A Universal Tag for Recombinant Proteins

A Universal Tag for Recombinant Proteins

ARCHIVES OF BIOCHEMISTRY AND BIOPHYSICS Vol. 329, No. 2, May 15, pp. 215–220, 1996 Article No. 0211 A Universal Tag for Recombinant Proteins1 Weipin...

215KB Sizes 5 Downloads 119 Views

ARCHIVES OF BIOCHEMISTRY AND BIOPHYSICS

Vol. 329, No. 2, May 15, pp. 215–220, 1996 Article No. 0211

A Universal Tag for Recombinant Proteins1 Weiping Luo,* T. Chyau Liang,† Jane M. Li,* Jer-Tsong Hsieh,‡ and Sue-Hwa Lin*,2 †Department of Biochemistry, The University of Texas Medical School at Houston, and Departments of *Molecular Pathology and ‡Urology, The University of Texas M. D. Anderson Cancer Center, 1515 Holcombe Boulevard, Houston, Texas 77030

Received November 22, 1995

Incorporation of tags into recombinant proteins can facilitate their identification and purification. In addition, these tags can also be used to monitor the trafficking or localization of the recombinant proteins inside the cells. Several such tags have been developed. However, the lengths of these tags make it cumbersome to incorporate them into the desired proteins. Typically, one must subclone the desired cDNA into a plasmid containing the tag sequence at a suitable restriction site or ligate a synthetic oligonucleotide containing the tag sequence at a suitable restriction site in the cDNA of the desired protein. These manipulations can be avoided, if one uses a short peptide tag that can be incorporated by PCR. We show here that a short peptide tag, RYIRS, can be easily incorporated at the C termini of recombinant proteins by PCR. We also showed that by using a mAb specific for this peptide sequence, the tagged proteins could be easily detected in Western blot analysis, immunofluorescence staining, and immunoprecipitation. The interactions between this tag sequence and the mAb have been well characterized. One can take advantage of this information and control the reactivities between the tagged proteins and the mAb by varying the lengths of the peptide tags. Furthermore, we showed that this tag can be used to monitor whether a recombinant protein is properly translated and terminated because the interactions between this tag sequence and the mAb requires that the tag be at the C-terminus of the protein. q 1996 Academic Press, Inc.

Incorporation of peptide tags into recombinant proteins can facilitate the identification and purification 1 This work was supported by a Grant-In-Aid from the American Heart Association, Texas Affiliate, to T.C.L. and NIH grants to J.T.H. (CA59939), S.H.L. (GM43189), and Core Grant CA16672. 2 To whom correspondence and reprint requests should be addressed. Fax: (713) 794-4672.

of these proteins by antibodies that react specifically with the tagged sequences. In addition, these tags can also be used to monitor the trafficking or localization of the recombinant proteins inside the cells. Several such tags have been developed. The most commonly used tags include HA-tag (1), FLAG tag (2), and Myctag (3). However, the lengths of these tags make them relatively cumbersome to incorporate into the desired proteins. Typically, one must subclone the desired cDNA into a plasmid containing the tag sequence at a suitable restriction site or ligate a synthetic oligonucleotide containing the tag sequence into a suitable restriction site in the cDNA of the desired protein. The binding epitopes of many well-characterized antipeptide antibodies are composed of six or fewer amino acids. If the antibody recognition sequence is short enough, the peptide tags can be incorporated into the cDNA of the desired proteins by PCR. In that case, one can avoid the tedious cloning procedures and the need to find suitable restriction sites. Furthermore, if the tag-antibody interactions are well characterized, one may be able to control the affinity of the tag to the antibody by adjusting the length of the peptide tag. Therefore, a well-characterized short peptide tag is more desirable than the ones in current use. We recently generated a monoclonal antibody (IRSmAb) against the synthetic dodecapeptide NPDSEIARYIRS. The epitope recognized by this antibody was mapped to the C-terminal pentapeptide, RYIRS. Furthermore, we found that the C-terminal free carboxyl group and the side chains of R, I, R, and S constituted the binding determinants.3 We also found that the Cterminal tripeptide, IRS, was sufficient for interaction with the mAb, albeit with a slightly lower affinity. These observations suggest that this peptide sequence may be useful as a universal tag for recombinant proteins. Since this peptide sequence is short, it may be 3 Liang, T. C., Luo, W., Hsieh, J. T., and Lin, S. H. (1996) the accompanying paper.

215

0003-9861/96 $18.00 Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.

