HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 277, Issue 49, 47149-47159, December 6, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/49/47149    most recent M209582200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Nappi, V. M. Articles by Petti, L. M. Search for Related Content PubMed PubMed Citation Articles by Nappi, V. M. Articles by Petti, L. M. Social Bookmarking                      What's this?

Molecular Examination of the Transmembrane Requirements of the Platelet-derived Growth Factor  Receptor for a Productive Interaction with the Bovine Papillomavirus E5 Oncoprotein^*

Valerie M. Nappi, Julia A. Schaefer, and Lisa M. Petti

From the Albany Medical College, Albany, New York 12208

Received for publication, September 18, 2002

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The small transmembrane E5 protein of bovine papillomavirus (BPV) transforms cells by forming a stable complex with and activating the platelet-derived growth factor  receptor (PDGFR). The E5/PDGFR interaction is thought to involve specific physical contacts between the transmembrane domains of the two proteins. Lys⁴⁹⁹ at the extracellular juxtamembrane position and Thr⁵¹³ within the transmembrane domain of the PDGFR are required for the interaction and are predicted to contact analogously positioned residues in the E5 protein. Here, mutagenic analysis of the transmembrane region of the PDGFR was performed to further characterize the nature of the E5/PDGFR interaction. We show that the receptor transmembrane domain, with minimal extracellular and intracellular sequence, is sufficient for the interaction. In addition, we provide evidence that the polar nature of Thr⁵¹³ as well as its positioning along the transmembrane -helix is important for the interaction. We also identify the receptor transmembrane amino acids Ile⁵⁰⁶ and Leu⁵²⁰ as additional requirements for the interaction. Because Lys⁴⁹⁹, Thr⁵¹³, Ile⁵⁰⁶, and Leu⁵²⁰ all align along the same face of the predicted PDGFR transmembrane -helix, our data support the model that the PDGFR contacts the E5 protein via multiple amino acids along a single -helical interface.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Bovine papillomavirus type 1 (BPV-1)¹ induces fibropapillomas in cattle and can tumorigenically transform cultured rodent fibroblast lines (1, 2). The E5 open reading frame of BPV-1 is primarily responsible for the transforming activity of BPV-1 (3-6). E5 encodes a small, 44-amino acid transmembrane protein that exists as a dimer and localizes mostly to cytoplasmic membranes (7, 8). The manner by which E5 functions to achieve transformation has been studied rather extensively. Previous studies have shown that the platelet-derived growth factor (PDGF)  receptor (PDGFR), a transmembrane receptor tyrosine kinase, is the primary cellular target of the E5 protein. Specifically, the E5 protein forms a complex with and constitutively activates the PDGFR (9, 10), and activation of this receptor by E5 is required for E5-mediated transformation (11, 12). Evidence suggests that the E5 protein binds as a dimer to two PDGFR molecules and thereby promotes receptor dimerization, resulting in receptor autophosphorylation and stimulation of its intrinsic kinase activity (13). Once the receptor is tyrosine phosphorylated, key cytoplasmic substrates can bind to the receptor and transmit signaling cascades eventuating in cell proliferation (14). Activation of the PDGFR by the E5 protein occurs independent of the native ligand, PDGFBB (11).

There have been several reports that the E5 protein can complex with other cellular transmembrane proteins such as other growth factor receptors (15, 16), -adaptin (17), and the 16-kDa subunit of the vacuolar H⁺-ATPase (18-20). However, in many of these studies, an interaction with E5 was demonstrated under conditions of overexpression, which may artificially enhance nonspecific interactions. Indeed, it has been shown that although the E5 protein can interact with several different growth factor receptors under conditions of transient overexpression in COS cells (16), it can interact only with the PDGFR when stably expressed at normal levels (16, 21). Furthermore, since the ability of E5 to interact with these other proteins does not correlate with its transforming activity, the biological significance of such interactions has not been established. Therefore, the PDGFR appears to be the most specific target of the E5 protein, and complex formation with this receptor is clearly related to a biochemical (receptor activation) and biological (cellular transformation) response (9-12).

In attempts to characterize the E5-PDGFR complex, mutagenic analysis of both proteins has been performed. Initial studies indicated that the E5 protein and the PDGFR contact each other via transmembrane domain interactions, suggesting a novel mechanism of activation for this receptor (22-24). Subsequent studies identified two potential sites of contact between the transmembrane regions of these two proteins. Specifically, it was shown that Lys⁴⁹⁹ at the outer juxtamembrane position and Thr⁵¹³ at a central transmembrane position within the receptor were required for stable complex formation with the E5 protein (23). Interestingly, the analogously positioned Asp³³ and Gln¹⁷, respectively, in the E5 protein, were found to be necessary for the transforming activity of E5 (25, 26) as well as its ability to form a complex with and activate the PDGFR (27). Replacing Lys⁴⁹⁹ in the receptor with Asp, Glu, or Ala hindered an interaction with E5, whereas an Arg substitution was tolerated, suggesting a requirement for a positive charge at this position (23). Similar studies revealed a requirement for a negative charge at the corresponding juxtamembrane position (Asp³³) of the E5 protein (27-29). Furthermore, it was shown that only amino acids with side groups capable of hydrogen bond formation could functionally substitute for Gln¹⁷ in E5 and permit an interaction with the PDGFR (20). Thus, it was proposed from these studies that complex formation between the PDGFR and the E5 protein involves an electrostatic interaction between Asp³³ of E5 and Lys⁴⁹⁹ of PDGFR and hydrogen bond formation between Gln¹⁷ of E5 and Thr⁵¹³ of the receptor (23, 28-30). It was also found that dimerization of E5, which is mediated by two extracellular cysteines, is necessary for transformation and stable complex formation with the PDGFR (25, 27). This implies that dimerization of E5 promotes a conformation suitable for making contacts with the PDGFR.

