Proc. Natl. Acad. Sci. USA
`Vol. 76, No. 9, pp. 4350-4354, September 1979
`Biochemistry
`
`Electrophoretic transfer of proteins from polyacrylamide gels to
`nitrocellulose sheets: Procedure and some applications
`(ribosomal proteins/radioimmunoassay/fluorescent antibody assay/peroxidase-conjugated antibody/autoradiography)
`HARRY ToWBIN*, THEOPHIL STAEHELINt, AND JULIAN GORDON*t
`*Friedrich Miescher-Institut, P. 0. Box 273, CH-4002 Basel, Switzerland; and tPharmaceutical Research Department, Hoffman-La Roche,
`CH-4002 Basel, Switzerland
`Communicated by V. Prelog, June 12, 1979
`
`A method has been devised for the electro-
`ABSTRACT
`phoretic transfer of proteins from polyacrylamide gels to ni-
`trocellulose sheets. The method results in quantitative transfer
`of ribosomal proteins from gels containing urea. For sodium
`dodecyl sulfate gels, the original band pattern was obtained
`with no loss of resolution, but the transfer was not quantitative.
`The method allows detection of proteins by autoradiography
`and is simpler than conventional procedures. The immobilized
`proteins were detectable by immunological procedures. All
`additional binding capacity on the nitrocellulose was blocked
`with excess protein; then a specific antibody was bound and,
`finally, a second antibody directed against the first antibody.
`The second antibody was either radioactively labeled or con-
`jugated to fluorescein or to peroxidase. The specific protein was
`then detected by either autoradiography, under UV light, or by
`the peroxidase reaction product, respectively. In the latter case,
`as little as 100 pg of protein was clearly detectable. It is antici-
`pated that the procedure will be applicable to analysis of a wide
`variety of proteins with specific reactions or ligands.
`
`Polyacrylamide gel electrophoresis has become a standard tool
`in every laboratory in which proteins are analyzed and purified.
`Most frequently, the amount and location of the protein are of
`interest and staining is then sufficient. However, it may also be
`important to correlate an activity of a protein with a particular
`band on the gel. Enzymatic and binding activities can some-
`times be detected in situ by letting substrates or ligands diffuse
`into the gel (1, 2). In immunoelectrophoresis, the antigen is
`allowed to diffuse (3) or electrophoretically move (4) against
`antibody. A precipitate is then formed where the antigen and
`antibody interact. Modifications have been described in which
`the antigen is precipitated by directly soaking the separation
`matrix in antiserum (5, 6). The range of gel electrophoretic
`separation systems is limited by the pore size of the gels and
`diffusion of the antibody. The systems are also dependent on
`concentration and type of antigen or antibody to give a physi-
`cally immobile aggregate.
`Analysis of cloned DNA has been revolutionized (7) by the
`ability to fractionate the DNA electrophoretically in polyac-
`rylamide/agarose gels first and then to obtain a faithful replica
`of the original gel pattern by blotting the DNA onto a sheet of
`nitrocellulose on which it is immobilized. The immobilized
`DNA can then be analyzed by in situ hybridization. The power
`of immobilized two-dimensional arrays has been extended to
`the analysis of proteins by use of antibody-coated plastic sheets
`to pick up the corresponding antigen from colonies on agar
`plates (8). Sharon et al. (9) have used antigen-coated nitrocel-
`lulose sheets to pick up antibodies secreted by hybridoma clones
`growing in agar.
`In this report we describe a procedure for the transfer of
`
`The publication costs of this article were defrayed in part by page
`charge payment. This article must therefore be hereby marked "ad-
`vertisement" in accordance with 18 U. S. C. §1734 solely to indicate
`this fact.
`
`proteins from a polyacrylamide gel to a sheet of nitrocellulose
`in such a way that a faithful replica of the original gel pattern
`is obtained. A wide variety of analytical procedures can be
`applied to the immobilized protein. Thus, the extreme versa-
`tility of nitrocellulose binding assays can be combined with
`high-resolution polyacrylamide gel electrophoresis. The pro-
`cedure brings to the analysis of proteins the power that the
`Southern (7) technique has brought to the analysis of DNA.
