Proc. Natl. Acad. Sci. USA Vol. 89, pp. 7801-7805, August 1992 Biochemistry

Direct identification of residues of the epidermal growth factor receptor in close proximity to the amino terminus of bound epidermal growth factor (cross-linking/protein sequendng/binding site)

RANDALL L. WOLTJER*t, THOMAS J.

LuKASt, AND JAMES V. STAROS*§¶

Departments of *Biochemistry, tPharmacology, and Molecular Biology, Vanderbilt University, Nashville, TN 37235

Communicated by Stanley Cohen, May 19, 1992

ABSTRACT We have recently developed a kinetically controlled, step-wise affinity cross-linking technique for specific, high-yield, covalent linkage of murine epidermal growth factor (mEGF) via its N terminus to the EGF receptor. EGF receptor from A431 cells was cross-linked to radiolabeled mEGF (MImEGF) by this technique and the 12'I-mEGF-receptor complex was purified and denatured. Tryptic digestion of this preparation gave rise to a unique radiolabeled peptide that did not comigrate with trypsin-treated 125I-mEGF in SDS/Tricine gels but that could be immunoprecipitated with antibodies to mEGF. The immunoprecipitated peptide was isolated by electrophoresis in SDS/Tricine gels, eluted, and sequenced. The sequence was found to correspond to that of a tryptic peptide of the EGF receptor bginning with Gly-85, which is in domain I, a region N terminal to the first cysteine-rich region of the receptor. Selective loss of signal in the 17th sequencing cycle suggests that the point of attachment of N-terminally modified 125I-mEGF to the receptor is Tyr-101 The data presented here provide identification by direct protein microsequencing of a site of interaction of EGF and the EGF receptor.

It is convenient to define the extracellular portion of the human EGF receptor in terms of the following domains (26): a cysteine-poor N-terminal region (domain I, residues 1-146), and a second cysteine-poor region (domain III, residues 333-460) that separates two cysteine-rich regions (domain II, residues 147-332; domain IV, residues 461-621). The bulk of existing evidence, including the results of all cross-linking studies reported to date, points to domain III as containing determinants for receptor binding to EGF (20, 23, 24, 26, 27); however, some studies have also implicated domain I (23, 25). We have recently developed a kinetically controlled, stepwise affinity cross-linking technique to achieve high yields of the EGF receptor covalently linked to the N terminus of murine EGF (mEGF) (30). Here we report purification of cross-linked species and conditions for denaturation and tryptic digestion of 125I-mEGF-linked receptor. A radiolabeled receptor fragment that did not comigrate in SDS/ Tricine gels with products resulting from tryptic digestion of 125I-mEGF was further purified; direct protein microsequencing revealed the uniquely migrating peptide to be derived from domain I of the extracellular portion of the EGF receptor. 11

Epidermal growth factor (EGF), a 6040-Da, single-chain polypeptide hormone (1, 2), binds to specific membrane receptors in target cells to exert effects on cell growth and differentiation (reviewed in ref. 3). Rapid effects of EGF binding to its receptor, a single-chain (170-kDa) transmembrane glycoprotein, include receptor dimerization (4-9) and stimulation of a protein tyrosine kinase (10, 11) intrinsic to the EGF receptor (12-14), which gives rise to receptor autophosphorylation (12, 15) and phosphorylation of intracellular substrates (reviewed in ref. 3). Much attention has been focused on the interaction of EGF with particular receptor sites and the relationship of this interaction to short- and long-term biological responses (reviewed in refs. 3 and 1618). A variety of techniques have been used in past work to investigate the ligand-binding region of the EGF receptor, including preparation of antibodies against the receptor that compete with EGF for binding (19-23), and functional analysis of chicken/human receptor chimera (23, 24) and receptor deletion mutants (25). Covalent cross-linking of 1251-labeled EGF (125I-EGF) to the receptor has been used in previous work, in which identification of the portion of the receptor to which 125I-EGF was linked was deduced from the electrophoretic mobility and immunochemical reactivity of the labeled products of proteolytic and glycosylytic digests (26, 27). Most recently, EGF bound to the soluble, extracytoplasmic portion of the receptor has been visualized in electron microscopic images (28) and crystallized for x-ray diffraction studies (29).