We recently obtained data suggesting that several other amino acids within the PDGFR transmembrane domain besides Lys⁴⁹⁹ and Thr⁵¹³ may be required for a stable interaction with the E5 protein (31). Here, additional mutagenesis analysis of the PDGFR was performed to further identify and characterize the PDGFR requirements for this interaction. First, we showed that a minimal segment of the PDGFR consisting primarily of the transmembrane domain is capable of forming a complex with the E5 protein. We also provide evidence that Thr⁵¹³ is important for the interaction by virtue of its polar nature and its position along the transmembrane -helix. Finally, we identify Ile⁵⁰⁶ and Leu⁵²⁰ as additional receptor transmembrane amino acids that play a role in the interaction. This stands to reason because Ile⁵⁰⁶ and Leu⁵²⁰ are predicted to align with Lys⁴⁹⁹ and Thr⁵¹³ along the same face of the PDGFR transmembrane -helix when in a left-handed coiled coil complex with another transmembrane -helix (23). Taken together, these data suggest that the PDGFR contacts the E5 protein via multiple amino acids aligned along a single -helical interface.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Construction of Truncated and Mutant Receptors-- A doubly truncated receptor construct containing primarily the transmembrane domain of the PDGFR was created by ligating the truncated portion of a receptor construct lacking the extracellular domain with that of one lacking the intracellular domain at a common transmembrane site. The first step was to attach the COOH-terminal 13 amino acids of the human PDGFR (the epitope recognized by the PDGF receptor-specific antibody used in these studies) to the COOH terminus of the truncated receptor construct lacking most of the intracellular domain, pECTM (gift from C. Heldin, Ludwig Institute for Cancer Research, Uppsala, Sweden; Ref. 32). A double-stranded oligonucleotide linker with the sense sequence, 5'-GCCCTGCGCCTCGAGCGGAAGCAGAGGATAGCTTCCTGTAAGCT-3', encoding this epitope and containing SacI compatible ends (gift from D. DiMaio, Yale University) was inserted in-frame into the SacI site located at the stop codon of the receptor open reading frame in pECTM. The resulting construct pECTM-epi was subcloned into the pLXSN retrovirus vector by standard methods, generating pECTM-epi-LXSN. Another truncated receptor construct, TPR (gift from C. Heldin), lacking most of the extracellular domain, was also subcloned into the pLXSN retrovirus vector (11), generating pTPR-LXSN. We exploited the XcmI site located in the transmembrane domain of the human PDGFR and the SacII site located in pLXSN, and ligated the XcmI-SacII fragment of pECTM-epi-LXSN to the SacII-XcmI fragment of pTPR-LXSN to generate pTMPR. Sequence analysis confirmed that this TMPR lacks extracellular amino acids 38-530 and intracellular amino acids 574-1060, corresponding to the published sequence of the human receptor (33).

Site-directed mutagenesis was performed using the QuikChange procedure (Stratagene) as described by the manufacturer to introduce single or double amino acid substitutions into the transmembrane domain of the PDGFR. For each mutation, complimentary oligonucleotides were designed to contain the appropriate base pair mismatches with respect to the wild type receptor sequence (33, 34) that would achieve the desired mutation(s). In the case of I506A, a SpeI site was established by introduction of a silent mutation concomitantly with the I506A substitution, which allowed for screening of potential mutants by SpeI digestion. Templates included the murine or human (either wild type or mutant) PDGFR cDNA cloned into the LXSN retroviral vector, which also contains the G418 resistance gene as a selectable marker. The plasmid DNA products from site-directed mutagenesis were sequenced to confirm the presence of the desired mutations.

Cell Culture-- The Phoenix ecotropic retrovirus producer cell line was obtained from the ATCC with permission from Dr. Gary Nolan (Stanford University) and maintained in Dulbecco's minimal essential medium with high glucose supplemented with 10% fetal bovine serum. Ba/F3 murine hematopoetic cells were maintained as previously described (11) in RPMI 1640 medium supplemented with 10% fetal bovine serum, antibiotics, 50 µM -mercaptoethanol, and 10% WEHI conditioned medium as a source of IL-3 (RPMI/IL-3).

Production of Retrovirus-- The various PDGFR constructs, E5, and v-sis were introduced into Ba/F3 cells by retroviral mediated gene transfer. The recombinant retroviral vectors used were the receptor-LXSN constructs described above and E5 or v-sis subcloned into the pBabepuro retroviral vector, which contains the puromycin resistance marker. High titer ecotropic retrovirus was obtained from these retroviral vectors as described previously (35). Briefly, Phoenix ecotropic cells grown to 70-80% confluence in 60-mm dishes were transfected with 10 µg of plasmid DNA using the calcium phosphate method. The next day the media was replaced, and 24 h later the virus-containing supernatant from each dish was collected and filtered through a 0.45-µm syringe filter.

Establishment of Stable Ba/F3 Cell Lines-- Retroviral infection of Ba/F3 cells was performed as described previously with some modifications (11). First, Ba/F3 cells stably expressing E5 or v-sis were established by infecting ~5 × 10⁶ cells with ~1-2 × 10⁵ colony forming units of pBabepuro-E5 or pBabepuro-v-sis ecotropic retrovirus in 10 ml of RPMI/IL-3 supplemented with 4 µg/ml Polybrene. Cells expressing no viral oncogene were generated in parallel by infection with retrovirus derived from the pBabepuro vector alone. 48 h post-infection, 2 ml of the infected cells was added to 8 ml of selective media (RPMI/IL-3 containing 1 µg/ml puromycin (Sigma)). After reaching a density of ~10⁶ cells/ml, cells were passaged again under selection conditions. Selection was repeated through 2-4 additional passages until 100% of a mock-infected culture died, thus establishing stable cell lines. The resulting cells expressing E5, v-sis, or no viral oncogene (Puro) were then infected with recombinant ecotropic retroviruses containing the various LXSN-receptor constructs as described above. Stable cell lines were generated using selective media containing 1 mg/ml G418 (Gemini) as well as puromycin.

Assay for IL-3-independent Growth-- Ba/F3 cells expressing a receptor construct without (Puro) or with E5 or v-sis were grown to an approximate density of 1 × 10⁶ cells/ml, washed twice with phosphate-buffered saline, and resuspended in an equal volume of RPMI lacking IL-3 (RPMI/IL-3; RPMI 1640 medium supplemented with only 1% fetal bovine serum, antibiotics, 50 µM -mercaptoethanol, and without WEHI conditioned medium). Approximately 5 × 10⁵ of these cells were inoculated into 10 ml of RPMI/IL-3, incubated at 37 °C, and monitored for growth. Total or viable cells were counted using a hemacytometer at various times after seeding. For experiments testing the murine PDGFR, Ba/F3 cells that proliferated at least 20-fold during a 10-day period were considered IL-3-independent. For experiments involving the human PDGFR, cells that proliferated at least 10-fold during a 10-day period were considered IL-3-independent. Multiple independently derived cell lines of each genotype were tested to the confirm results.

Immunoprecipitation and Immunoblotting-- Ba/F3 cells were lysed by incubation in cold EBC buffer (50 mM Tris-HCl, pH 8.0, 120 mM NaCl, 0.5% Nonidet P-40) supplemented with 1 mM phenylmethylsulfonyl fluoride, 2 mM sodium orthovanadate, 20 mg/ml leupeptin, and 20 mg/ml aprotinin on ice for 15 min. Lysates were cleared of nuclei and cell debris by centrifugation in a microcentrifuge at 15,000 rpm for 10 min at 4 °C and the supernatant extracts were used for immunoprecipitation. To immunoprecipitate the E5 protein and any associated protein, 750-1100 µg of extracted protein was incubated overnight at 4 °C with ~10 µl of a rabbit polyclonal antibody directed against the 16 COOH-terminal amino acids of the E5 protein (gift from D. DiMaio). To immunoprecipitate the wild type and mutant forms of the PDGFR, 500-1200 µg of extracted protein was incubated overnight at 4 °C with 5-12 µl of a rabbit polyclonal antibody directed against the COOH-terminal 13 amino acids of the human PDGFR (gift from D. DiMaio). Following incubation with primary antibody, extracts were incubated with 60 µl of a 1:1 slurry of protein-A-Sepharose CL-4B beads (Amersham Biosciences) in Tris-buffered saline (TBS) containing 10% bovine serum albumin for 60 min at 4 °C. Beads were then washed 3-5 times with cold EBC buffer. For the experiments presented in Figs. 3A and 5A, the PDGFR was precipitated from cell extracts through affinity purification of glycosylated proteins with wheat germ lectin (WGL)-Sepharose beads (Amersham Biosciences). Approximately 100 µl of a 1:1 slurry of WGL beads was incubated with 800-1000 µg of EBC extracts overnight at 4 °C and then washed as described above. Protein complexes were dissociated from beads by boiling in 2× Laemmli protein sample buffer.