/ 6b17$$9404

04-19-96 14:31:24

arca

AP: Archives

216

LUO ET AL.

incorporated into any proteins by PCR. Furthermore, the antibody-binding determinants on this peptide are well characterized, it may be possible to fine tune its affinity with the antibody by adjusting the length of the peptide tag. The requirement of the C-terminalfree carboxyl group for antibody binding also provides a useful indicator for monitoring whether the tagged proteins are properly translated and terminated. To test the utility of RYIRS peptide as a universal tag, we have used PCR to incorporate this sequence at the C-termini of several recombinant proteins. We found that the recombinant proteins tagged with this peptide sequence are easily detected by Western blot analysis, immunofluorescence staining, and immunoprecipitation. EXPERIMENTAL Expression of fusion proteins in Escherichia coli. Several proteins were expressed in E. coli as maltose-binding protein (MBP) fusions. The cDNA coding for the first Ig-domain of C-CAM14 (LD1) was inserted in frame into the vector pMal-cRI (New England Biolabs, Beverly, MA) downstream of the malE gene, which encodes MBP. To construct the plasmid pMal-LD1, which encodes the first Ig domain of C-CAM1 (LD1, amino acids 35 to 139) fused to the C terminus of MBP, the LD1 coding sequence was first obtained by PCR amplification of the full-length C-CAM1 cDNA. Oligo 31, which contains the N-terminal sequence of C-CAM1, and oligo 33, containing the complementary sequence to the C-terminal portion of LD1, were used as primers (Table I). Restriction sites for EcoRI and PstI were included in the 5* primer and 3* primer, respectively. The conditions used for PCR were 40 cycles of 1 min at 947C to denature the template DNA, 1 min at 507C to anneal the primers, and 3 min at 727C to extend the DNA. The 0.3-kb amplified fragment was cloned into the TA cloning vector pCRII (Invitrogen, San Diego, CA) and its sequence was confirmed by sequencing. The insert was then excised from pCRII by digestion with restriction enzymes EcoRI and PstI. The excised fragment was then inserted into the EcoRI/PstI cloning sites of pMal-cRI to generate plasmid pMal-LD1. The plasmid pMal-LD1c, which codes for the first 51 amino acids of C-CAM1 fused to the C-terminus of MBP, was constructed in a similar manner, except that the oligos used for PCR were oligo 31 and oligo 39, which is complementary to the sequence coding for amino acids 79 to 85 of C-CAM1 (Table I). Plasmids of fusion proteins containing various lengths of the tag sequence (pMal-LD1-5aa, pMal-LD1-4aa, pMal-LD1-3aa, and pMalLD1-2aa) were constructed in two steps. In the first step, the DNA sequence coding for the first Ig domain of C-CAM1 with various lengths of the tag attached to its C terminus were created by PCR with the full-length C-CAM1 cDNA as the template. Oligo 31 was used as the 5* primer. Oligos 34, 35, 36, and 37, which contain a sequence complementary to the last 21 nucleotides of the first Igdomain coding sequence of C-CAM1, portions of the IRS-mAb recognition sequence, a termination codon, and a PstI restriction site, were used as the 3* primers for pMal-LD1-5aa, pMal-LD1-4aa, pMal-LD13aa, and pMal-LD1-2aa, respectively (Table I). The PCR products were subcloned into pCRII (Invitrogen) and sequenced to ensure that there were no nucleotide substitution. The inserts were then excised from pCRII by digestion with PstI and EcoRI and subcloned into plasmid pMal-cRI at the PstI/EcoRI sites. 4 Abbreviations used: MBP, maltose-binding protein; C-CAM1, a cell adhesion molecule with apparent molecular weight of 105 kDa; Ig, immunoglobulin; LD1, first immunoglobulin domain of C-CAM1.