MATERIALS AND METHODS Materials. mEGF was prepared as described (31); 1251mEGF was prepared as described (32), using 1 mg of unlabeled mEGF per ml as a carrier for radiolabeled ligand. The cross-linking reagent sulfo-N-succinimidyl-4-(fluorosulfonyl)benzoate (SSFSB) was synthesized as described (30). A431 cells were grown to confluence in Dulbecco's modified Eagle's medium (GIBCO) supplemented with 10%o calf serum (GIBCO). Trypsin/EDTA (lx) was from GIBCO. Shed membrane vesicles from A431 cells were prepared as described (15). ATP was purchased from Boehringer Mannheim. Triton X-100 was obtained from Aldrich, and glycerol was from Fisher. Anti-phosphotyrosyl antibody (APY) was prepared as described (33, 34) and was used as purified antibody coupled to cyanogen bromide-activated Sepharose 4B (35). Agarose-bound wheat germ lectin (WGL) was obtained from Vector Laboratories, and NN',N"-triacetylchiAbbreviations: APY, Sepharose-coupled anti-phosphotyrosyl antibody; DSS, disuccinimidyl suberate; EGF, epidermal growth factor; mEGF, murine EGF; PAS, protein A-Sepharose CL-4B; PTH, phenylthiohydantoin; SSFSB, sulfo-N-succinimidyl-4-(fluorosulfonyl)benzoate; WGL, agarose-bound wheat germ lectin. tPresent address: Department of Molecular Biology, Vanderbilt University, Nashville, TN 37235. ITo whom reprint requests should be addressed at: Department of Molecular Biology, Vanderbilt University, Box 1820, Station B, Nashville, TN 37235. IA preliminary account of some of this work was presented at the 1990 Annual Meeting ofthe American Society for Biochemistry and Molecular Biology (47).

The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact. 7801

7802

Biochemistry: Woltjer et al.

totriose was from Sigma. lodoacetamide and dithiothreitol were purchased from Fluka; trypsin (L-1-tosylamido-2phenylethyl chloromethyl ketone treated) was obtained from Worthington. Anti-EGF immune antiserum was a generous gift from S. Cohen (Vanderbilt University). Protein A-Sepharose CL-4B (PAS) was from Pharmacia. For gels from which peptides were isolated for sequencing, acrylamide and NN'methylenebisacrylamide from BDH were used; for other gels, these were obtained from Bio-Rad. SDS was purchased from Serva, and ammonium persulfate and N,N,N',N'tetramethylethylenediamine were from Bio-Rad. Mercaptoacetic acid (sodium salt) was obtained from Aldrich. All other chemicals were obtained from Sigma and were reagent grade or better. Affinity Cross-Linking of 125I-mEGF to the EGF Receptor in A431 Cells. For preliminary experiments, 1251-mEGF was covalently cross-linked to the EGF receptor in A431 cell membrane vesicles with SSFSB as described elsewhere (30). For sequencing studies, intact A431 cells were used as a source of EGF receptor as follows: confluent A431 cells were detached from T75 flasks with gentle agitation for 30 min at room temperature with 5 ml of trypsin/EDTA (lx) per flask. Harvested cells were pelleted by centrifugation at maximum speed for 5 min in a Sorvall GLC-1 centrifuge maintained at 40C and supernatants were discarded; the yield was -150 ,ul of packed cells per flask. Cells were washed twice by resuspension in 30 vol of an ice-cold solution of 20 mM Hepes, pH 8.0/10 mM iodoacetic acid/i mM EGTA/0.1 mM phenylmethylsulfonyl fluoride, followed by centrifugation as described above and discarding of supernatants. Washed A431 cells were stored at -70°C or used immediately. 17-5I-mEGF (specific activity, 30,000 cpm/,ug) was derivatized with SSFSB and separated from excess reagent by gel filtration as described (30) and applied at a final concentration of 0.12 ,uM to 4-5 ml of washed A431 cells suspended in 160 ml of ice-cold buffer containing 50 mM Hepes (pH 8.0), 0.1 mM Na3VO4, 5 mM MgCl2, 1 mM MnCl2, and 40 p.M ATP. Cross-linking of 1251-mEGF to the EGF receptor and receptor autophosphorylation were allowed to proceed with gentle agitation for 4 h at 4°C, with addition of 6.4 ,umol of concentrated fresh ATP at 3.5 h. Centrifugation (27,000 x g) for 20 min at 4°C followed, the supernatant was discarded, and pelleted cells were solubilized by alternately pipetting and Vortex mixing them at 4°C in an ice-cold solution of 40 ml of 20 mM Hepes (pH 7.4), 5% Triton X-100, 10% (vol/vol) glycerol, 1 mM EGTA, 0.1 mM Na3VO4, 5 mM MjCl2, 1 mM MnCl2, and 100 pM ATP. Solubilized cells were clarified by centrifugation (213,000 x g) for 20 min at 4°C. Anti-Phosphotyrosyl Affinity Purification of EGF Receptor Covalently Linked to 12'I-mEGF. The supernatant from labeled, solubilized A431 cells was divided into four aliquots, and each aliquot was applied to a hydrated vol of 1 ml of APY resin moistened by APY buffer [20 mM Hepes (pH 7.4), 0.2% Triton X-100, 10%o glycerol, 1 mM EGTA, and 0.1 mM Na3VO4]. Adsorption of receptor to the resin was accomplished with gentle rocking of the slurry for 2 h at 4°C. Centrifugation of the slurry in the same manner as described above for A431 cells was followed by washing the pellet four times by resuspension in APY wash buffer (APY buffer supplemented with 150 mM NaCl, 5 mM MgCl2, 1 mM MnCl2, and 0.1 mM ATP); centrifugation; and discarding of supernatants as described above. Labeled receptor was eluted by rocking the resin overnight at 4°C in 2 ml of APY wash buffer supplemented with 150 mM NaCl, 5 mM phenyl phosphate, and 0.05% sodium azide. The resin was pelleted as described above, and the supernatant was collected as eluate. The resin was resuspended in another 2 ml of elution buffer, centrifugation and supernatant collection were repeated immediately, and eluates were pooled.