SDS-PAGE and immunoblot analysis was performed as described previously (23). Briefly, immunoprecipitates were either electrophoresed on an SDS-7.5% polyacrylamide gel and transferred to nitrocellulose (for PDGF receptor or phosphotyrosine imunoblotting) or electrophoresed on an SDS-15% polyacrylamide gel and transferred to Immobilon (Millipore) (for E5 immunoblotting). Blots were blocked for 2 h in milk buffer (5% nonfat dry milk in TBST (10 mM Tris-HCl, pH 7.4, 167 mM NaCl, 1% Tween 20)), then incubated overnight with either a 1:2000 or 1:1000 dilution of monoclonal anti-phosphotyrosine antibody P-Tyr-100 (Cell Signaling) or 4G10 (Upstate Biotechnology), respectively, or a 1:500-1:1000 dilution of the anti-PDGFR or anti-E5 antiserum described above. Following incubation with primary antibody, immunoblots were washed 5 times, 10 min each, in either TBST buffer for phosphotyrosine and E5 immunoblots or TNET buffer (10 mM Tris-HCl, pH 7.4, 50 mM NaCl, 1% Tween 20) for PDGF receptor immunoblots. Each blot was then incubated for 1 h with a 1:5000 dilution of a protein A (Pierce; for PDGFR or E5 blots) or goat anti-mouse IgG (Pierce; for anti-phosphotyrosine blots) horseradish peroxidase conjugate, washed as above, and subjected to enhanced chemiluminescence (ECL) detection (Amersham Biosciences) as described by the manufacturer. For the experiment presented in Fig. 7A, after ECL detection of the phosphotyrosine immunoblot, primary and secondary antibodies were removed according to the stripping protocol provided by the manufacturer. In brief, membranes were incubated in stripping buffer (100 mM -mercaptoethanol, 2% SDS, 62.5 mM Tris-HCl, pH 6.7) for 30 min at 60 °C. The stripped blot was then washed and subjected to ECL detection to ensure that the stripping process was effective. The membrane was then washed in TBST several times, blocked in 5% milk-TBST, and subjected to anti-PDGFR immunoblotting as described above.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Previous work identified two specific amino acids of the PDGFR, Lys⁴⁹⁹ at the extracellular juxtamembrane position and Thr⁵¹³ within the transmembrane domain, that are required for an interaction with the BPV E5 protein (23). To further elucidate the nature of the E5/PDGFR interaction and to determine whether other amino acids in the receptor are required for the interaction, additional mutagenesis analysis of the receptor was performed. Receptor mutants were examined for an interaction with the E5 protein in the mouse hematopoietic Ba/F3 cell line because these cells do not express endogenous PDGF receptors, which might otherwise obscure the analysis of mutant receptors. Furthermore, these cells provide a convenient assay for the functional cooperation of receptor mutants with E5 because they normally depend on IL-3 for survival and proliferation (36). Co-expression of a growth factor receptor and its cognate ligand in these cells can alleviate this requirement for IL-3 (37) apparently because activation of the receptor by the ligand results in a compensatory proliferative signal. Hence, expression of the PDGFR with v-sis, which encodes the viral homologue of PDGF BB (38), or the BPV E5 protein in these cells allows for IL-3-independent growth (11). Therefore, after co-expressing E5 with various PDGFR mutants in Ba/F3 cells, we were able to assess the ability of receptor mutants to form a complex with the E5 protein, to undergo activation by E5, and to cooperate with E5 to induce a proliferative response. The functional integrity of each receptor mutant was ascertained by determining the ligand-induced responsiveness to v-sis.

The Transmembrane Domain of the PDGFR Is Sufficient for an Interaction with the E5 Protein-- Previous work established that the transmembrane domain of the PDGFR is required for complex formation with the E5 protein and implicated that the transmembrane domains of the two proteins may contact each other directly (23, 28, 30). Here, we asked if the transmembrane domain of the receptor by itself is sufficient to interact with the E5 protein. To address this question we constructed a truncated receptor consisting primarily of the PDGFR transmembrane domain with only 6 and 30 amino acids derived from the extracellular and intracellular domains, respectively, and tested its ability to form a stable complex with the E5 protein. In this truncated receptor construct a segment of the human PDGFR containing the juxtamembrane lysine, transmembrane domain, and 17 adjacent cytoplasmic amino acids is linked to the COOH-terminal 13 amino acids, the epitope recognized by the PDGFR antiserum used in these studies (Fig. 1A). A recombinant retrovirus carrying the construct for the truncated receptor was introduced by retroviral infection into Ba/F3 cells expressing either the BPV E5 gene or no exogenous gene. Cells stably expressing the truncated receptor were established after selection for a G418 resistance marker present in the retroviral vector. Cells were first analyzed for expression of the truncated receptor by immunoblot analysis with the PDGFR antiserum. As expected, the truncated receptor was expressed as a small protein with an apparent molecular mass of ~14.5 kilodaltons (Fig. 2, top, left two lanes). Similar amounts of the truncated receptor were expressed regardless of whether or not E5 was co-expressed. To assess the ability of the truncated receptor to form a stable complex with the E5 protein, cell extracts were immunoprecipitated with an E5 antiserum, and E5 immunoprecipitates were subjected to immunoblotting for the PDGFR. Fig. 2 (top, fourth lane) shows that a significant amount of the truncated receptor could be co-immunoprecipitated with the E5 protein. This co-immunoprecipitation was not because of cross-reactivity of the E5 antibody with the truncated receptor, because no truncated receptor was precipitated in the absence of E5 expression (Fig. 2, top, third lane). Thus, co-immunoprecipitation of the truncated receptor with the E5 protein indicated the presence of a stable complex between the two proteins. Others were able to independently reproduce these results using the same truncated receptor construct.² Furthermore, complex formation between this truncated receptor and the E5 protein was also observed in human diploid fibroblasts (data not shown). Therefore, these results suggest that the PDGFR transmembrane domain is sufficient for an interaction with the E5 protein.

View larger version (21K): [in this window] [in a new window]   Fig. 1.   Structure of the truncated and mutant receptors. A, schematic depiction of the truncated TMPR receptor compared with the full-length wild type human PDGFR oriented in the membrane. Note that the truncated mutant retains the juxtamembrane Lys, the transmembrane Thr, and the COOH-terminal epitope, but not the extracellular ligand binding domain or most of the intracellular region containing the split kinase domain (diagonal lines). B, amino acid sequence of the transmembrane region of the mutant receptors compared with that of the wild type (WT) receptor. Underlined sequence denotes transmembrane domain. The amino acid sequence of the human PDGFR transmembrane domain is identical to that of the mouse except for an Ile instead of a Val at position 514 (33, 34). Boxed residues are those that have been mutated. Numbers indicate amino acid positions in the mouse PDGFR sequence (34).