/ 6b17$$9404

04-19-96 14:31:24

arca

To produce the fusion proteins, E. coli DH5a was first transformed with the plasmids described in the previous section, and then translation was induced by the addition of 0.3 mM IPTG according to the procedures provided by the manufacturer (New England Biolabs). Expression of a membrane protein in insect cells. The full-length cDNA coding for the human red visual pigment hs7 (4) was excised from a plasmid by digestion with EcoRI and subcloned into the plasmid pBluescript (Strategene, La Jolla, CA). The sequence encoding RYIRS was incorporated into the 3* end of the coding region by PCR with the primers oligo-hs5* and oligo-hs3* (Table I) and the fulllength cDNA as the template. The 0.35-kb PCR product, which encoded the C-terminal half of the hs7 protein, was subcloned into pSK and the nucleotide sequence of the double-stranded DNA was determined. No nucleotide substitutions were found. This 0.35-kb DNA fragment was further digested with SphI and EcoRV and ligated with the N-terminal portion of the hs7 cDNA at the SphI site. The full-length hs7 cDNA with RYIRS at the C terminus was also subcloned into the baculoviral transfer vector pVL-1392 (Invitrogen) at the EcoRI and SmaI sites to generate the plasmid pVL-hs7. Recombinant baculoviruses were generated by cotransfection of pVLhs7 and wild-type baculovirus into insect cells as described by the supplier (PharMingen, San Diego, CA). Expression of a cytosolic protein in insect cells. To express the first Ig domain of C-CAM1 with RYIRS at the C terminus (LD1-5aa) in the sf9 cells, the DNA fragment encoding LD1-5aa was excised from pCRII with EcoRI and PstI as described in the previous section. This DNA fragment was then subcloned into the baculoviral transfer vector pAcSG-His-NT-A (PharMingen) to generate the plasmid pAcSG-LD1-5aa. Recombinant baculoviruses expressing LD1-5aa were generated as described above. Generation of monoclonal antibodies. Monoclonal antibody against the peptide NPDSEIARYIRS was generated as described elsewhere.3 Monoclonal antibody against MBP was generated by injecting a fusion protein containing MBP. Balb/c mice were each immunized at 1-, 15-, and 30-day intervals with 50 mg of fusion protein emulsified with complete Freund’s adjuvant for the first injection and incomplete Freund’s adjuvant for the subsequent injections. Three days after the last immunization, spleen cells were harvested from the mice and fused with myeloma cell line SP2/0. The hybrid cells were cultured in Iscove’s modified Dulbecco’s medium (Gibco/ BRL) with 20% fetal calf serum. Ascites were generated in Balb/c mice and used without further purification. Immunoblot analysis. Aliquots of lysates of E. coli or infected Sf9 cells were boiled in SDS sample buffer and analyzed by SDS–PAGE (5). Western immunoblotting with monoclonal antibodies was performed as previously described (6), except that horseraddish peroxidase-conjugated goat anti-mouse antibody was used as the second antibody and the signal was detected with an Enhanced Chemiluminescence kit (Amersham, Arlington Heights, IL). Immunoprecipitation. To label the proteins with 125I-BoltonHunter reagent (7) (New England Nuclear), aliquots of E. coli lysate were boiled in denaturation solution (1 mM EDTA, 1 mM dithiothreitol, and 1% SDS) for 5 min. The proteins were then allowed to react with 125I-Bolton-Hunter reagent at room temperature for 16 h. Aliquots (20 ml) of the labeled proteins were diluted into 500 ml of immunoprecipitation buffer (150 mM NaCl, 1 mM EDTA, 0.5% deoxycholate, 1% NP-40, and 25 mM Tris–HCl, pH 7.5) and 3 ml of IRS-mAb was added. The resultant mixtures were incubated for 2 h at room temperature before 20 ml of protein G sepharose (Pharmacia, Piscataway, NJ) was added. The mixture was then incubated at 47C overnight with constant mixing. The protein G sepharose beads were collected by centrifugation and washed sequentially with immunoprecipitation buffer, a mixture of immunoprecipitation buffer and saturated sodium chloride (9:1 v/v) and immunoprecipitation buffer containing 0.1% SDS. The washed beads were then boiled in SDS sample buffer and analyzed by SDS–PAGE (5).

AP: Archives

217

UNIVERSAL PEPTIDE TAG TABLE I

Nucleotides Used in Plasmid Construction Oligomer Oligo Oligo Oligo Oligo Oligo Oligo Oligo Oligo Oligo a

31 33 34 (5aa) 35 (4aa) 36 (3aa) 37 (2aa) 39 hs-5* hs-3*

Strand

Sequencesa

sense antisense antisense antisense antisense antisense antisense sense antisense

GGAATTCCAAGTCACCGTAGACGCTGTG GCTGCAGACGAAATTGCACAGACGTTTG CTGCAGCTAGGATCTTATATATCGGGGGTATACACGAAATTGCA (44) CTGCAGCTAGGATCTTATATAGGGGTATACACGAAATTGCA (41) CTGCAGCTAGGATCTTATGGGGTATACACGAAATTGCA (38) CTGCAGCTAGGATCTGGGGTATACACGAAATTGCA (35) CTGCAGCTAGGATCTTATATATCGTGCAA TGGCAAAGCAGCAGAAAGAGTC GCGATATCAGGATCTTATATATCGTGCAGGCGATACCGAGGACAC

The recognition sequence for PstI is in boldface type. The sequence complementary to the IRS-mAb recognition site is underlined.