Proc. Natl. Acad. Sci. USA 89 (1992)

WGL Affinity Purification of Eluate from the Anti-

Phosphotyrosyl Resin. A total of 16 ml of pooled APY eluate

was applied to 0.5 ml of hydrated WGL resin, concentrated MgCl2 was added to a final concentration of 10 mM, and the slurry was rocked overnight at 4TC. The WGL resin was sedimented by using the same centrifugation techniques as with APY resin described above, and the supernatant was discarded. Resin-bound receptor was washed by resuspension in 4 ml of ice-cold WGL wash buffer containing 20 mM Hepes (pH 7.4), 10%1 glycerol, 0.2 M NaCl, and 10 mM MgCl2 and pelleted as described above; the supernatant was discarded. Elution was accomplished by rocking the resin for 6 h in 0.5 ml of WGL wash buffer without MgCl2 but supplemented with 3 mM NN',N"-triacetylchitotriose, followed by centrifugation as described above for the resin, and collection of the supernatant. The WGL resin was eluted an additional seven times in rapid succession using the same procedure, and eluates were pooled. Tryptic Digestion of Purified EGF Receptor Cross-Linked to '"I-mEGF. The WGL eluate was dialyzed overnight at 4TC in Spectrapor no. 2 membrane dialysis tubing (Mr cutoff, 12,000-14,000) that had been pretreated by boiling for 20 mi in Milli-Q-filtered water (Millipore), against 4 liters of MilliQ-filtered water with one change of water. Aliquots of the dialyzed material were added repeatedly to a 1-ml Reacti-Vial (Pierce) and dried in a Speed-Vac concentrator (Savant). EGF receptor cross-linked to 125I-mEGF, visible as an offwhite residue at the tip of the vial, was dissolved in 100 td of a solution of 8 M urea/0.4 M ammonium bicarbonate and reduced and carboxyamidomethylated as described (36). Digestion with trypsin was carried out with stirring at 3TC in a vol of 0.4 ml, with a total of 4 pug of trypsin added in three aliquots over 24 h. Digestion was halted with the addition of a 12x molar excess of soybean trypsin inhibitor. Immunoprecipitation of EGF-Linked Tryptc Peptides. The tryptic digest was transferred to a 5-ml glass tube, and 0.6 ml of immunoprecipitation buffer (50 mM TrisHCI, pH 8.0/ 0.02% SDS/150 mM NaCl), 0.4 ml of anti-EGF immune antiserum, and 0.2 ml of hydrated, packed PAS resin were added. The resulting slurry was rocked overnight at 4TC. The PAS resin was pelleted for 10 min at 40C in the GLC-1 centrifuge as described above for other resins, and the supernatant was removed; the beads were then washed six times by resuspension in 0.8 ml of immunoprecipitation buffer, followed by centrifugation and removal of supernatants as described above for other resins. To immunoprecipitate additional 12-5I-mEGF-linked peptide, the first four supernatants obtained in the procedure described above were pooled, and 0.4 ml of hydrated, packed PAS resin and 0.3 ml of anti-EGF immune antiserum were added. The resulting slurry was rocked for 3.5 h at 40C before pelleting and washing six times as described above with 3 ml of immunoprecipitation buffer per wash. Electrophoresis of Anti-EGF Immunopreciptates. Electrophoresis in SDS/Tricine gels was performed as described (37) using gel phases with the following composition [the nomenclature is that of Hjerten (38)]: stacking phase, 4% T, 3% C; spacer phase, 10%1 T, 3% C; separating phase, 16.5% T, 6% C. Glycerol (13%) was present in the separating phase. Gels were preelectrophoresed with 7.5 mg of mercaptoacetic acid. To the 0.6 ml of washed, combined PAS beads was added an equal volume of SDS-containing 2x electrophoresis buffer (as in ref. 37, except that 100 mM dithiothreitol was used as the reductant), and this suspension was subjected to electrophoresis at 100 V for 18 h. An overnight autoradiograph was obtained of the wet gel mounted on Whatman 3 mm paper and sealed between plastic sheets. Elution of Immunoprecipitated Peptides from Polyacrylamide Gels. Using the autoradiograph as a guide, the region of the wet gel containing the labeled peptide of interest was