View larger version (40K): [in this window] [in a new window]   Fig. 2.   Complex formation of the TMPR-truncated receptor with the E5 protein. Extracts from Ba/F3 cell lines expressing the TMPR-truncated receptor with or without E5 were immunoprecipitated (IP) with an anti-E5 (E5) or anti-PDGFR (PR) rabbit antiserum. The truncated receptor is recognized by the anti-PR antiserum because it retains the epitope to which the antiserum was raised. Immunoprecipitates were run on a SDS-15% polyacrylamide gel, transferred to Immobilon, and subjected to either PDGFR or E5 immunoblotting. The position and size, in kDa, of molecular mass standards are indicated by the numbers on the left. Note that the TMPR receptor has an electrophoretic mobility corresponding to 14.5 kDa, which is much faster than the mobility of the 200-kDa full-length wild type PDGFR. The presence of the truncated receptor (TMPR) in E5 immunoprecipitates is indicative of complex formation between TMPR and E5. Each lane represents ~110 or 1100 µg of extracted protein for the PDGFR and E5 immunoprecipitates, respectively.

Polar Amino Acid Substitutions of Thr⁵¹³ Are Tolerated for the Interaction-- Mutagenesis studies demonstrated that the juxtamembrane Lys⁴⁹⁹ and transmembrane Thr⁵¹³ of the PDGFR are required for an interaction with the BPV E5 protein (23). Whereas K499D and K499E substitution mutants were defective for the interaction, a K499R substitution mutant retained the ability to bind to and be activated by E5, implicating a requirement for a positive charge at this position (23). To further assess the nature of the requirement at transmembrane position 513, we used site-directed mutagenesis to replace Thr⁵¹³ with Ser or Asn, two different polar amino acids, generating receptor mutants T513S and T513N (Fig. 1B). The T513S substitution also was introduced in the receptor in conjunction with a K499R substitution, creating the double substitution mutant K499R/T513S (Fig. 1B). Retroviral infection was used to introduce the genes encoding the wild type, T513S, K499R/T513S, or T513N receptor into Ba/F3 cells engineered to express E5, v-sis, or no viral oncogene. Stable cell lines were established and the ability of the receptor constructs to interact with and cooperate with the E5 protein was assessed.

First, we examined the cell lines expressing the T513S and K499R/T513S receptor mutants. PDGFR immunoblot analysis (Fig. 3A, lower panel) revealed that the mutant and wild type receptors were expressed at similar levels in the various different cell lines. As is typically observed, two different exogenous PDGFR forms were evident, a slower migrating mature form and a faster migrating incompletely processed form. Next, the tyrosine phosphorylation status of these receptor mutants was assessed by phosphotyrosine immunoblot analysis and served as an indication of receptor activation. Briefly, glycoproteins were affinity purified from cell extracts with WGL-Sepharose, subjected to SDS-PAGE, and immunoblotted with an anti-phosphotyrosine antibody (Fig. 3A, upper panel). Because no endogenous tyrosine-phosphorylated glycoproteins similar in size to the PDGFR could be detected in Ba/F3 cells infected with the empty vector-containing retrovirus (Fig. 3A, LXSN lanes), only the exogenously expressed PDGFR, when activated, should be detected by this analysis. The wild type receptor when expressed alone in Ba/F3 cells displayed only minimal tyrosine phosphorylation. As expected and shown previously (11, 23) the wild type receptor co-expressed with E5 or v-sis was abundantly tyrosine phosphorylated, indicating that the E5 protein and the normal ligand activate the wild type receptor to a similar extent. Neither the T513S nor the K499R/T513S mutant when expressed alone exhibited significant tyrosine phosphorylation, suggesting that neither the single nor double amino acid substitution(s) resulted in a condition of constitutive receptor activation. Both the T513S and the K499R/T513S mutant receptors were substantially tyrosine-phosphorylated when co-expressed with v-sis, illustrating that the T513S and K499R substitutions do not negatively effect the capacity of the PDGFR to respond to ligand. Importantly, both receptor mutants were tyrosine-phosphorylated to the same extent as the wild type receptor when co-expressed with E5. Thus, the T513S mutation alone or in conjunction with the K499R mutation did not inhibit receptor activation by E5. Reproducible results were obtained when the PDGF receptor was isolated from extracts by PDGFR immunoprecipitation (data not shown).

View larger version (29K): [in this window] [in a new window]   Fig. 3.   Biochemical and functional analysis of the T513S and K499R/T513S receptor mutants. Ba/F3 cells expressing the wild type PDGFR (WT), mutant receptor T513S or K499R/T513S, or no receptor (LXSN) in the presence (+) or absence () of E5 or v-sis were established as described under "Experimental Procedures." A, glycosylated proteins were isolated from cell extracts by precipitation with WGL-Sepharose, run on an SDS-7.5% polyacrylamide gel, and transferred to nitrocellulose. Blots were subjected to anti-phosphotyrosine (PY) or anti-PDGF receptor (PR) immunoblotting to detect tyrosine-phosphorylated or total receptor levels, respectively. B, the E5 protein and PDGFR-E5 complexes were isolated from cell extracts by immunoprecipitation with an anti-E5 rabbit antiserum. Immune complexes were run on SDS-7.5 (upper panel) or 15% (lower panel) polyacrylamide gels, transferred to nitrocellulose or Immobilon, and subjected to PDGFR (PR) or E5 immunoblotting, respectively. Complex formation between the receptor and E5 is indicated by the presence of receptor in E5 immunoprecipitates. Arrows on the right delineate the mature (m) and precursor (p) forms of the PDGFRs (PR) as well as the E5 protein. Each lane represents 750 µg (A, upper panel, and B) or 170 µg (A, lower panel) of extracted protein. C, Ba/F3 cells co-expressing the indicated receptor construct with E5, v-sis, or the empty retroviral vector (Puro) were seeded into 10 ml of medium lacking IL-3 at an approximate density of 5 × 104 cells/ml. Cells were incubated in the absence of IL-3 and counted after 14 days to assess cell survival and growth. Cells not expressing exogenous receptor in the presence of E5 or v-sis did not grow in the absence of IL-3 (not shown). The graph depicted is a representative for several different sets of independently derived cell lines.

The ability of these mutant receptors to physically associate with the E5 protein was determined by a co-immunoprecipitation experiment involving PDGF receptor immunoblot analysis of E5 immunoprecipitates (Fig. 3B). As shown previously (23), both mature and precursor forms of the wild type PDGFR could be co-precipitated with the E5 protein from extracts of Ba/F3 cells (Fig. 3B, upper panel). No receptor was detected in E5 immunoprecipitates from cells not expressing E5, thereby excluding the possibility of cross-reactivity of the E5 antiserum with the receptor. Similarly, the T513S and K499R/T513S mutant receptors each could be co-immunoprecipitated with the E5 protein (Fig. 3B; data not shown). The amount of each co-precipitated mutant receptor was comparable with that of the wild type receptor. These results suggest that the T513S and K499R/T513S mutant receptors were fully capable of forming a physical complex with the E5 protein.