`
`MATERIALS AND METHODS
`Immunogens and Immunization Procedures. Escherichia
`coli ribosomal proteins L7 and L12 were extracted (10) from
`50S subunits and purified as described (11) by ion-exchange
`chromatography on carboxymethyl- and DEAE-cellulose.
`Antibodies were raised in a goat by injecting 250 jig of protein
`emulsified with complete Freund's adjuvant intracutaneously
`distributed over several sites. Bacillus pertussis vaccine (1.5 ml
`of Bordet-Gengou vaccine, Schweizerisches Serum- und
`Impfinstitut, Bern, Switzerland) was given subcutaneously with
`every antigen injection. Booster injections of the same formu-
`lation were given on days 38, 79, and 110. The animal was bled
`on day 117.
`Subunits from chicken liver ribosomes (12) were combined
`in equimolar amounts, and 200-iAg aliquots were emulsified
`with 125 Al of complete Freund's adjuvant injected at one in-
`traperitoneal and four subcutaneous sites into BALB/c mice.
`Booster injections of 400 ,ig of ribosomes in saline were given
`intraperitoneally on days 33, 57, 58, and 59. The animals were
`bled on day 71.
`Electrophoretic Blotting Procedures. Proteins were first
`subjected to electrophoresis in the presence of urea either in two
`dimensions (12) or in one-dimensional slab gels corresponding
`to the second dimension of the same two-dimensional system.
`The proteins were then transferred to nitrocellulose sheets as
`follows. The physical assembly used is shown diagrammatically
`in Fig. 1. A sheet of nitrocellulose (0.45 ,im pore size in roll
`form, Millipore) was briefly wetted with water and laid on a
`scouring pad (Scotch-Brite) which was supported by a stiff
`plastic grid (disposable micropipette tray, Medical Laboratory
`Automation, Inc., New York). The gel to be blotted was put on
`the nitrocellulose sheet and care was taken to remove all air
`bubbles. A second pad and plastic grid were added and rubber
`bands were strung around all layers. The gel was thus firmly
`and evenly pressed against the nitrocellulose sheet. The as-
`sembly was put into an electrophoretic destaining chamber with
`the nitrocellulose sheet facing the cathode. The chamber con-
`tained 0.7% acetic acid. A voltage gradient of 6 V/cm was ap-
`plied for 1 hr.
`For polyacrylamide electrophoresis in the presence of sodium
`dodecyl sulfate (13) instead of urea, the procedure was as de-
`t To whom reprint requests should be addressed.
`
`4350
`
`GNE 2004
`Page 1
`
`

`
`0
`
`0
`
`0
`
`1
`
`0
`
`*I*
`
`A
`
`Biochemistry: Towbin et al.
`
`1
`
`2
`
`~~~~+
`
`3 -"A0
`1~~~
`
`r
`
`0
`
`0
`
`0
`
`0
`
`F/
`
`1
`
`I
`
`I
`
`6
`
`Assembly for electrophoretic blotting procedure. 1,
`FIG. 1.
`Electrodes of destainer; 2, elastic bands; 3, disposable pipette-tip tray;
`4, nitrocellulose sheets; 5, polyacrylamide gel; 6, Scotch-Brite pads.
`Assembly parts are shown separated for visualization only.
`scribed above except that the polarity of the electrodes was
`reversed and the electrode buffer was 25 mM Tris-192 mM
`glycine/20% (vol/vol) methanol at pH 8.3.
`Staining for Protein. The blot may be stained with amido
`black (0.1% in 45% methanol/10% acetic acid) and destained
`with 90% methanol/2% acetic acid (see ref. 14).
`Immunological Detection of Proteins on Nitrocellulose.