Biochemistry: Woltjer et al. excised, and the sample of wet gel was homogenized in 5 ml of Milli-Q water. The minced acrylamide was incubated with agitation for 4 h at 40C and then pelleted for 10 min at 40C as described above for resins. The supernatant was retained, and peptide elution with 5 ml of water was repeated as described above, but with overnight incubation of the resuspended acrylamide pellet. Pooled eluates were concentrated to 0.1 ml of viscous liquid using a Speed-Vac evaporator. To this liquid was added 0.15 ml of methanol, and the resulting mixture was added in aliquots of 75 A1 to a protein support disc (Porton Instruments, Tarzana, CA) in a 5-ml glass tube. After addition of each aliquot, the disc was dried for 30 min in a Speed-Vac concentrator, washed by agitation for 1 min with 1 ml of methanol followed by removal of methanol by pipetting, and redried as described above. After application of all aliquots, two additional methanol washes were performed as described above, except that the disc was left to soak in methanol for 1 h at room temperature before removal of methanol. Sequendng of EGF-Linked Peptides. Sequencing of Porton disc-supported samples was performed on an Applied Biosystems model 475A sequencer. Phenylthiohydantoin (PTH) amino acid derivatives were separated on a model 120A on-line analyzer by the column and separation protocol provided by the manufacturer. Chromatographic data were collected and analyzed with an Applied Biosystems 900A data controller with the supplied data acquisition software.

RESULTS AND DISCUSSION In a previous paper, we reported the development of the technique used here to achieve specific, covalent attachment of SSFSB-modified 125I-mEGF to >60%o of specific binding sites for EGF through the a-amino group of 125I-mEGF (30). To contribute to understanding the sites of interaction of EGF with its receptor, we have isolated and determined the amino acid sequence of a tryptic peptide containing the site in the EGF receptor to which the N terminus of 1251-mEGF had been cross-linked. As described above, 1251-mEGF was affinity cross-linked to EGF receptor-rich A431 cells. Under the conditions described, the cross-linking of radiolabeled ligand to the receptor is essentially complete within 4 h (39). Experiments with shed membrane vesicles from A431 cells showed that the EGF cross-linked to the receptor in this manner stimulates autophosphorylation of the receptor at least as well as an equivalent amount of free mEGF (30). Triton-solubilized EGF receptor cross-linked to 1251-mEGF could be purified to -80% homogeneity in a single affinity chromatographic step, as assayed by silver staining of SDS gels of APY eluates, with recovery of >70%o of receptor-linked radioactivity (39). For the purpose of removing excess Triton from the sample, APY eluates were applied to WGL resin, and, after washing the resin in detergent-free buffer, >70o of the applied radioactivity was successfully eluted. Electrophoresis of WGL eluates in SDS gels revealed that approximately half of the eluted radioactivity comigrated with the EGF receptor, and the rest comigrated with free EGF. Control experiments (unpublished data) indicated that the covalent cross-link of 1251-mEGF to the EGF receptor is as stable as the covalent integrity of the EGF receptor itself; hence, it was concluded that the presence of free 1251-mEGF in eluates even after extensive purification of the EGF receptor is due to binding of 125I-mEGF to the receptor without covalent attachment. After dialysis for the purpose of removing salts and glycerol, 98% of the radioactivity present in the WGL eluate was recovered in the dialysate and concentrated. EGF receptor cross-linked to 125I-mEGF was reduced, carboxyamidomethylated, and subjected to tryptic digestion. An aliquot of the digest was fractionated by electrophoresis,

Proc. Natl. Acad. Sci. USA 89 (1992)