Immunofluorescent staining. Immunofluorescent staining of insect cells with monoclonal antibodies was performed as previously described (8), except that the secondary antibody was a fluoresceinconjugated goat anti-mouse antibody.

55 kDa was also detected with anti-MBP mAb (Fig. 1B). The identity of this 55-kDa protein is unknown. When these fusion proteins were probed with IRSmAb, as expected, only the 49-kDa MBP-LD1c fusion

RESULTS

IRS-mAb Recognition of the C-Terminal Tag The IRS-mAb was generated against a dodecapeptide, NPDSEIARYIRS, derived from C-CAM1, which is an epithelial cell-adhesion molecule of the Ig superfamily (6). In ELISA using synthetic peptides, we found that the binding epitope for this mAb was located within the Cterminal pentapeptide, RYIRS. Further analysis revealed that the side chains of R, I, R, and S and the Cterminal-free carboxyl group of the peptide constitute the antibody-binding determinants.3 In addition, we found that a peptide containing only IRS tripeptide (instead of the pentapeptide, RYIRS) at its C-terminus was able to react with the mAb, albeit with a lower affinity. The specific and well-defined binding pattern of this mAb and the fact that its affinity can be modulated by the length of the target peptide suggest that it might be useful in monitoring proteins tagged with the peptide sequence. To test this hypothesis, the peptide sequence was incorporated into MBP-fusion proteins, MBP-LD1 and MBPLD1c (Fig. 1A). MBP-LD1 contains the entire first Ig domain of C-CAM1 fused to the C-terminus of MBP; hence, the dodecapeptide NPDSEIARYIRS is embedded in the middle of the C-CAM first domain. MBP-LD1c is a truncated form of MBP-LD1 such that the dodecapeptide NPDSEIARYIRS is exactly at the C terminus of MBPLD1c. In Western immunoblotting using a mAb against MBP, the expression of MBP control, MBP-LD1, and MBP-LD1c was evident and proteins with expected apparent molecular masses of 50, 55, and 49 kDa, respectively, were revealed by this antibody. However, with MBP-LD1c, in addition to the expected 49-kDa protein, another protein with an apparent molecular mass of

/ 6b17$$9404

04-19-96 14:31:24

arca

FIG. 1. Western immunoblots of MBP-fusion proteins containing the tag sequences. (A) Schematic diagram indicating the location of the dodecapeptide in the MBP-fusion protein constructs, MBP-LD1 and MBP-LD1c. These plasmids were transfected into E. coli and protein expression was induced by addition of IPTG. (B and C) The immunoreactivity of MBP-LD1 and MBP-LD1c as detected by antiMBP monoclonal antibody (B) or IRS-mAb (C) in 1:500 dilution. The arrows indicate the corresponding protein positions.

AP: Archives

218

LUO ET AL.

protein was detectable (Fig. 1C). This result shows that IRS-mAb can react with the tag sequence in fusion proteins, only if the tag sequence is located at the C terminus. This Western blot result is consistent with our earlier peptide-ELISA results and demonstrates that the IRS-mAb can be used in Western blot analysis in addition to ELISA. Minimal C-Terminal Length for IRS-mAb Recognition By mutational analysis using synthetic peptides and ELISA, we have previously shown that the C-terminal IRS tripeptide was sufficient for binding with the IRSmAb.3 To determine whether proteins containing shorter tags can also react with IRS-mAb in Western immunoblots, we incorporated tags with various lengths into the C termini of the MBP-LD1 fusion proteins. The structures of these tagged fusion proteins are shown in Fig. 2A. As observed with MBP-LD1c, two proteins immunoreactive with anti-MBP mAb were detected for each construct and only the smaller ones in each construct were reactive with IRS-mAb (Fig. 2B). It is also clear from this experiment that there is a correlation between the lengths of the tags and the reactivities of these proteins with IRS-mAb. This result suggests that one can control the reactivities between the tagged proteins and the IRS-mAb by adjusting the lengths of this tag. This result also suggests that the shortest tag detectable by this antibody is the IRS tripeptide, consistent with the results from the peptideELISA studies.3 Because the ability of an antibody to react with protein antigens in Western blots does not always correlate with its ability to immunoprecipitate the same proteins, we have also tested the ability of IRS-mAb to immunoprecipitate the tagged proteins. As shown in Fig. 3, IRS-mAb was able to immunoprecipitate proteins containing RYIRS at their C-termini. However, the efficiencies of immunoprecipitation decreased when the tag sequence is shortened. These relative efficiencies as a function of the lengths of the tags parallel those from the Western blot analysis. Immunofluorescent Detection of a Membrane Protein Tagged with RYIRS To test the utility of IRS-mAb in immunofluorescent localization of proteins, we used human red visual pigment protein hs7 as a model. RYIRS was incorporated into the full-length hs7 cDNA by PCR. The resultant cDNA was used to generate the recombinant baculovirus hs7-5aa. This tagged protein was then expressed in insect cells. Western blot analysis using IRS-mAb detected several proteins between 50 and 150 kDa (Fig. 4A). These proteins were absent from the wild-type baculovirus-infected cell lysates, suggesting that they