`The electrophoretic blots (usually not stained with amido black)
`were soaked in 3% bovine serum albumin in saline (0.9%
`NaCl/10 mM Tris-HCl, pH 7.4) for 1 hr at 40'C to saturate
`additional protein binding sites. They were rinsed in saline and
`incubated with antiserum appropriately diluted into 3% bovine
`serum albumin in saline also containing carrier serum with
`concentration and species as indicated in the legends. The sheets
`were washed in saline (about five changes during 30 min, total)
`and incubated with the second (indicator) antibody directed
`against the immunoglobulins of the first antiserum. As indicator
`antibodies we used '25I-labeled sheep anti-mouse IgG. This had
`been purified with affinity chromatography on Sepharose-
`immobilized myeloma proteins and labeled by a modified
`version of the chloramine T method in 0.5 ml with 0.5 mg of
`IgG and 1 mCi of Na'25I (1 Ci = 3.7 X 1010 becquerels) for 60
`sec at room temperature. The specific activity was approxi-
`mately 1.5 yCi/Ag of IgG. 125I-Labeled IgG was diluted to 106
`cpm/ml in saline containing 3% bovine serum albumin and 10%
`goat serum, and 3 ml of this solution was used for a nitrocellulose
`sheet of 100 cm2. Incubation was in the presence of 0.01% NaN3
`for 6 hr at room temperature. The electrophoretic blots were
`washed in saline (five changes during 30 min, total) and thor-
`oughly dried with a hair dryer. The blots were exposed to Kodak
`X-Omat R film for 6 days.
`Fluorescein- and horseradish peroxidase-conjugated rabbit
`anti-goat IgG (Nordic Laboratories, Tilburg, Netherlands) were
`reconstituted before use according to the manufacturer's in-
`structions. Fluorescein-conjugated antibodies were used at 1:50
`
`Proc. Natl. Acad. Sci. USA 76 (1979)
`
`4351
`
`dilution in saline containing 3% bovine serum albumin and 10%
`rabbit serum. After incubation for 30 min at room temperature,
`the blots were washed as above and inspected or photographed
`with a Polaroid camera under long-wave UV light through a
`yellow filter.
`Horseradish peroxidase-conjugated IgG preparations were
`used at 1:2000 dilution in saline containing 3% bovine serum
`albumin and 10% rabbit serum. The blots were incubated for
`2 hr at room temperature and washed as described above. For
`the color reaction (15), the blots were soaked in a solution of 25
`,.g of o-dianisidine per ml/0.01% H202/10 mM Tris-HCI, pH
`7.4. This was prepared freshly from stock solutions of 1% o-
`dianisidine (Fluka) in methanol and 0.30% H202. The reaction
`was terminated after 20-30 min by washing with water. The
`blots were dried between filter paper. Drying considerably
`reduced the background staining. The blots were stored pro-
`tected from light.
`
`RESULTS
`Electrophoretic Transfer of Ribosomal Proteins from
`Polyacrylamide Gels to Nitrocellulose Sheets. Most proteins
`or complexes containing protein adsorb readily to nitrocellulose
`filters (16), whereas salts, many small molecules, and RNA are
`usually not retained. These binding properties are widely used
`for binding assays with nitrocellulose filters. We found that
`proteins were retained on these filters equally well when carried
`towards the filter in an electric field. If the electric field was
`perpendicular to a slab gel containing separated proteins (see
`Fig. 1), we obtained a replica of the protein pattern on the ni-
`trocellulose sheet. This is demonstrated with ribosomal proteins
`from E. coli; a conventionally stained gel (Fig. 2A) and a stained
`electrophoretic blot of an identical gel (Fig. 2B) are shown. All
`ribosomal proteins from chicken liver and E. coli ribosomes
`detectable on two-dimensional gels could be seen on the elec-
`trophoretic blots produced from them. An example of a blot
`from a two-dimensional gel is given in Fig. 3. When the original
`polyacrylamide gel was stained after blotting, no protein could
`be detected. Thus, the blotting procedure removed all protein
`from the gel.
`To establish whether the proteins removed from the gels were
`quantitatively deposited on the nitrocellulose sheet, we sepa-
`rated 3H-labeled proteins from chicken liver 60S ribosomal
`subunit by two-dimensional electrophoresis and compared the
`radioactivity that could be recovered from the blot with that
`recovered directly from the gel (Table 1). Single proteins or
`groups of poorly separated proteins were cut out and radioac-
`tivity was measured after combustion of the samples. The results
`were within the variability inherent to two-dimensional anal-
`yses. Variations could be accounted for by variable transfer of
`proteins into the second dimension gel and the acuity with
`which spots can be cut out.