To examine the functionality of complexes formed between each receptor construct and the E5 protein, we took advantage of the fact that co-expression of the PDGFR with E5 or v-sis can functionally replace the IL-3 growth requirement of Ba/F3 cells (11). Ba/F3 cell lines expressing wild type, T513S, or K499R/T513S with or without E5 or v-sis were seeded into media lacking IL-3, monitored for growth, and counted. A representative graph that shows the cell densities achieved after a 14-day incubation in the absence of IL-3 is depicted in Fig. 3C. As expected, none of these receptors when expressed alone could permit IL-3-independent growth of Ba/F3 cells. As shown previously (11, 23), the wild type PDGFR when expressed with v-sis or with E5, conferred an IL-3-independent growth phenotype, indicating that the wild type PDGFR is able to functionally cooperate with either viral oncoprotein in Ba/F3 cells. Both the T513S and K499R/T513S mutant receptors, when expressed with E5 or v-sis, were able to induce IL-3-independent growth to a similar extent as the wild type receptor. Thus, the mutant receptors were not only functionally responsive to ligand, but also exhibited a wild type ability to induce proliferation in response to E5. Taken together, our results indicate that the conservative T513S substitution in the transmembrane domain of the wild type PDGFR is tolerable for a physical and functional interaction with the E5 protein.

Next, we examined the effect of the T513N mutation on receptor activity and signaling in response to E5. Tyrosine phosphorylation of the T513N receptor mutant was assessed by phosphotyrosine immunoblotting of PDGFR immunoprecipitates (Fig. 4A, upper panel). When expressed alone, the T513N mutant was only minimally tyrosine phosphorylated, indicating that the T513N mutation did not constitutively activate the receptor. Expression with E5 resulted in abundant tyrosine phosphorylation of the T513N mutant, suggesting that it was perfectly capable of being activated by the E5 protein. Interestingly, the precursor form of the T513N mutant was substantially more tyrosine phosphorylated than the wild type receptor in response to E5. As expected, the T513N mutant, like the wild type receptor, was considerably tyrosine phosphorylated when expressed with v-sis, suggesting that it also could be activated by ligand. The increases in tyrosine phosphorylation observed for this mutant in response to E5 or v-sis were not because of corresponding increases in the level of receptor expressed (Fig. 4A, lower panel).

View larger version (25K): [in this window] [in a new window]   Fig. 4.   Biochemical and functional analysis of the T513N receptor mutant. Ba/F3 cells expressing the wild type PDGFR (WT) or the T513N receptor mutant with (+) or without () E5 or v-sis were established as described under "Experimental Procedures." A, the exogenous wild type or mutant PDGFR was immunoprecipitated (PRIP) from cell extracts and subjected to anti-phosphotyrosine (PY) immunoblotting to determine the level of activated receptor in the cells (upper panel). The blot was then re-probed with the anti-PDGFR (PR) antiserum to determine the total level of receptor expressed in the cells (lower panel). Each lane represents ~450 µg of extracted protein that was immunoprecipitated. Arrows on the right indicate the mature (m) and precursor (p) forms of the PDGFRs (PR). B, IL-3-independent growth of Ba/F3 cells expressing the T513N mutant with E5. Ba/F3 cells expressing the indicated receptor without (Puro) or with E5 or v-sis were seeded into 10 ml of medium lacking IL-3 at an approximate density of 5 × 104 cells/ml and counted after 7 days. The graph depicted is representative of several different sets of independently derived cell lines.

To assess the ability of the T513N mutant to induce a proliferative signal in response to E5 or v-sis, an IL-3 independence assay was performed. Ba/F3 cells expressing the T513N mutant, with or without E5 or v-sis, were incubated in the absence of IL-3 and counted 7 days later (Fig. 4B). Cells expressing the T513N mutant alone lost viability and were unable to proliferate. In contrast, co-expression of the T513N mutant with either E5 or v-sis allowed the cells to proliferate in the absence of IL-3 as efficiently as the wild type receptor with E5 or v-sis. These results suggest that the T513N mutant was fully competent for a functional interaction with the E5 protein. Thus, two different polar amino acids can functionally substitute for Thr⁵¹³ in the receptor with respect to a productive interaction with the E5 protein.

Altering the Location of Thr⁵¹³ in the Transmembrane Domain of the PDGFR Is Inhibitory for an Interaction with the E5 Protein-- Based on a predicted -helical structure of the PDGFR transmembrane domain, both the transmembrane Thr⁵¹³ and the juxtamembrane Lys⁴⁹⁹ were proposed to fall on the same helical face (23) (Fig. 9). A previously constructed human PDGFR mutant, in which the transmembrane Thr was replaced with a Leu (23), was used as a template to examine the significance of the positioning of this Thr along the transmembrane -helix. Specifically, we introduced an I514T substitution into this mutant, generating the T513L/I514T double substitution mutant (Fig. 1B). In this mutant the position of the transmembrane Thr is effectively shifted one residue from its native location to an adjacent helical face with respect to the juxtamembrane Lys (Fig. 9, compare the position of residues 513 and 514). To determine the effect of the I514T substitution when the Thr is maintained at its native location, this substitution was also introduced into the wild type receptor, generating the I514T mutant (Fig. 1B). The I514T and T513L/I514T mutants were then assessed for their ability to physically and functionally interact with the E5 protein in Ba/F3 cells.

To determine whether these receptor mutants could be activated by the E5 protein in Ba/F3 cells, phosphotyrosine immunoblotting of WGL precipitates or PDGFR immunoprecipitates was performed as shown in Figs. 5A and 6A, upper panels. As shown previously (23), the wild type human PDGFR exhibited substantial tyrosine phosphorylation in response to either v-sis or E5 stimulation. Both the T513L/I514T and I514T mutants displayed a level of receptor tyrosine phosphorylation that was comparable with the wild type receptor when expressed with v-sis (Figs. 5A and 6A, respectively). This indicates that the amino acid substitutions in these mutants did not inhibit ligand-dependent activation of the receptor. The I514T receptor, when co-expressed with E5, exhibited a level of tyrosine phosphorylation comparable with the wild type receptor co-expressed with E5 (Fig. 6A). However, when co-expressed with E5, the T513L/I514T mutant displayed no detectable tyrosine phosphorylation, suggesting that the E5 protein was unable to induce activation of this receptor mutant. The inhibition of E5-induced activation of this receptor mutant was not because of decreased receptor expression, as receptor expression levels were similar in the cell lines examined (Fig. 5A, lower panel). These data suggest that it is the altered location of the transmembrane Thr in the T513L/I514T mutant rather than the I514T replacement itself that is responsible for abrogating receptor activation in response to the E5 protein.