/ 6b17$$9404

04-19-96 14:31:24

arca

FIG. 2. IRS-mAb Western blot reactivity as a function of the length of the tags. (A) Structures of fusion protein constructs containing tags with various lengths. (B and C) Immunoreactivity of MBP-LD1 fusion proteins containing tags with various lengths as detected by anti-MBP monoclonal antibody (B) or IRS-mAb (C). The arrows indicate the corresponding protein positions.

were specific recombinant viral products. The heterogeneity of proteins detected by IRS-mAb is most likely related to the observation by Okano et al. (9) that the red visual pigment protein could form dimeric or multimeric aggregates when analyzed by SDS–PAGE. It is worth noting that the IRS-mAb did not react with any protein from the wild-type virus-infected cells. This result suggests that the IRS-mAb is highly specific. This specificity is expected because it is highly unlikely for an untagged-protein to have the same peptide sequence at the C-terminus by chance. Immunofluorescent staining of the insect cells expressing the hs7-5aa showed that the protein was localized to the plasma membrane of the cells (Fig. 4C). As a control, MBP-mAb did not produce any fluorescent

AP: Archives

UNIVERSAL PEPTIDE TAG

FIG. 3. IRS-mAb immunoprecipitation reactivity as a function of the length of the tags. Immunoreactivity of MBP-LD1 fusion proteins containing tags with various lengths as detected by anti-MBP monoclonal antibody (A) or IRS-mAb (B). The arrows indicate the corresponding protein positions.

staining (Fig. 4B). The strong immunofluorescence staining of hs7-5aa on the plasma membrane suggests that IRS-mAb is suitable for detection and localization of membrane proteins expressed with the tag. Immunofluorescence Detection of a Soluble Protein Tagged with RYIRS To test whether IRS-mAb can be used for detecting a soluble protein expressed in the cytosol, the DNA

FIG. 4. Reactivity of IRS-mAb with tagged human red visual pigment protein expressed in insect cells. (A) Western immunoblot of human red visual pigment protein (hs7) tagged with RYIRS. Insect cells were infected with recombinant baculoviruses containing the sequences coding for the visual red pigment protein with RYIRS at the carboxyl terminus. After 72 h, cell lysates were analyzed by Western blotting using the IRS-mAb. Cell lysate prepared from insect cells infected with wild-type viruses was used as the negative control. Equal amounts of protein (70 mg as determined by the method described by Peterson (10)) were loaded on a 10% SDS– polyacrylamide gel. (B and C) Immunofluorescence staining of insect cells infected with the tagged hs7-recombinant viruses. Sf9 cells grown on coverslips were infected with the tagged hs7-recombinant viruses. After 72 h of incubation, the cells were fixed by formaldehyde treatment followed with Triton X-100. Immunostaining was performed with anti-MBP monoclonal antibody (B) or IRS-mAb (C).

