Created: 1st June 2000, last updated: 30th August 2000, © 2000 ABRF
Long Wen
Biology Department, Microchemical Core Facility, San Diego State University, San Diego, CA
Differential display polymerase chain reaction (DD-PCR) is a novel method for identification of differentially expressed genes by comparative display of arbitrarily amplified cDNA subsets. The basic principle of the DD-PCR is to systematically amplify messenger RNAs and then distribute their 3' termini on a denaturing polyacrylamide gel. Although this technology has been successfully applied in a large number of studies, few novel genes of interest have been identified, suggesting that the method needs further improvement. We discovered that primer purity is crucial. We show that by purifying primers using reverse-phase high-performance liquid chromatography, sampling of differentially expressed genes can be greatly enhanced, and relevant genes can be isolated. Using these purified primers in DD-PCR, when compared with the unpurified primers, it should be possible to identify threefold to fourfold differences in expression or differential expression in a fraction of the cell population. (J Biomol Tech 2000;11:87-91)
Key Words: differential display polymerase chain reaction, high-performance liquid chromatography purified primers, oligonucleotide synthesis, differentially expressed genes.
Address correspondence and reprint requests to: Long Wen, Microchemical Core Facility, Biology Department, San Diego State University, 5500 Campanile Drive, San Diego, CA 92182-4614 (email: lwen@sciences.sdsu.edu).
Differential display (DD) has been developed as a tool for the comparison, identification, and isolation of genes expressed as mRNA in various cells under designated conditions.1 This technique has been used successfully in isolating a number of important genes differentially expressed in many fundamental biologic processes such as embryogenesis and organogenesis, cell and tissue differentiation, long-term plasticity of the nervous system, and cellular responses to various stimuli.
The essence of the differential display method is to use, for reverse transcription, an anchored oligo-dT primer that anneals to the beginning of a subpopulation of the poly(A) tails of mRNAs. The anchored oligo-dT primers consist of 11 or 12 Ts plus two additional 3' bases that provide specificity. These are used in conjunction with a decamer oligonucleotide of arbitrarily defined sequence for the subsequent polymerase chain reaction (PCR) amplification. Primers and cycling conditions in DD-PCR protocols were originally designed so that 100 to 150 different primer combinations encompass most of the estimated 15,000 expressed genes.1-3
Although the method worked well in displaying the expected number of mRNAs per primer combination under optimal conditions, many groups to apply the technique become frustrated when the method fails to identify large numbers of novel and relevant genes. We hypothesize that these failures may result from the presence of incomplete primer sequences that compete with full-length primers in the method's underlying PCR reactions. Commercially available, automated, solid-phase synthesizers and chemical reagents can produce high product yields of standard and modified oligonucleotides.4 Despite constant improvements in synthesis instrumentation and chemistry, full-length products are often contaminated with unacceptable amounts of short-length, failure sequences. The purity requirement for an oligonucleotide product depends on its intended use. In the case of primers for sequencing or PCR amplifications, a crude reaction mixture may be used immediately following its synthesis, cleavage, and deprotection. For some applications, such as site-directed mutagenesis or therapeutic experiments, the analysis of crude reaction mixtures and subsequent purification of the full-length oligonucleotide product from the deleterious sequences or reaction byproducts in the synthesis mixture are mandatory.5 However, the quality requirements of the oligonucleotide primers using for DD-PCR applications are unknown. The use of standard nonpurified primers containing failure fragments may result in competition for amplification so that only a small fraction of potential products are sufficiently amplified and displayed.
In this study, we purified the primers by reverse-phase high-performance liquid chromatography (HPLC) to ensure that only full-length primers are obtained, and we determined the effect of primer purification on DD-PCR. We demonstrated that, by purifying primers, sampling of differentially expressed genes can be greatly enhanced and relevant genes can be isolated. As a result, genes specific to a process (eg, cell differentiation caused by stimulation of interleukin-2 [IL-2]) can be readily identified.
Oligonucleotide primers UTR-10 (5'-c/gTTc/gTTTc/ gTTTc/g-3'), UTR-11 (5'-gTAc/gc/gT-c/gc/gATTTA-3'), UTR-12 (5'-CATTGTTATATTA-3')6 and degenerate two-base anchored oligo-dT primers, T12MG, T12MA, T12MT, and T12MC, were synthesized in house on ABI 392 DNA synthesizers by the Microchemical Core Facility of San Diego State University. The primers were purified by reverse-phase HPLC to ensure only the full-length primers were obtained.