View larger version (26K): [in this window] [in a new window]   Fig. 5.   Biochemical and functional analysis of the T513L/I514T receptor mutant. Ba/F3 cells expressing the wild type human PDGFR (WT), the T513L/I514T mutant receptor, or no exogenous receptor (LXSN) with (+) or without () E5 or v-sis were established as described under "Experimental Procedures." The PDGFR (A) or the E5 protein (B) was precipitated from cell extracts as in Fig. 3. A, WGL were subjected to anti-phosphotyrosine (PY) immunoblot analysis for assessment of PDGFR activation (upper panel) or anti-PDGFR (PR) immunoblotting to determine the receptor expression levels (lower panel). B, E5 immunoprecipitates (E5IP) were subjected to PDGFR immunoblotting to assess the presence of a physical complex between the receptor and E5 (upper panel), or anti-E5 (E5) immunoblotting to detect E5 protein expression levels (lower panel). Each lane represents 670 (upper panel of A), 170 (lower panel of A), or 750 µg (B) of extracted protein. The arrows on the right denote the mature (m) and precursor (p) forms of the PDGFR (PR) and the E5 protein. C, IL-3-dependent phenotype of Ba/F3 cells co-expressing the T513L/I514T mutant and E5. Ba/F3 cells expressing the indicated receptor construct with E5, v-sis, or the empty retroviral vector (Puro) were seeded into medium lacking IL-3 at a density of 5 × 105 cells per 10 ml and counted 11 days later. Cells expressing no receptor (LXSN) with v-sis were counted after 9 days. The graph shown is representative of several different sets of independently derived cell lines.

View larger version (29K): [in this window] [in a new window]   Fig. 6.   Biochemical and functional analysis of the I514T receptor mutant. Ba/F3 cells expressing the wild type human PDGFR (WT) or the I514T mutant receptor with or without () E5 or v-sis were established as described under "Experimental Procedures." A, exogenously expressed wild type or mutant PDGFR was immunoprecipitated (PRIP) from cell extracts and subjected to immunoblotting with an anti-phosphotyrosine antibody (PY) to assess activated receptor levels (upper panel). The blot then was re-probed with an anti-PDGFR antiserum to determine the levels of total receptor expressed (lower panel). Each lane represents ~900 µg of extracted protein that was immunoprecipitated. Arrows on the right indicate the mature (m) and precursor (p) forms of the PDGFR (PR). B, IL-3-independent growth of Ba/F3 cells expressing the I514T mutant with E5. Ba/F3 cells expressing the indicated receptor without (Puro) or with E5 or v-sis were seeded into 10 ml of medium lacking IL-3 at an approximate density of 5 × 104 cells/ml and counted 13 days later. The graph depicted is representative of three different sets of independently derived cell lines.

To determine whether the T513L/I514T mutant could form a stable physical complex with the E5 protein, PDGFR immunoblot analysis of E5 immunoprecipitates was performed as shown in Fig. 5B. Both mature and precursor forms of the wild type receptor could be co-precipitated with the E5 protein. In contrast, no co-immunoprecipitation of the T513L/I514T mutant with the E5 protein could be detected. This lack of detectable co-immunoprecipitation was not because of a reduction of E5 expression, because the level of E5 expressed with the mutant receptor was similar to that expressed with the wild type receptor (Fig. 5B, lower panel). Instead, these results indicate that the T513L/I514T mutant lost the ability to form a stable physical complex with the E5 protein.

To determine the biological activity of the I514T and T513L/I514T mutants, IL-3 independence assays were performed. Ba/F3 cell lines co-expressing either mutant receptor with or without E5 or v-sis were incubated in media lacking IL-3 and monitored for growth. The graphs illustrated in Figs. 5C and 6B depict cell densities after culturing in the absence of IL-3 for 10 and 13 days, respectively. Ba/F3 cells co-expressing I514T or T513L/I514T with v-sis, like cells expressing the wild type receptor with v-sis, proliferated in the absence of IL-3, suggesting that neither the I514T substitution nor altering the location of the transmembrane Thr residue affects the ability of this receptor to functionally cooperate with ligand. Like the wild type receptor, the I514T mutant, when co-expressed with E5, conferred an IL-3-independent growth phenotype (Fig. 6B). In contrast, the T513L/I514T receptor mutant when expressed with E5 (Fig. 5C) was unable to induce IL-3-independent proliferation. Thus, the positional change of the transmembrane Thr in the T513L/I514T mutant, and not the I546T substitution by itself, is inhibitory for functional cooperation with the E5 protein. Taken together, these data demonstrate that the location of the Thr in the transmembrane domain of the PDGFR plays a significant role in a productive interaction with the E5 protein.

Ile⁵⁰⁶ in the PDGFR Is Required for a Productive Interaction with the E5 Protein-- Because the required Lys⁴⁹⁹ and Thr⁵¹³ are predicted to align along the same face of the PDGFR transmembrane -helix, additional amino acids located along this face of the -helix such as Ile at position 506 also may be required for stable protein-protein contacts with the E5 protein. To address this hypothesis, Ile⁵⁰⁶ was replaced with an alanine (Ala) and the resulting mutant, I506A (Fig. 1B), was tested for its ability to interact with the E5 protein in Ba/F3 cells.

First, cell lines expressing the wild type receptor or the I506A mutant receptor with or without E5 or v-sis were analyzed for receptor expression levels and receptor tyrosine phosphorylation (Fig. 7A). PDGFR was immunoprecipitated from cell lysates and subjected to SDS-PAGE followed by PDGFR or phosphotyrosine immunoblotting. As shown in Fig. 7A (upper panel), when expressed with v-sis the I506A mutant receptor exhibited a similar level of tyrosine phosphorylation as the wild type receptor with v-sis, suggesting that the I506A substitution does not effect ligand-induced receptor activation. Once again, when co-expressed with E5, the wild type receptor displayed high levels of tyrosine phosphorylation, indicative of its activation by E5. In stark contrast, the I506A receptor co-expressed with E5 exhibited no detectable tyrosine phosphorylation. This did not reflect a lack of I506A receptor expression, because the levels of the mutant and wild type receptor were comparable in the cell lines tested (Fig. 7A, lower panel). Therefore, these results indicate that the I506A mutant receptor could not be activated in response to E5.

View larger version (41K): [in this window] [in a new window]   Fig. 7.   Biochemical and functional analysis of the I506A PDGFR mutant. Ba/F3 cell lines stably expressing the wild type (WT) mouse PDGFR, the I506A mutant receptor, or no exogenous receptor (LXSN) in the presence (+) or absence () of E5 or v-sis were established as described under "Experimental Procedures." A, PDGFR (PR) was immunoprecipitated from cell extracts and then subjected to SDS-PAGE followed by immunoblot analysis using an anti-phosphotyrosine (PY) antibody for assessment of receptor activation (upper panel), or immunoblotted with anti-PDGFR antiserum for determination of the receptor expression levels (lower panel). The upper left phosphotyrosine immunoblot was stripped and re-probed with PDGFR antiserum and is shown at the lower left. Each lane represents 300 µg of extracted protein. Arrows on the right point to the mature (m) and precursor (p) forms of the PDGFR. B, IL-3-dependent growth of Ba/F3 cell lines expressing the I506A receptor mutant with E5. Ba/F3 cells expressing the indicated PDGFR construct without (Puro) or with E5 or v-sis were seeded into medium lacking IL-3 at a density of 5 × 105 cells per 10 ml and then counted 9 days later. The graph shown is representative of three sets of independently derived cell lines.