`At loads exceeding the capacity of nitrocellulose, losses of
`protein occurred. Titration with radioactive ribosomal proteins
`under blotting conditions showed that at concentrations below
`0.15 ,Lg/mm2 all protein was adsorbed. Overloading became
`apparent when a second sheet of nitrocellulose directly un-
`derneath the first one took up protein or when protein became
`visible on the cathodal surface of amido black-stained blots.
`The conservation of resolution together with the high re-
`covery of ribosomal proteins simplifies the procedure for au-
`toradiography. The common procedure involving drying of
`polyacrylamide gels under heat and reduced pressure (19),
`which is tedious and time consuming, may be eliminated. Be-
`cause the proteins become concentrated on a very thin layer,
`autoradiography from 14C- and m5S-labeled proteins should be
`highly efficient even without 2,5-diphenyloxazole impregnation
`
`GNE 2004
`Page 2
`
`

`
`4352
`
`Biochemistry: Towbin et al.
`
`Proc. Nati. Acad. Sci. USA 76 (1979)
`
`A
`
`B
`
`I~~~~~~~
`
`Fic..
`Electrophoretic blotting of ribosomal proteins from one-dimensional gels. Total ribosomal proteins from E. coli were separated on
`an 18% polyacrylamide slab gel containing 8 M urea. (A) A section of the gel was stained with Coomassie blue; (B) another section was electro-
`p)horetically blotted and the blot was stained as described in Materials and Methods. Electrophoresis was from left to right.
`and the blot were apparent. In spite of the apparently incom-
`(19). We have successfully obtained such autoradiograms from
`plete recovery, blots from polyacrylamide gels containing so-
`gels of MS-labeled proteins (not shown). Further, preliminary
`dium dodecyl sulfate may be used for detection of antigen in
`experiments with tritiated proteins have shown that dried blots
`the same way as described below for ribosomal proteins (un-
`may be processed for fluorography by brief soaking in 10%
`published experiments).
`diphenyloxazole in ether (20).
`Detection of Antigen by Antibody Binding on Blots In
`The above experiments were done with ribosomal proteins
`Situ. We found that proteins transferred to nitrocellulose sheets
`separated on polyacrylamide gels containing urea. We have
`remained there without being exchanged over several days.
`electrophoretically blotted proteins from sodium dodecyl sulfate
`Because a blot could be saturated with bovine serum albumin
`by the modified procedure also described in Materials and
`to block the residual binding capacity of the sheet, it can be
`Methods. Again, there was no loss of resolution. However,
`treated as a solid-phase immunoassay. In the following im-
`differences of staining intensities between proteins on the gel
`munological applications, we used indirect techniques
`throughout. Thus, antibody bound by the immobilized antigen
`was detected by a second, labeled antibody directed against the
`first antibody, and in each case excess unbound antibody was
`washed out.
`Table 1.
`
`A
`
`fts
`
`A.
`
`Electrophoretic blotting of ribosomal proteins from
`FIG. 3.
`two-dimensional gels. Proteins (35 ,ug) extracted from the 60S ribo-
`somal subunit of chicken liver (12) were separated by two-dimensional
`gel electrophoresis. (A) Gel stained with Coomassie blue; (B) blot of
`an identical gel. Electrophoresis: 1st dimension, from left to right
`(towards cathode); 2nd dimension, top to bottom.
`
`Efficiency of transfer of ribosomal proteins to
`nitrocellulose sheets
`Protein or
`group of
`Recovery on
`proteins
`blot, %
`analyzed
`3
`123
`4,4A
`104
`111
`5
`107
`6
`86
`7,8
`9
`80
`10
`112
`79
`11
`12, 16
`93
`13
`95
`15, 15A, 18
`125
`115
`17
`19
`139
`118
`21,23
`26
`114
`27
`143
`69
`28, 29
`31
`117
`33
`131
`Ribosomal large-subunit proteins from chicken liver were tritiated
`by reductive methylation (17) and separated by two-dimensional
`electrophoresis (12) in the presence of 35 ,ig of carrier protein. Two
`identical gels were run. One was stained; the other was electropho-
`retically blotted on a nitrocellulose sheet. Spots were identified ac-
`cording to our nomenclature for chicken ribosomes (12), which differs
`only in minor respects from that established for rat ribosomes (18).