The T1 cells, with or without IL-2 stimulation, were provided by Dr. Kathleen L McGuire (Biology Department, San Diego State University). Total cellular RNA was extracted from cells using STAT60 Reagent (Tel-Test, Inc., Friendswood, TX) according to the manufacturer's protocol.
DD-PCR was performed as previously described.1 First-strand cDNAs (20 µL) were synthesized for each RNA sample separately using one of four T12MG (where M is A, T, G, or C). 0.5 µg total cellular RNA, 4 µL 5X RT buffer, 20 µM dNTPs, and 200 U of Superscript II reverse transcriptase (Life Technologies, Gaithersburg, MD) at 42°C for 1 hour. PCR reactions (20 µL) were performed using 1X PCR buffer, 4 µM dNTPs, 0.5 µM 5'-UTR10 primer/3' T12MG anchored primer, 0.15 µL [alpha-32P] dCTP (3000 Ci/mM, NEN Life Science, Boston, MA), 1 µL of reverse transcription product, and 1 U of AmpliTaq DNA polymerase (Perkin-Elmer, Foster City, CA). The PCR cycling profile was 94°C for 5 minutes (94°C for 30 seconds, 42°C for 1 minute, 72°C for 30 seconds) for 40 cycles and then 72°C for 5 minutes. Denatured DDRT-PCR products were loaded onto a 6% denaturing polyacrylamide DNA sequencing gel and run for 3.5 to 4 hours. The gel was blotting onto filter paper, dried under vacuum on a gel dryer at 80°C for 1 hour, and then exposed to film at room temperature for 16 to 24 hours.
Differentially expressed amplicons were prepared as described.1 Briefly, bands were excised from acrylamide gels, placed in 100 µL of dH2O for 10 minutes and then boiled for 15 minutes. DDRT-PCR products were collected by centrifuging for 2 minutes and stored at -20°C. Reamplified cDNAs were purified from the 2% agarose gel using the QIAEX II kit (Qiagen Corp., Chatsworth, CA) and then stored at -20°C for cloning and sequencing analysis.
Reamplified DD-PCR products were ligated into the TOPO cloning site of pCR 4-TOPO (Invitrogen, Carlsbad, CA) using a T-A cloning approach and the recombinant clones were identified by restriction analysis. Plasmid DNA from clones with insert was prepared by miniprep (Qiagen). DNA sequencing was performed on an Applied Biosystem 377 Automatic DNA sequencer (Perkin-Elmer) in the Microchemical Core Facility, College of Science, San Diego State University. Sequences were compared with the National Center of Biotechnology Information nonredundant sequence database using the BLASTX and BLASTN programs. Significant sequence similarity at the nucleotide level was defined at least 96% identity over the entire length of the clone.
A Waters HPLC instrument was used for the analysis of synthetic oligonucleotides by reverse-phase chromatography. This instrument included a 600 Gradient Solvent Delivery System, a U6K Manual Injector or 717 Autosampler, a Temperature Control System, a 486 UV/VIS Absorbance Detector, and the Millennium32 Chromatography Manager. The packing material chosen for the separation was Waters Bondapak HC18HA (125 Å, 37 to 55 µm, 14% carbon load) contained in an 8 X 100-mm Radial-Pak scaling column. Sample loading and the separations were performed at pH 7.2 to enhance product recovery by minimizing product detritylation during sample loading and gradient elution as described.5
Reverse-phase HPLC has been documented to be an effective technique for the purification of synthetic oligonucleotides. Compared with the use of a simple syringe with Oligo-Pak cartridges, purification of DMT-protected oligonucleotides by reverse-phase HPLC offers the ability to purify large amounts of material in a single run and can deliver product of greater overall homogeneity compared with the use of a simple syringe with the Oligo-Pak cartridges. In our facility, we used a Waters Bondapak HC18HA column to separate the DMT-protected full-length product from branched chain synthesis products or DMT-containing short length sequences. The Millennium32 computer software program was used to integrate the peaks and to generate the percentage of the peak area. As indicated in Table 1, elution of these failure species during the gradient program normally occurs immediately before and after the elution of the tritylated full-length products.5 Good nucleoside resolution is obtained using this mode of chromatography. HPLC gradient elution protocols, coupled to proper fraction collection techniques, can be used to ascertain the quality of oligonucleotides obtained following its synthesis, cleavage, deprotection, and purification.