7803

with control lanes containing undigested 125I-mEGF-linked receptor, undigested 1751I-mEGF, and digested 125I-mEGF to assay for digest completeness. The presence of species in the digest of "251-mEGF-linked receptor that could be attributed to a tryptic fragment of I251-mEGF covalently linked to a tryptic fragment(s) of the EGF receptor was determined in the following manner. The mI radiolabel in '25I-mEGF is carried on tyrosyl residues; since all cleavage sites for trypsin are C-terminal to all tyrosyl residues, a complete tryptic digest of 125I-mEGF can be expected to give rise to a single radiolabeled peptide derived from the N-terminal portion of 5I-mEGF (Fig. 1). Furthermore, since cross-linking proceeds through the N terminus of 125I-mEGF, linkage to a single EGF receptor peptide would likewise be expected to give rise to a unique radiolabeled peptide. The results of digests with trypsin (Fig. 2) approximate these expectations. Trypsin-cleaved 125I-mEGF was visualized on autoradiographs as a diffuse band that migrated more slowly than intact '25I-mEGF. This anomalous migration may be due to the acidic nature of mEGF and the loss of hydrophobic residues upon cleavage with trypsin. A radiolabeled species comigrating with trypsin-digested 125I-mEGF was present in the tryptic digest of EGF receptor that had been cross-linked to 125I-mEGF; this is expected, due to the presence of the bound but unlinked 125I-mEGF carried through the receptor purification described above. Additionally present in the digest of 125I-mEGF-linked receptor, however, was a species with approximately the same migration as intact l251-mEGF. It was concluded that this uniquely migrating band is likely to represent a receptor-derived peptide linked to the N terminus of trypsin-treated "-'ImEGF. Attempts to transfer this peptide electrophoretically to a poly(vinylidene difluoride) sequencing membrane (40) gave rise to poor yields of transferred protein, probably because the hydrophobic residues in mEGF that may mediate binding to poly(vinylidene difluoride) are lost after tryptic digestion. Moreover, multiple PTH amino acid derivatives were detected in each sequencing cycle when a poly(vinylidene difluoride) membrane containing a small amount of bound peptide was subjected to sequence analysis (unpublished data). It was apparent that a single electrophoretic step was not sufficient to separate the products of the tryptic digest and that other media for peptide adsorption for sequencing were indicated.

0

FIG. 1. Modifications present in mEGF, which was affinity cross-linked to the EGF receptor. The N terminus of SSFSBmodified mEGF bears the fluorosulfonylbenzoyl moiety through which cross-linking to the EGF receptor occurs. Tyrosyl residues (*) carry the radiolabel in MI-mEGF. Sites of tryptic cleavage of reduced carboxyamidomethylated mI-mEGF are denoted by t.

7804

Biochemistry: Woltjer et al. 2

A

3

4

3

3

1

Proc. Nad. Acad. Sci. USA 89 (1992)

2

-F:.

-'I

-IF -(,1t

FIG. 2. Tricine gel electrophoresis of the products of tryptic digestion of mlI-mEGF-linked receptor. Samples of intact or trypsindigested 125I-mEGF or 125I-mEGF-linked receptor were prepared and separated by electrophoresis as described in the text; the figure consists of autoradiographs of unfixed, dried polyacrylamide gels. A431 membrane vesicles were used as a source of receptor in A, which depicts intact 1251-mEGF-linked receptor (EGFR-EGF) with copurified u5I-mEGF (EGF) (lane 1), digested 125I-mEGF-linked receptor (lane 2), intact 125I-mEGF (lane 3), and digested 125I-mEGF (lane 4). The application of trypsin to lUI-mEGF-linked receptor under the conditions described is seen to generate two bands in the low molecular weight region of the gel. The more slowly m ting band comigrates with trypsin-digested 'zI-mEGF (tEGF), and the remaining band, which comigrates with intact 125I-mEGF, was attributed to a tryptic fragment of the EGF receptor linked to trypsin-digested lzI-mEGF (tEGFR-EGF). In B, A431 cells were the source of the purified, trypsin-treated 125I-mEGF-linked receptor of lane 1, and two less-well-resolved bands can again be distinguished. The more quickly migrating band comigrates with the intact 'ZImEGF present in lane 2. The sample from which the aliquot of lane 1 was derived was further purified and sequenced successfully.

Serum containing antibody directed against native mEGF was found also to immunoprecipitate reduced, carboxyamidomethylated, trypsin-treated 125I-mEGF (39). Hence, antimEGF antibody and PAS were applied to the tryptic digest of the purified preparation of EGF receptor linked to 1251mEGF; 49%o of the radioactivity present in the digest resided in washed immunoprecipitates. Electrophoresis and autoradiography of the immunoprecipitate revealed the presence of both trypsin-treated 125I-mEGF and the species believed to be 125I-mEGF-linked EGF receptor fragment. Approximately 26 pmol of the latter species (as estimated from the radioactivity present) was eluted from the gel, and 13 pmol of eluate was successfully applied to a Porton protein support disc. 1251-mEGF-linked peptides derived from two large-scale

affinity cross-linking experiments were sequenced; experiment B was carried out by using essentially the same procedures as in experiment A described above. Sequencing chromatograms from both experiments indicated the presence of a single peptide (Fig. 3), which was recognized as the fragment produced by tryptic cleavage at Arg-84 in domain I Cycle