`Corresponding spots or groups of spots were cut from the gel and the
`blot. Their radioactivity was determined after conversion to tritiated
`water in a sample oxidizer (Oxymat).
`
`GNE 2004
`Page 3
`
`

`
`Biochemistry: Tow.in et al.
`BProc. Natl. Acad. Sci. USA 76 (1979)
`
`4353
`
`A
`
`Detection of E. coli ribosomal proteins L7 and L12 by (A) horseradish peroxidase- and (B) fluorescein-conjugated antibodies. Total
`FIG. 4.
`ribosomal proteins from E. coli were separated and blotted as in Fig. 2. The anti-L7/L12 serum had a titer of 340 pmol of 70S ribosomes per
`ml of serum as determined by turbidity formation (20). Incubation was for 2 hr at room temperature in goat antiserum diluted 1:10 in saline
`containing 3% bovine serum albumin and 10%o rabbit carrier serum and then with conjugated anti-goat IgG. In each case the lower strip is a control
`with preimmune antiserum. Electrophoresis was from left to right.
`In Fig. 4 the detection of E. coli ribosomal proteins L7 and
`L12 with a goat serum specific for proteins L7 and L12 is
`shown. L7 is identical to L12, except for its N-acetylated
`NH2-terminal amino acid (21). L7 and L12 fully crossreact
`immunologically (22) and are separated on acidic polyacryl-
`amide gels (21). Both peroxidase- (Fig. 4A) and fluorescein-
`conjugated (Fig. 4B) antibodies were able to reveal immuno-
`globulin that was specifically retained by proteins L7 and L12.
`In each case, the lower gel is a control with preimmune serum.
`Peroxidase-conjugated antibodies were far more sensitive than
`fluorescein-conjugated ones. They could therefore be used at
`much higher dilution. This also permitted the detection of very
`small amounts of antigen. With a rabbit serum (23) we could
`detect 100 pg of L7 and L12 with serum and incubation con-
`ditions similar to those of the experiment described in Fig. 4 (not
`shown).
`Because we can use the procedure to detect a specific anti-
`body reacting with a specific protein after electrophoresis in
`polacrylamide, we should also be able to determine which
`proteins have elicited antibodies in a complex mixture of im-
`munogens. In the experiment of Fig. 5, individual sera of five
`
`mice immunized with chicken liver ribosomes were tested. We
`used 125I-labeled sheep anti-mouse immunoglobulins to detect
`the presence of mouse immunoglobulins. In all mice, antibodies
`were preferentially produced against slowly moving proteins,
`presumably of high molecular weight. The procedure can thus
`characterize the antigen population against which specific
`antibodies have been raised in a mixture of immunogens.
`DISCUSSION
`The electrophoretic blotting technique described here produces
`replicas of proteins separated on polyacrylamide gels with high
`fidelity. We obtained quantitative transfer with proteins from
`gels containing urea. This was established here with ribosomal
`proteins. More generally, nitrocellulose membranes have been
`used to retain proteins from dilute solutions for their subsequent
`quantitative determination (16). Still, there remains the possi-
`bility that certain classes of protein do not bind to nitrocellulose.
`In this case absorbent sheets other than nitrocellulose or dif-
`ferent blotting conditions may be helpful.
`We have demonstrated that proteins immobilized on nitro-
`cellulose sheets can be used to detect their respective antibodies.
`
`FIG. 5.