To determine whether the inability of the I506A receptor mutant to be activated by E5 translates to a defective proliferative response to E5, an IL-3 independence assay was performed. The graph in Fig. 7B depicts the density of cells after a 9-day incubation in the absence of IL-3. Expression of the wild type receptor or the I506A mutant alone was not capable of conferring IL-3-independent growth on Ba/F3 cells. When expressed with v-sis, the I506A mutant receptor was capable of stimulating IL-3-independent proliferation to a similar cell density as the wild type receptor expressed with v-sis, illustrating that the I506A substitution did not alter normal receptor signaling in response to ligand. However, unlike the wild type receptor, the I506A mutant receptor when expressed with E5 was unable to induce proliferation in the absence of IL-3. This indicates that this receptor mutant was unable to interact with the E5 protein in a functionally productive manner. In summary, these data provide evidence that Ile⁵⁰⁶ is an additional amino acid within the PDGFR transmembrane domain necessary for an optimal and completely productive interaction with the E5 protein.

Leu⁵²⁰ in the PDGFR Is Required for a Fully Productive Interaction with the E5 Protein-- Lys⁴⁹⁹, Ile⁵⁰⁶, and Thr⁵¹³, the three PDGFR amino acids thus far shown to be required for a productive interaction with the E5 protein, all are predicted to align along a single face of the -helix of the receptor transmembrane domain. This supports the hypothesis that the transmembrane domain of the receptor adopts a structure that enables multiple contacts with the E5 protein along a single face of the -helix. To test this hypothesis further we examined the role of Leu⁵²⁰ of the receptor in the interaction because it is also predicted to align along this "active" -helical face (Fig. 9). Site-directed mutagenesis was utilized to replace Leu⁵²⁰ with either an Ala or an Ile, and the resultant receptor mutants L520A and L520I (Fig. 1B), respectively, were analyzed for their ability to become activated and induce proliferation in response to E5 in Ba/F3 cells.

Receptor activation was determined by phosphotyrosine immunoblot analysis of PDGFR immunoprecipitates from extracts of Ba/F3 cells expressing the wild type PDGFR, L520A, or L520I receptor mutants with or without E5 or v-sis. As shown in Fig. 8A, substantial tyrosine phosphorylation was detectable for each receptor when co-expressed with v-sis, but not when expressed alone. This result suggests that these amino acid substitutions at position 520 neither induce a state of constitutive activation nor inhibit receptor activation in response to ligand. When co-expressed with E5 the L520I receptor mutant displayed an abundant level of tyrosine phosphorylation, which appeared even greater than that of the wild type receptor expressed with E5. In addition, the L520A receptor mutant was tyrosine phosphorylated to a similar extent as the wild type receptor in response to E5. (Although tyrosine phosphorylation of the precursor form of this mutant was somewhat diminished.) Any induction of receptor tyrosine phosphorylation that was observed could not be explained by a corresponding increase in total receptor levels (Fig. 8A, lower panel). Thus, neither the L520A nor the L520I substitution appeared to significantly inhibit the ability of the receptor to be activated by the E5 protein.

View larger version (35K): [in this window] [in a new window]   Fig. 8.   Biochemical and functional analysis of the L520A and L520I PDGFR mutants. Ba/F3 cell lines stably expressing the wild type (WT) mouse PDGFR, or the L520A or L520I mutant receptor in the presence or absence () of E5 or v-sis were established as described under "Experimental Procedures." A, PDGFR (PR) was immunoprecipitated from cell extracts and then subjected to SDS-PAGE followed by phosphotyrosine (PY; upper panel) or PDGFR (PR; lower panel) immunoblotting for assessment of receptor activation or expression levels, respectively. Each lane represents ~600 (upper panel) or 200 µg (lower panel) of extracted protein. Arrows on the right point to the mature (m) and precursor (p) forms of the PDGFR. B, IL-3 independence assay of Ba/F3 cell lines expressing the L520A or L520I receptor mutants. Ba/F3 cells expressing the indicated receptor construct without (Puro) or with E5 or v-sis were seeded into medium lacking IL-3 at a density of 5 × 105 cells/10 ml and then counted 9 days later. The graph shown is representative of four different sets of independently derived cell lines.

To examine the ability of the L520I and L520A mutants to functionally cooperate with E5, an IL-3 independence assay was performed. Stable Ba/F3 cell lines expressing the wild type receptor, L520I, or L520A alone or with E5 or v-sis were incubated in the absence of IL-3 and monitored for growth. Fig. 8B is a representative graph showing cell densities after a 9-day incubation in the absence of IL-3. As expected, cell lines expressing each receptor in the absence of E5 or v-sis remained IL-3 dependent for growth. Ba/F3 cells co-expressing the L520I mutant with E5 proliferated in the absence of IL-3 as well if not better than cells co-expressing the wild type receptor with E5. In contrast, the L520A receptor mutant when expressed with E5 only inefficiently permitted IL-3-independent growth. Although cells of this genotype eventually grew in the absence of IL-3 (data not shown), this growth was substantially delayed (typically by 7 days or more) compared with cells expressing the wild type receptor and E5. Thus, the L520A, but not the L520I, substitution appears to be inhibitory for functionally cooperating with the E5 protein. This was not because of a general inability of the L520A mutant to induce a proliferative signal, as it was able to induce IL-3-independent growth as efficiently as the wild type receptor when expressed with v-sis. This suggests that Leu⁵²⁰ in the transmembrane domain of the PDGFR is important for a fully productive interaction with the E5 protein.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The data presented here expands the current understanding of protein-protein interactions between the BPV E5 protein and the PDGFR. Previous mutagenesis studies showed that the juxtamembrane Lys⁴⁹⁹ and the transmembrane Thr⁵¹³ in the PDGFR and the analogously positioned Asp³³ and Gln¹⁷, respectively, in the E5 protein play an integral role in complex formation (23, 27, 28, 30). Because of the nature of these specific amino acid requirements, it was proposed that the PDGFR and the E5 protein make two contacts: 1) an electrostatic interaction between charged residues at the juxtamembrane position; and 2) hydrogen bonding between polar residues at a central transmembrane position. Furthermore, it was speculated that the PDGFR contacts the E5 protein via a single transmembrane -helical interface (23). Our data solidifies this model and suggests that more than two receptor amino acids may be involved in making contacts with the E5 protein along this active face of the transmembrane alpha helix.