1

#

of the receptor, by comparison to a tryptic cleavage map of the cDNA-derived receptor sequence. Under conditions of complete digest, this peptide would be predicted to contain 21 receptor residues, in addition to 41 residues from linked, trypsin-treated I251-mEGF. PTH-derivatized amino acids attributable to the amino acid sequence of mEGF were not detected, lending further support to earlier work (30), which suggested that cross-linking occurred through the terminal amino group of 125I-mEGF. Lysyl, tyrosyl, cysteinyl, and histidyl residues are the most likely targets in the receptor for reaction with 175I-mEGF bearing a reactive fluorosulfonylbenzoyl moiety. Linkages to lysyl or tyrosyl residues would be expected to be stable to sample treatment with trifluoroacetic acid during the course of sequencing (41) and would give rise to chromatograms with low levels of the corresponding PTH-derivatized amino acids in cycles in which these residues would be predicted from the receptor sequence. Fig. 4 shows that signals from tyrosyl residues are present at approximately the expected yields in sequencing cycles 4, 5, and 9; these residues, then, appear not to have been sites of linkage to 1zI-mEGF. The signal for Tyr-101 of cycle 17, however, is much diminished in both experiments A and B. This lends support to the hypothesis that Tyr-101 is a site of linkage to 125I-mEGF; and the nearly quantitative absence of the signal for Tyr-101 in sequencing chromatograms implies that Tyr-101 is virtually the only site of linkage. The accelerated loss of sequenceable peptide in cycles after cycle 17 is consistent with this hypothesis, since sequencing cycle 17 would have resulted in cleavage of the remainder of the receptor firgment from w'I-mEGF, which may have helped tether the fragment to the sequencing support. 1.0

0.0

o -1.0 '0 0 2 la

-1.5

U2

2 -2.0 2 4 6 8 10 12 14 16 18 2 Do

'4.4

0

*1 0.5 1-1

.0 0.5

Experiment B

XXMYYEXSYALAVLSX-DA

-1.5

101

0 .00

~0.0

-1.0

84

B

10.0

la on

XXMYYXNSYALAVLXN-D

Receptor Pos.

--

o -0.5

Experiment A

ReoeptorSeq. IRGNMYYENSYALAVLSNYDANKT

A

0.5

-0.5

17

-...

*0

1

2 4 6 8 10 12 14 16 18 20

Cycle

105

FIG. 3. Sequence analysis of a tryptic peptide of the EGF receptor affinity cross-linked to 125I-mEGF. Identified residues in two experiments (A and B) are aligned with the corresponding portion of the cDNA-derived EGF receptor sequence. The sequencing cycle is given in the top line, and the positions ofrelevant receptor residues are in the bottom line. X, sequencing cycles for which residues could not be identified due to contaminants in chromatograms, which precluded identification and quantitative evaluation of residues. Dashes indicate sequencing cycles in which a residue could not be identified despite the absence of contaminant peaks at or near the anticipated residue in sequencing chromatograms.

FIG. 4. Yields of PTH-derivatized amino acid derivatives observed during sequence analysis of a tryptic peptide of the receptor affinity cross-linked to mI-mEGF. The background-subtracted yields of the residues identified in Fig. 3 are plotted as a function of sequencing cycle number. Line represents least-squares fit of the logarithm of the yields to the cycle number, where cycle 17 was excluded from the fit calculation. (A) Results from experiment A. (B) Results from experiment B described in the text. Initial sequencing yields were estimated to be 509O, and apparent repetitive yields were -85%. PTH-derivatized amino acids were indistinguishable from background noise after cycle 18 in experiment A and after cycle 19 in experiment B.

