Electrophoresis 1995, 16, 1977-1981 2-D Database of rat liver proteins 1977 N. Leigh Anderson' Ricardo Esquer-Blasco' Jean-Paul Hofmann' Lydie Meheus' Jos Raymackers* Sandra Steine? Frank Witzmann4 Norman G. Anderson' 'Large Scale Biology Corporation, Rockville, M D 21nnogenetics NV, Ghent 3Sandoz Pharma Ltd, Drug Safety Assessment, Toxicology, Basel 4Molecular Anatomy Laboratory, Indiana University Purdue University Columbus, Columbus, IN An updated two-dimensional gel database of rat liver proteins useful in gene regulation and drug effect studies We have improved upon the reference two-dimensional (2-D) electrophoretic map of rat liver proteins originally published in 1991 (N. L. Anderson et al., Electrophoresis 1991, 12, 907-930). A total of 53 proteins (102 spots) are now identified, many by microsequencing. In most cases, spots cut from wet, Coo- massie Blue stained.2-D gels were submitted to internal tryptic digestion [2], and individual peptides, separated by high-performance liquid chromatography (HPLC), were sequenced using a Perkin-Elmer 477A sequenator. Additional spots were identified using specific antibodies. Figure 1 shows the current annotated 2-D map of F344 rat liver, analyzed using the Iso-DALT system (20 X 25 cm gels) and BDH 4-8 carrier ampholytes. Both the map itself and the master spot number system remain the same as shown in the original publication. Table 1 lists the important features of each identification shown, including the gel position, pl, and MI for the most abundant or most basic form of each protein. Using this extended base of identified spots, a series of four improved calibration functions has been derived for the pIand SDS-MI axes (the first two of which are shown in Fig. 2A and B). Both forward and reverse functions are derived, so that one can compute the physical properties of a spot with a given gel location, or inversely compute the gel position expected for a protein having given physical properties: (1) (2) (31 (4) - YRATLiVER - fMr-RATLIVER y (MrSEQUENCE-DERIVED) XRATLiVER = fpi-RATLiVER x (P~SEQUENCE-DERIVED) M,GEL-DERiVED = fRATLiVER Y-Mr (YR*TL,VE~ PzGEL-DERIVED = fRATLIVER X-pI (XRATLIVER) A spreadsheet program (in Microsoft Excel) was devel- oped to facilitate flexible computation of pZ's from amino acid sequence data, and the results were entered into a relational database (Microsoft Access). A table of spot positions and sequence-derived PI'S and M,'s was fitted with a large series of analytic equations using Tablecurve (Jandel Scientific), and the four conversion Eqs. (1)-(4), relating computed p l and gel X coordinate, or computed molecular weight and gel Y coordinate, were selected, based on criteria of simplicity, goodness of fit and favorable asymptotic behavior. Table 2 lists the equations and coefficients. Application of Eqs. (3) and (4) to a spot's X and Y coordinates, given in [l], produce improved M, estimates, and allow computation of pZ Correspondence: Dr. Leigh Anderson, Large Scale Biology Corpora- tion, 9620 Medical Center Drive, Rockville, MD 20850-3338 USA (Tel: +301-424-5989; Fax: +301-762-4892; email: leigh@lsbc.com) Keywords: Two-dimensional polyacrylamide gel electrophoresis / Liver I Map / Identification / Calibration directly in pH units, instead of in terms of positions rela- tive to creatine phosphokinase (CPK) charge standards. The inverse Eqs. (1) and (2) were used to compute the gel positions of a series of p l and MI tick marks. These tick marks were plotted with SigmaPlot (Jandel), together with fiducial marks locating several prominent spots, and the resulting graphic was aligned over the syn- thetic gel image (computed by Kepler from the master gel pattern) using Freelance (Lotus Development). Maps were printed as Postscript output from Freelance, either in black and white (as shown here) or in color, where label color indicates subcellular location (available from the first author upon request). We have also used the rat liver 2-D pattern as presented here to calibrate the pat- terns of other samples. Using mixtures of rat liver and mouse liver samples, for example, we made composite 2-D patterns that allow use of the rat pattern to standar- dize both axes of the mouse pattern. This was accompli- shed by deriving transformations relating the rat and mouse X, and separately the rat and mouse Y, axes (Table 2, lower half; Fig. 2C and D) based on a series of spots that coelectrophorese in these closely related spe- cies. These functions were then applied to derive equa- tions relating the mouse liver X and Y to pl and SDS-M, (Eqs. 5 and 6 below). The resulting standardized 2-D pat- tern for B6C3F1 mouse liver is shown in Fig. 3. MIMOUSELIVER - fRATLiVER Y-Mr UMOUSELIVER Y-RATLIVER y - ( YMOusELivER)) (5) (XMousE LIVER)) (6) PIMOUSELIVER = fRATLiVER x-pi UMOUSELIVER X-mTLiVER x A slightly more complex approach can be used to stand- ardize samples that have few or no spots co-electropho- resing with rat liver proteins. In this case, a 2-D gel is prepared with a mixture of the two samples, and four functions (forward and backward, each for X and Y) are derived relating each sample's own master pattern to the composite. The required functions are then applied in a nested fashion to yield the desired result (using rat plasma as an example): MruTPLASMA - fLTLivER Y-MI URATPLASMA+LIVER Y-RATLIVER y - URATPLASMA Y-RATPLASMA+LiVER y (YmTpLAsMA))) (7) 0 VCH Verlagsgesellschaft mbH, 69451 Weinheim, 1995 0173-0835/95/1010-1977 $5.00+.25/0