TABLE 1
Reverse-Phase HPLC Analysis of Synthetic Oligonucleotide Before Purification
|
|
||||||
| Area (%) | ||||||
|
|
||||||
| Non-DMT- Containing Failure Sequence |
Full-Length Oligonucleotide Product |
Trityl-on Failure |
||||
|
|
||||||
| T12MG | 13.31 | 84.85 | 1.84 | |||
| UTR10 | 5.33 | 94.54 | 0.14 | |||
| UTR11 | 4.72 | 94.37 | 0.91 | |||
| UTR12 | 2.04 | 95.93 | 2.03 | |||
|
|
||||||
Four mRNA samples were prepared from T1 cells after 20 minutes, 2 hours, 6 hours, or 24 hours of IL-2 stimulation and then were subjected to differential display analysis. T1 cells cultivated without the addition of IL-2 were used as a control. DD-PCR was performed on isolated mRNA using the original procedures of Liang and Pardee1 with purified or nonpurified primers. Figure 1A shows DNA bands generated by DD-PCR using a combination of two purified primers (T12MG + UTR 10) on matched pairs of normal and IL-2-stimulated T1 cell mRNAs. At least 12 fragments of different sizes were present that were upregulated (up 1 to 12) only for the sample from IL-2-stimulated T1 cell lines. Five DD-PCR fragments were identified as being downregulated (down 1 to 5) in IL-2-stimulated T1 cell lines compared with normal T1 cells.
FIGURE 1. Examples of differentially expressed RNAs identified by differential display polymerase chain reaction (DD-PCR). DD-PCR using purified (A) and nonpurified (B) primers T12MG and UTR-10 was performed on cDNA templates corresponding to 20 ng of total RNA from normal T1 cells (lane 1) or T1 cells cultured for 20 minutes, 2 hours, 6 hours, and 24 hours (lanes 2 through 5) in interleukin-2 (IL-2) as described in Materials and Methods. Arrows indicate bands that appeared to be differentially expressed between IL-2-nonstimulated and -stimulated samples. Upward pointing arrows indicate some of the products expressed preferentially in the IL-2-stimulated cells. Downward pointing arrow indicate the products expressed preferentially in the normal cells.
Figure 1B shows the same experiment done with nonpurified primers. The major band patterns generated by using the nonpurified primers were similar to those obtained with the purified primers; however, using the unpurified primers, we generated four cDNA fractions that were previously identified being upregulated or downregulated from the RNA of IL-2-stimulated cell lines. Of these fragments, three were differentially upregulated in IL-stimulated cells, and one was found to be downregulated from the normal cells after IL-2 stimulation (Table 2).
TABLE 2
Retrospective Analysis of Purified and Unpurified Oligomer Usage Among Differentially Expressed Bands from DD-PCR
|
|
||||||
| Fragments Isolated |
Purified Primera |
Unpurified Primera |
DNA Sequence Match Category |
|||
|
|
||||||
| Up 1 | Yes | No | Yes | |||
| Up 2 | Yes | Yes | NT | |||
| Up 3 | Yes | Yes | Yes | |||
| Up 4 | Yes | No | Yes | |||
| Up 5 | Yes | No | Unknown | |||
| Up 6 | Yes | Yes | Yes | |||
| Up 7 | Yes | No | Yes | |||
| Up 8 | Yes | No | NT | |||
| Up 9 | Yes | No | NT | |||
| Up 10 | Yes | No | Unknown | |||
| Up 11 | Yes | No | Yes | |||
| Up 12 | Yes | No | Yes | |||
| Down 1 | Yes | No | Yes | |||
| Down 2 | Yes | No | Yes | |||
| Down 3 | Yes | Yes | NT | |||
| Down 4 | Yes | No | NT | |||
| Down 5 | Yes | No | Yes | |||
|
|
||||||
aThe intensity of the DD-PCR products was identified visually from DD-PCR gels exposed to film. Band identified is denoted yes, and no denotes no DD-PCR products.
The 17 different PCR fragments previously identified as being differentially expressed in normal and IL-2-stimulated T1 cell lines were excised and reamplified with the same primer used in the original DD-PCR reactions. Reamplified DD-PCR products were subcloned into pCR 4-TOPO vector. Because different cDNA species can be contained theoretically within one eluted differentially expressed band,7 at least 10 colonies were selected from each cloned fragment. The corresponding DNA inserts were sequenced and a search for homologous nucleic acid sequences was performed in the databases at NCBI using the software BLAST. As summarized in Table 2, 10 inserts showed homology with the known genes and two inserts showed no homology with the known genes.
