P87-T

Yates, III, J.R., P54-T, P59-M

Yeung, A., P15-T, R4

Yeung, A.T., T2

Yoshikawa, M., P132-T

Young, M.K., P34-S

Young, P., P39-T, P62-M

Young, S.G., P43-S

Yu, S., P13-S, P17-M

Yuan, Y.C., P143-M

Yuen, S.W., P76-S, P77-M, P84-T, P97-S

Z

Zacharias, D.A., P95-M

Zahn, M., P121-S

Zappacosta, F., P94-S

Zaugg, F., P110-M

Zhang, B-H., P140-M, P141-T

Zhang, J., P98-M

Zhou, H., P66-T

Zielke, H., P3-T

Zieske, L.R., P76-S, P77-M

Zon, J., P2-M

Zrolka, A.M., P134-M

Zwahlen, P., P73-S

Return to table of contents