`Detection of immunoglobulin from individual mice directed against ribosomal proteins from chicken liver. Total protein from chicken
`liver ribosomes (12) was electrophoretically separated and blotted as in Fig. 2. Sera were obtained from five individual mice immunized against
`combined 40S and 60S subunits. The antisera were diluted 1:50 in saline containing 3% bovine serum albumin and 10% goat carrier serum. The
`l)lots were incubated in 250 ,l of the diluted antiserum for 6 hr at room temperature. The blots were combined and treated with 125I-labeled
`sheep anti-mouse IgG and autoradiographed. Electrophoresis was from left to right.
`
`GNE 2004
`Page 4
`
`

`
`4354
`
`Biochemistry: Towbin et al.
`
`With radioactively labeled or peroxidase-conjugated antibodies
`the method is sensitive enough to detect small amounts of
`electrophoretically separated antigen, and this simple procedure
`can also be used to show the presence of small amounts of
`antibody in a serum of low titer. Because the antigen is immo-
`bilized on a sheet, the antibody is not required to form a pre-
`cipitate with the antigen. The blotting technique therefore has
`the potential for immunoelectrophoretic analysis of proteins
`by using binding of Fab fragments or binding of antibodies
`against a single determinant, such as monoclonal antibodies
`produced by hybridomas (24). This could not be done by cur-
`rent immunoelectrophoretic techniques. If hybridoma clones
`are obtained from a mouse immunized with impure immu-
`nogen, it will be possible to use the technique to screen for clones
`making antibody directed against a desired antigen. Provided
`the desired antigen has a characteristic mobility in polyacryl-
`amide gel electrophoresis, the appropriate clone can be selected
`without ever having pure antigen.
`The procedure described here also has potential as a tool for
`screening pathological sera containing auto-antibodies-e.g.,
`those against ribosomes (25-27). The precise identification of
`the immunogenic components may be a useful diagnostic tool
`for various pathological conditions.
`A further advantage of immobilization of proteins on nitro-
`cellulose is the ease of processing for autoradiography. Con-
`ventional staining, destaining, and drying of polyacrylamide
`gels takes many hours, and the exact drying conditions are ex-
`tremely critical, especially for 18% gels as used in the second
`dimension for ribosomal proteins (12). When the proteins are
`transferred to a nitrocellulose support, as described here, the
`electrophoretic blotting takes 1 hr, staining and destaining less
`than 10 min, and drying an additional 5 min. This is thus both
`faster and simpler than conventional procedures, and it elimi-
`nates the tedious and hazardous procedure of soaking the gels
`in diphenyloxazole (19).
`The technique has been developed to detect specific antisera
`against ribosomal proteins. However, it is applicable to any
`analytical procedure depending on formation of a protein-
`ligand complex. With the blotting technique, the usual proce-
`dure of forming a complex in solution and retaining it on a
`membrane would have to be reversed: the protein, already
`adsorbed to the membrane, would have to retain the ligand
`from a solution into which the membrane is immersed. Inter-
`actions that can possibly be analyzed in this way include hor-
`mone-receptor, cyclic AMP-receptor, and protein-nucleic acid
`interactions. The ligand may also be a protein. Enzymes sepa-
`rated on polyacrylamide gels could also be conveniently lo-
`calized on blots by in situ assays. A critical requirement for
`these applications is that the protein is not damaged by the
`adsorption process and that binding sites remain accessible to
`ligands and substrates. In this respect, considerations similar
`to those in affinity chromatography and insoluble enzyme
`techniques pertain.
`The method could also be adapted to the procedure of
`Cleveland et al. (28) for the analysis of proteins eluted from
`bands in polyacrylamide gels by one-dimensional fingerprints:
`one could label by iodination in situ' on the nitrocellulose and
`then carry out the proteolytic digestion.
`
`2.
`3.
`
`4.
`5.
`
`6.
`
`7.
`8.
`
`9.
`
`10.
`
`11.
`
`12.
`
`13.
`14.
`
`15.
`
`16.
`
`17.
`18.
`
`Proc. Natl. Acad. Sci. USA 76 (1979)
`In preliminary experiments we have attempted to identify
`ribosomal RNA binding proteins by binding RNA to ribosomal
`proteins immobilized on nitrocellulose by the procedure of this
`paper, followed by staining for RNA (unpublished data), and
`have found a tendency for nonspecific binding. However, J.
