During the early part of 1990 we had two sequencers to evaluate in our laboratory plus a five year old ABI 475A. The three sequencers were all "on-line" and were the following: APPLIED BIOSYSTEMS 475A, PORTON 2090E, MILLIGEN 6600.
We obtained 6 samples from the ABRF to do a direct comparison on the three sequencers. The samples were ABRF90SEQ which were on PVDF membrane. These experiments were done before the sequence of ABRF-9OSEQ was published. Two samples were run on each of the three machines. The two samples sequenced on the MilliGen were coupled using the SequeNet chemistry from MilliGen. These two samples had to be extensively washed with methanol and water before sequencing to remove residual poly(allylamine)polymer.
No assignments on ABRF-9OSEQ after residue 28 were expected since residue 29 is a cysteine coupled to a carrier protein. The Porton 2090E gave the best results with this sample. The two runs were the 39th and 46th on this instrument. The 5 year old ABI 475A also performed quite well with this sample. The runs were made with the standard "old style" cartridge (ABI 475A indicates a 470A with on-line HPLC and a 900 computer). More recently we have been evaluating the slotted "Blott" cartridge and have found that it improves sequencing performance on PVDF. These runs were numbers 1589 and 1591 on the ABI. The results on the MilliGen sequencer were very poor in comparison to the other instruments. We came to the conclusion that the SequeNet chemistry is of little value for sequencing proteins and peptides blotted to membrane. It is difficult and time-consuming to remove all of the polymer. In our experience most proteins and peptides are not washed off the PVDF membrane. This is especially true for the newer membranes obtained from Bio-Rad and ABI. The runs on the MilliGen were the 38th and 39th made. The results of the six runs are presented in the table below.
The SecY protein (an integral membrane protein from E. coli obtained from W. Wickners' laboratory) was sequenced on all three sequencers as well. Sequencing of the 21K SecY protein blotted to PVDF gave the same type of results as with the ABRF-9OSEQ sample. About 50 pmol of the blotted protein was sequenced for 25 cycles on each of the three instruments. The correct sequence was called to residue 25 in both the AB1475A and the Porton 2090E. Again the blotted protein sequenced poorly in the MilliGen 6600. Only 12 correct calls were made. The alanine in step one was obtained in poor yield and of the three Iysine residues (3,12,19) only number 12 was called tentatively on the Milligen 6600. Arginine residues 20 and 21 as well as glycine residues 13, 14, and 16 were not called. Needless to say the Porton 2090E along with the ABI 475A is now a permanent part of the UCLA Protein Microsequencing Facility.
Data has been included from a second facility (Al Smith, Beckman Center, Stanford University). The ABI 470A with on-line HPLC was 6 years old and was updated with the regulator assembly. The ABI 473A was new and was used in the pulsed-liquid configuration. Polybrene filters were used for these two samples (but not for the 6 run in the UCLA facility). On the sequencer run on the 473 Asp and Asn merged into a single peak during the run. These residues are normally resolved and readily identified. The Stanford calls were only made to residue 20.
Return to the The ABRF Home Page
Created: 21st August 1995
Last modified: 21st October 1995