DD-PCR should allow the display of all of the estimated 15,000 unique species of mRNA transcribed in the cell at any given time.1 The number of primer combinations necessary to visualize all of the different messages as bands in polyacrylamide gels has been suggested to range from 240 to 312.1,8 The introduction of DD and related techniques has contributed to the recent shift in focus from DNA genetics to expression genetics. The analysis of changes in gene expression in cells that underwent a particular step in differentiation, dedifferentiation, or carcinogenesis is of prime interest in molecular biology.
We sought to increase the number of bands visualized while minimizing the number of reverse transcription reactions. In this study, we documented the importance of the purity of the primers in the DD-PCR method. We showed that by using purified primers in DD-PCR experiments it is possible to detect more differentially expressed genes than using nonpurified primers. By purifying the primers and using the conditions previously described for DD-PCR,1 we observed an average threefold to fourfold increase on the number of bands obtained by DD-PCR, providing a simple approach to decrease the total number of primer combinations required to visualize 15,000 bands. These bands are reproducible. This may result from the A-T rich nature of the 3' untranslated region of many genes, which could be recognized by the primers. As shown in Table 2, on extraction and cloning of 12 single bands, all fragments were bona fide primer/oligo-dT products.
DD-PCR is distinguished from related methods by a low stringency, competitive PCR step that use primer pairs to target the 3' ends of messenger RNAs. Despite constant technologic improvements, coupling efficiency in each synthesis cycle in the automated solid-phase DNA synthesizers remains greater than 98% but below 100%. Depurination and strand cleavage result in "failure sequences" or shorter total lengths that contaminate the desired full-length synthetic product.9 Using nonpurified primers containing many failed short fragments may compete for amplification so that only a small fraction of potential products are sufficiently amplified and displayed. As a consequence, using purified primers in DD-PCR, compared with nonpurified primers, it should be possible to identify threefold to fourfold differences in expression or differential expression bands. A second benefit of increasing the number of bands in DD-PCR experiments is the resulting accumulation of an enormous number of tags representing differentially expressed genes, which could be prioritized for further study, a prerequisite for the study of more complex biological questions.
I would like to thank Dr. Kathleen L McGuire for cell specimens. I would also like to thank all the members of the Microchemical Core Facility at SDSU for their help and patience, in particular Felise Wolven, Katrine Verdun, Victor Seguritan, and I-Wei Feng. I also thank Dr. Anca Segall for helpful discussion.
1. Liang P, Pardee AB. Differential display of eukaryotic messenger RNA by means of the polymerase chain reaction. Science 1992;257:967-971.
2. Bauer D, Muller H, Reich J, Riedel J, Ahrenkiel V, Warthoe P, Strauss M. Identification of differentially expressed mRNA species by an improved display technique (DDRT-PCR). Nucleic Acids Res 1993;21:4272-4280.
3. Liang P, Averboukh L, Pardee AB. Distribution and cloning of eukaryotic mRNAs by means of differential display: refinements and optimization. Nucleic Acids Res 1993;21:3269-3275.
4. Brown T, Brown DJS. Modern machine-aided methods of oligodeoxyribonucleotides. In Eckstein F (ed): Oligonucleotides and Analogues. New York: IRL Press at Oxford University Press, 1991:1-24.
5. Warren WJ, Vella G. Analysis and purification of synthetic oligonucleotides by high-performance liquid chromatography. In Agrawal A (ed): Protocols for Oligonucleotide Conjugates: Synthesis and Analytical Techniques. Totowa, NJ: Humana Press, 1994:223-264.
6. Daniel G, Amanda GF, Merkenschlager M. Rational primer design greatly improves differential display-PCR (DD-PCR). Nucleic Acids Res 1997;25:2239-2240.
7. Liang P, Pardee AB. Recent advances in differential display. Curr Opin Immunol 1995;7:274-280.
8. Sambrook J, Fritsch EF, Maniatis T. In Molecular Cloning: A Laboratory Manual, 2nd ed. Cold Spring Harbor, NY: Cold Spring Harbor Press, 1989.
9. Paivinen AM, Aguiar H, Reiss P, Bonner A. New challenges in automated DNA synthesis. J Anal Purif 1986;Oct:28-37.