Biochemistry: Woltjer et al. The cDNA-derived receptor sequence shows that Tyr-101 is followed by the tryptic cleavage sites Lys-105, Lys-109, and Arg-114; that no cysteinyl or histidyl residues exist within residues 85-114; and that no additional tyrosyl residues are located within residues 102-114. Therefore, if Tyr101 is not the site of linkage to 125I-mEGF, Lys-105 and Lys-109 would be the next most likely candidates. In two previous cross-linking studies (26, 27), methods that did not involve direct sequencing were used to identify domain III of the EGF receptor as containing sites of linkage to 125I-mEGF. With the result reported here, the results from cross-linking experiments as a whole complement and add detail to studies of receptor mutants, which imply roles for both domains I and III in the binding of EGF to its receptor. It should be noted that affinity cross-linking is essentially always an "exo" labeling technique in the terminology of Baker (42)-i.e., with the labeling reaction occurring outside of the binding site itself. That the cross-linking reagent used in other studies, disuccinimidyl suberate (DSS) (43), which was also expected to mediate cross-linking through the N terminus of 125I-mEGF, was longer and more flexible than the SSFSB used here may account for the different results obtained for sites of crosslinking to the receptor. Moreover, in the direct cross-linking experiments reported elsewhere, the possibility of initial modification by DSS of receptor sites that are readily accessible to the cross-linking reagent, followed by cross-linking to residues of 125I-mEGF other than its N terminus, cannot be excluded. DSS could be expected to display some reactivity toward, for example, tyrosyl residues of 125I-mEGF that could be present in high local concentration at the site of receptor modification by DSS. The thorough characterization of cross-linker-modified mEGF in our studies (30), however, eliminates the possibility of cross-linking at sites other than the N terminus of 125I-mEGF. The affinity cross-linking described here was performed with membrane-resident, intact EGF receptor. The most specific cross-linking result published to date (27), as well as all electron microscopic and x-ray crystallographic imaging of the EGF receptor's ligand binding domain, were derived from work with secreted or truncated receptor forms. Although studies with conformation-sensitive antibodies have failed to detect differences between these species and the extracytoplasmic portion of the intact receptor (27), it is known that the intact, unsolubilized receptor binds 125ImEGF with 100-fold higher affinity than the secreted or truncated ligand-binding portion (28, 44, 45). It is conceivable that differences in the affinity with which 125I-mEGF is bound to its receptor could be reflected in differences in receptor residues in proximity to the N terminus of cross-linkerderivatized 12-I-mEGF. Interestingly, a recent report indicates that DSS-mediated cross-linking of 125I-mEGF to the EGF receptor may not occur in all preparations of the extracytoplasmic portion of the receptor (46). The cross-linking methods described here enabled the first elucidation, by direct protein microsequencing, of a site of interaction of EGF and the EGF receptor. It is hoped that these techniques have further potential for contributing to our understanding of how EGF interacts with its receptor, and for laying a foundation for the asking of detailed questions about the functions of EGF binding. We acknowledge the assistance of U. Barnela in the preparation and radiolabeling of mEGF. We are grateful to D. Sanchez for her work with A431 cell cultures. We thank Dr. L. Weclas-Henderson for the preparation of SSFSB. This work was supported by Grants R01 DK25489, R01 GM30861, T32 GM08320, T32 GM07347, and

2 S07 RR05424-30-41 from the National Institutes of Health.

Proc. Natl. Acad. Sci. USA 89 (1992) 1. 2. 3. 4. 5. 6. 7.

8. 9. 10. 11. 12. 13. 14.

15. 16. 17. 18. 19. 20. 21. 22. 23.

24. 25. 26. 27. 28.

29. 30. 31. 32. 33. 34. 35. 36.

37. 38. 39. 40. 41. 42. 43. 44.

45. 46. 47.