/ 6b17$$9404

04-19-96 14:31:24

arca

219

FIG. 5. Immunofluorescence detection of a soluble tagged protein expressed in insect cells. Insect cells grown in a flask or on coverslips were infected with recombinant baculovirus containing the sequences coding for the first Ig-domain of C-CAM1(LD1) with RYIRS at its carboxyl terminus. Seventy-two hours after infection, an aliquot of the cells grown in the flask was lysed and examined for expression of LD1-5aa by western immunoblotting using the IRSmAb (A). The cells grown on coverslips were fixed by formaldehyde treatment without (B) or with (C) Triton X-100, and immunostaining was performed with the IRS-mAb.

sequence encoding LD1 containing the tag for RYIRS was cloned into the baculoviral vector pAcSG-HisNT-A (Invitrogen, CA) to generate the recombinant baculovirus LD1-5aa. The LD1-5aa protein expressed from this vector contains six histidines at the N terminus and RYIRS at the C terminus. Western blot analysis using IRS-mAb showed that LD1-5aa expressed from the insect cells had an apparent molecular mass of 16 kDa (Fig. 5A), consistent with the predicted size for the protein product. Because the sequence encoding the signal peptide of C-CAM1 was absent, the fusion protein product was expected to remain in the cytosolic compartment of insect cells. Immunofluorescence staining of the insect cells expressing LD1-5aa showed that IRS-mAb could detect the expressed protein only after the cells were permeablized with detergent (Figs. 5B and 5C). These results suggest that RYIRS should also be useful in monitoring the trafficking and localization of recombinant proteins inside the cells. DISCUSSION

Incorporation of peptide tags into recombinant proteins can facilitate their identification, localization, and purification. An ideal tag should be short and hydrophilic so that it can be easily incorporated into any protein, will be exposed on the surface of the protein, and will not interfere with the normal function of the protein. In this report, we show that the RYIRS peptide sequence has most of these and other desired properties. First, the 5-amino acid recognition sequence is relatively short and thus can be easily incorporated into

AP: Archives

220

LUO ET AL.

the desired proteins by PCR. Second, this tag sequence contains charged and hydrophilic residues that make it likely to be exposed on most proteins and thus easily detected. Third, as shown in this study, this sequence can be used to determine the topology of membrane proteins by incoporating this sequence at their C-termini. Fourth, because the carboxyl group is necessary for the IRS-mAb recognition, this tag could be used to determine the correct reading frame of a cDNA clone by incorporating the tag sequence in all three possible reading frames. Only the construct in which the tag sequence is read through in frame will be recognized by the IRS-mAb. Similarly, this tag is useful in monitoring whether a recombinant protein is properly translated and terminated. When expressing a recombinant protein, one sometimes encounters extra protein products other than the expected ones. As shown in Figs. 1–3, extra protein products with larger apparent molecular weights were detected by MBP antibodies but not by IRS-mAb. These extra products could arise from modification of the transfected plasmids by the hosts. As demonstrated in this study, the correct protein products with the tag at the C-termini could be easily identified. In summary, we showed that the tag consists of a short peptide, RYIRS, together with the mAb specific for this sequence can be of general use because the tagged proteins can be easily identified by Western immunoblotting, immunoprecipitation, and immunofluo-

/ 6b17$$9404

04-19-96 14:31:24

arca

rescent staining.5 The main advantage of using such a short tag is its ease of incorporation into any protein by PCR. ACKNOWLEDGMENT We thank Dr. S.-K. Shyu for providing the tagged hs7 cDNA in a baculoviral vector.

REFERENCES 1. Field, J., Nikawa, J.-I., Broek, D., MacDonald, B., Rodgers, L., Wilson, I. A., Lerner, R. A., and Wigler, M. (1988) Mol. Cell. Biol. 8, 2159–2165. 2. Brizzard, B. L., Chubet, R. G., and Vizard, D. L. (1994) Biotechniques 16, 730–735. 3. Evan, G., Lewis, G., Ramsey, G., and Bishop, J. M. (1985) Mol. Cell. Biol. 5, 3610–3616. 4. Nathans, J., Thomas, D., and Hogness, D. S. (1986) Science 232, 193–202. 5. Laemmli, U. K. (1970) Nature 227, 680–685. 6. Lin, S. H., Culic, O., Flanagan, D., and Hixson, D. C. (1991) Biochem. J. 278, 155–161. 7. Bolton, A. E., and Hunter, W. M. (1973) Biochem. J. 133, 529– 539. 8. Cheung, P. H., Luo, W., Qiu, Y., Zhang, X., Earley, K., Millirons, P., and Lin, S. H. (1993) J. Biol. Chem. 268, 24303–24310. 9. Okano, T., Fukada, Y., Artamonov, I. D., and Yoshizawa, T. (1989) Biochemistry 28, 8848–8856. 10. Peterson, G. L. (1977) Anal. Biochem. 83, 346–356. 5 While our supply lasts, we will provide this antibody to researchers interested in using this tag.

AP: Archives