`Steinberg, H. Weintraub, and U. K. Laemmli (personal com-
`munication) have independently developed a similar procedure
`for identifying DNA binding proteins.
`We thank Drs. J. Schmidt and F. Dietrich for advice and help with
`immunization procedures and Mrs. M. Towbin for advice on setting
`up the peroxidase assay.
`Gordon, A. H. (1971) in Laboratory Techniques in Biochemistry
`1.
`and Molecular Biology, eds. Work, T. S. & Work, E. (North-
`Holland, Amsterdam), p. 62.
`Williamson, A. R. (1971) Eur. J. Immunol. 1, 390-394.
`Grabar, P. & Williams, C. A. (1955) Biochim. Biophys. Acta 17,
`67-74.
`Laurell, C.-B. (1965) Anal. Biochem. 10, 358-361.
`Zubke, W., Stadler, H., Ehrlich, R., Stoffler, G., Wittmann, H.
`G. & Apirion, D. (1977) Mol. Gen. Genet. 158, 129-139.
`Showe, M. K., Isobe, E. & Onorato, L. (1970) J. Mol. Biol. 107,
`55-69.
`Southern, E. M. (1975) J. Mol. Biol. 98, 503-517.
`Broome, S. & Gilbert, W. (1978) Proc. Natl. Acad. Sci. USA 75,
`2746-2749.
`Sharon, J., Morrison, S. L. & Kabat, E. A. (1979) Proc. Natl. Acad.
`Sci. USA 76, 1420-1424.
`Hamel, E., Koka, M. & Nakamoto, T. (1972) J. Biol. Chem. 247,
`805-814.
`Moller, W., Groene, A., Terhorst, C. & Amons, R. (1972) Eur. J.
`Biochem. 25, 5-12.
`Ramjoue, -H.-P. R. & Gordon, J. (1977) J. Biol. Chem. 252,
`9065-9070.
`Laemmli, U. K. (1970) Nature (London) 227,680-685.
`Schaffner, W. & Weissmann, C. (1973) Anal. Biochem. 56,
`502-514.
`Avrameas, S. & Guilbert, B. (1971 Eur. J. Immunol. 1, 394-
`396.
`Kuno, H. & Kihara, H. K. (1967) Nature (London) 215, 974-
`975.
`Moore, G. & Crichton, R. R. (1974) Biochem. J. 143, 604-612.
`McConkey, E. H., Bielka, H., Gordon, J., Lastick, S. M., Lin, A.,
`Ogata, K., Reboud, J.-P., Traugh, J. A., Traut, R. R., Warner, J.
`R., Welfle, H. & Wool, I. G. (1979) Mol. Gen. Genet. 169,
`1-6.
`Bonner, W. M. & Laskey, R. L. (1974) Eur. J. Biochem. 46,
`83-88.
`Randerath, K. (1970) Anal. Biochem. 34, 188-205.
`Terhorst, C., Wittmann-Liebold, B. & Moller, W. (1972) Eur.
`J. Biochem. 25, 13-19.
`Stoffler, G. & Wittmann, H. G. (1971) J. Mol. Biol. 62, 407-
`409.
`Howard, G., Smith, R. L. & Gordon, J. (1976) J. Mol. Biol. 106,
`623-637.
`Kohler, G. & Milstein, C. (1976) Eur. J. Immunol. 6,511-519.
`Schur, P. H., Moroz, L. A. & Kunkel, H. G. (1967) Immuno-
`chemistry 4, 447-453.
`Miyachi, K. & Tan, E. M. (1979) Arthritis Rheum. 22, 87-93.
`Gerber, M. A., Shapiro, J. M., Smith, H., Jr., Lebewohl, 0. &
`Schaffner, F. (1979) Gastroenterology 76, 139-143.
`Cleveland, D. W., Fischer, S. G., Kirschner, M. W. & Laemmli,
`U. K. (1977) J. Biol. Chem. 252, 1102-1106.
`
`19.
`
`20.
`21.
`
`22.
`
`23.
`
`24.
`25.
`
`26.
`27.
`
`28.
`
`GNE 2004
`Page 5