7805

Cohen, S. (1960) Proc. Natl. Acad. Sci. USA 46, 302-311. Cohen, S. (1962) J. Biol. Chem. 237, 1555-1562. Carpenter, G. & Wahl, M. I. (1990) Handb. Exp. Pharmacol. 95,69-171. Boni-Schnetzler, M. & Pilch, P. F. (1987) Proc. Natd. Acad. Sci. USA 84, 7832-7836. Yarden, Y. & Schlessinger, J. (1987) Biochemistry 26, 1434-1442. Yarden, Y. & Schlessinger, J. (1987) Biochemistry 26, 1443-1451. Cochet, C., Kashles, O., Chambaz, E. M., Borrello, I., King, C. R. & Schlessinger, J. (1988) J. Biol. Chem. 263, 3290-3295. Northwood, I. C. & Davis, R. J. (1988) J. Biol. Chem. 263, 7450-7453. Fanger, B. O., Stephens, J. E. & Staros, J. V. (1989) FASEB J. 3, 71-75. Carpenter, G., King, L., Jr., & Cohen, S. (1979) J. Biol. Chem. 254, 4884-4891. Ushiro, H. & Cohen, S. (1980) J. Biol. Chem. 255, 8363-8365. Buhrow, S. A., Cohen, S. & Staros, J. V. (1982) J. Biol. Chem. 257, 4019-4022. Buhrow, S. A., Cohen, S., Garbers, D. L. & Staros, J. V. (1983) J. Biol. Chem. 258, 7824-7827. Ullrich, A., Coussens, L., Hayflick, J. S., Dull, T. J., Gray, A., Tam, A. W., Lee, J., Yarden, Y., Libermann, T. A., Schlessinger, J., Downward, J., Mayes, E. L. V., Whittle, N., Waterfield, M. D. & Seeburg, P. H. (1984) Nature (London) 369, 418-424. Cohen, S., Ushiro, H., Stoscheck, C. & Chinkers, M. (1982) J. Biol. Chem. 257, 1523-1531. Gill, G. N., Bertics, P. J. & Santon, J. B. (1987) Mol. Cell. Endocrinol. 51, 169-186. Staros, J. V., Fanger, B. O., Faulkner, L. A., Palaszewski, P. P. & Russo, M. W. (1989) in Receptor Phosphorylation, ed. Moudgil, V. K. (CRC, Boca Raton, FL), pp. 227-242. Ullrich, A. & Schlessinger, J. (1990) Cell 61, 203-212. Murthy, U., Basu, A., Rodeck, U., Herlyn, M., Ross, A. H. & Das, M. (1987) Arch. Biochem. Biophys. 252, 549-560. Wu, D., Wang, L., Sato, G. H., West, K. A., Harris, W. R., Crabb, J. W. & Sato, J. D. (1989) J. Biol. Chem. 264, 17469-17475. Defize, L. H., Boonstra, J., Meisenhelder, J., Kruijer, W., Tertoolen, L. G. J., Tilly, B. C., Hunter, T., van Bergen en Henegouwen, P. M. P., Moolenaar, W. H. & de Laat, S. W. (1989) J. Cell Biol. 109, 2495-2507. Bellot, F., Moolenaar, W., Kris, R., Mirakhur, B., Verlaan, I., Ullrich, A., Schlessinger, J. & Felder, S. (1990) J. Cell Biol. 110, 491-502. Lax, I., Fischer, R., Ng, C., Segre, J., Ullrich, A., Givol, D. & Schlessinger, J. (1991) Cell Regul. 2, 337-345. Lax, I., Bellot, F., Howk, R., Ullrich, A., Givol, D. & Schlessinger, J. (1989) EMBO J. 8, 421-427. Lax, I., Bellot, F., Honegger, A., Schmidt, A., Ullrich, A., Givol, D. & Schlessinger, J. (1990) Cell Regul. 1, 173-188. Lax, I., Burgess, W. H., Bellot, F., Ullrich, A., Schlessinger, J. & Givol, D. (1988) Mol. Cell. Biol. 8, 1831-1834. Wu, D., Wang, L., Chi, Y., Sato, G. H. & Sato, J. D. (1990) Proc. Nail. Acad. Sci. USA 87, 3151-3155. Lax, I., Mitra, A. K., Ravera, C., Hurwitz, D. R., Rubinstein, M., Ullrich, A., Stroud, R. M. & Schlessinger, J. (1991) J. Biol. Chem. 266, 13828-13833. Gunther, N., Betzel, C. & Weber, W. (1990) J. Biol. Chem. 265, 22082-22085. Woltjer, R. L., Weclas-Henderson, L., Papayannopoulos, I. A. & Staros, J. V. (1992) Biochemistry 31, in press. Savage, C. R., Jr., & Cohen, S. (1972) J. Biol. Chem. 247, 7609-7611. Carpenter, G. & Cohen, S. (1976) J. Cell Biol. 71, 159-171. Frackelton, A. R., Jr. (1983) Cancer Cells 3, 339-345. Wahl, M. I., Daniel, T. 0. & Carpenter, G. (1988) Science 241, 968-970. Frackelton, A. R., Jr., Ross, A. & Eisen, H. (1983) Mol. Cell. Biol. 3, 1353-1360. Stone, K. L., LoPresti, J. B., Crawford, J. B., DeAngelis, R. & Williams, K. R. (1989) in A Practical Guide to Protein and Peptide Purification for Microsequencing, ed. Matsudaira, P. T. (Academic, San Diego), pp. 31-74. Schlgger, H. & von Jagow, G. (1987) Anal. Biochem. 166, 368-379. Hjerten, S. (1962) Arch. Biochem. Biophys. 99, 466-467. Woltjer, R. L. (1992) Dissertation (Vanderbilt Univ., Nashville, TN). Matsudaira, P. (1987) J. Biol. Chem. 262, 10035-10038. Annamalai, A. E. & Colman, R. F. (1981) J. Biol. Chem. 256, 1027610283. Baker, B. R. (1967) Design of Active-Site-Directed Irreversible Enzyme Inhibitors (Wiley, New York). Pilch, P. F. & Czech, N. P. (1979) J. Biol. Chem. 254, 3375-3381. Basu, A., Raghunath, M., Bishayee, S. & Das, M. (1989) Mol. Cell. Biol. 9, 671-677. Greenfield, C., Hiles, I., Waterfield, M. D., Federwisch, M., Wollmer, A., Blundell, T. L. & McDonald, N. (1989) EMBO J. 8, 4115-4123. Rubinstein, M., Felder, S., Lax, I., Zhou, M., Ullrich, A. & Schlessinger, J. (1992) FASEB J. 6, A94 (abstr.). Woljer, R. L. & Staros, J. V. (1990) FASEB J. 4, A2208 (abstr.).

Direct identification of residues of the epidermal growth factor receptor in close proximity to the amino terminus of bound epidermal growth factor.

We have recently developed a kinetically controlled, step-wise affinity cross-linking technique for specific, high-yield, covalent linkage of murine e...
1MB Sizes 0 Downloads 0 Views