We first provide evidence that the transmembrane region of the receptor is sufficient for complex formation with the E5 protein. A dually truncated receptor construct with large portions of the intracellular and extracellular domains deleted was still able to physically complex with the E5 protein. Besides the signal peptide sequence and the transmembrane domain, this truncated receptor is predicted to contain only five extracellular amino acids (residues 1-4 linked to Lys⁴⁹⁹) as well as 17 juxtamembrane cytoplasmic amino acids linked to the COOH-terminal epitope. The 17 juxtamembrane cytoplasmic amino acids are not likely to play a role in the interaction because there are only 3 amino acids in the E5 protein that are predicted to be intracellular and these are not essential for its transforming activity (25). Also, the COOH-terminal epitope of the receptor is not likely to be involved in the interaction because receptor antibodies raised against this epitope can co-immunoprecipitate E5-PDGFR complexes (10). Thus, we conclude that the receptor transmembrane domain with the juxtamembrane Lys⁴⁹⁹ is the minimal contiguous region required for an interaction with the E5 protein.

It was previously suggested that the transmembrane Thr⁵¹³ of the PDGFR interacts with the required Gln¹⁷ in the transmembrane domain of the E5 protein via hydrogen bonding. In support of this hypothesis, extensive mutagenesis of Gln¹⁷ in E5 revealed that only amino acids capable of hydrogen bond formation (typically polar in nature) could maintain the transforming activity of E5 and permit an interaction with and activation of the PDGFR (30). In this paper we tested the tolerability of replacing Thr⁵¹³ with two different polar amino acids, Ser and Asn. We found that Asn as well as Ser could functionally replace Thr at position 513 for a fully productive interaction with the E5 protein. Because the side chain of Asn does not contain a hydroxyl group, these results suggest that it is the overall polar nature of the Thr side chain rather than its hydroxyl group that allows for an interaction with the E5 protein. This is consistent with the hypothesis that Thr⁵¹³ contacts Gln¹⁷ in the E5 protein via hydrogen bonding.

Further analysis of the nature of the Thr⁵¹³ requirement revealed the importance of its location within the transmembrane domain for maintenance of a productive interaction with E5. We found that changing the position of the Thr to the carboxyl adjacent site completely abrogated the interaction. Because this change in location effectively shifted the Thr to an adjacent face of the proposed transmembrane -helix (Fig. 9), the proper position of the Thr along this -helix is strictly required for the interaction. This is consistent with the model that the positioning of the required Lys⁴⁹⁹ and Thr⁵¹³ along the same face of the transmembrane -helix allows both residues to simultaneously contact the E5 protein along a single -helical interface (23) (Fig. 9). In this case, altering the location of the Thr would be detrimental to the interaction if the -helical structure of the receptor transmembrane domain is not flexible enough to spatially compensate for relocation of this residue.

View larger version (22K): [in this window] [in a new window]   Fig. 9.   Helical wheel representation of the PDGFR transmembrane domain. The helical wheel diagram of the murine PDGFR transmembrane domain is shown for an -helix in a left-handed coiled coil. Note that the required Lys499, Thr513, Ile506, and Leu520 (boxed residues) would all align along the same face of the -helix if in a left-handed coiled coil complex with the transmembrane domain of another protein. Adapted from Petti et al. (23).

Our results also suggest that besides Lys⁴⁹⁹ and Thr⁵¹³, Ile⁵⁰⁶ and Leu⁵²⁰ are additional receptor transmembrane amino acids required for an optimal interaction with the E5 protein. Interestingly, Ile⁵⁰⁶ and Leu⁵²⁰ are predicted to be located on the same face of the transmembrane -helix as Lys⁴⁹⁹ and Thr⁵¹³ when in a left-handed coiled coil complex (Fig. 9). This lends further support to the model that a single interface of the receptor transmembrane -helix interacts with the E5 protein and suggests that more than two sites of contact are involved. It has been shown for glycophorin A that multiple transmembrane amino acids spaced at a regular interval play a role in its dimerization via protein-protein interactions between transmembrane domains (39-44). Specifically, every 7th amino acid residue within the glycophorin A transmembrane domain is required for dimerization of this protein (42, 44). Because 7 amino acids correspond to two turns of a canonical -helix having 3.6 amino acids per turn, the amino acids required for dimerization of glycophorin A were predicted to be on the same face of the transmembrane -helix. Similarly, the required Lys⁴⁹⁹, Ile⁵⁰⁶, Thr⁵¹³, and Leu⁵²⁰ of the PDGFR may be on the same face of the transmembrane -helix because they are also spaced seven residues apart and may be situated at every second turn of the helix.

Although a L520A substitution did not appear to inhibit E5-induced receptor tyrosine phosphorylation, it did inhibit the ability of the receptor to stimulate a proliferative signal in response to E5. Therefore, it is likely that the mutation somehow inhibited the functioning of the E5-PDGFR complex without inhibiting a physical association with the E5 protein. Because this mutation did not inhibit the ability of the receptor to cooperate with v-sis, it was specifically detrimental to the quality of the interaction with the E5 protein rather than normal receptor signaling. Thus, it is possible that Leu⁵²⁰ plays a role in stabilizing the proper conformation of the E5-induced receptor dimer for generating a proliferative signal. Recently, Bell et al. (45) provided evidence that the orientation of the transmembrane dimer interface between two Neu receptor molecules can dictate the orientation of cytoplasmic kinase domains and affect receptor kinase activity. Similarly, the E5 dimer most likely links two receptor molecules together in a particular conformation that is conductive for transmitting a transforming signal. The inhibitory effect of an Ala but not an Ile at position 520 could be explained if a hydrophobic side chain at this position is required for maintaining a functional receptor conformation. For example, the side chain of Leu⁵²⁰ could stabilize the conformation of the receptor by weakly interacting with the side chain of Leu¹⁰ at the corresponding position in the E5 protein. This is consistent with evidence that Leu¹⁰ of E5 is important for complex formation with the PDGFR (46). Alternatively, the methyl group of the Leu⁵²⁰ side chain may be involved in hydrophobic contacts with phospholipids within the cell membrane and thus stabilize the transmembrane -helix. Several studies using in vitro membrane systems have shown that protein-phospholipid interactions play a role in maintaining helical stability and flexibility of the transmembrane domain of the protein as well as membrane lipid packing (47-50).

Unlike a L520A substitution, an I506A replacement obliterated an interaction with E5, as indicated by complete inhibition of E5-induced receptor activation and mitogenic activity. Others have found that an I506V substitution was inhibitory for a functional interaction but not a physical interaction with the E5 protein.² Thus, a hydrophobic side chain containing two methyl groups, such as that in Ile and Val, may be required for a physical interaction with the E5 protein. However, the longer side chain of Ile compared with Val may be necessary for an intermolecular contact that holds the receptor in a conformation suitable for generating a proliferative signal.

It is possible that the hydrophobic side chain of Ile⁵⁰⁶ interacts directly with the side chains of Leu²⁴, which is located at the corresponding transmembrane position in the E5 protein. This hypothesis is substantiated by a recent study, which determined that mutating several Leu residues within E5, including Leu²⁴, to Ala inhibited complex formation with the PDGFR (46). Furthermore, Mattoon and DiMaio (51) reported that Leu²⁴ of the E5 protein may be located on the same dimer interface as the required Gln¹⁷ and thus may be positioned for an interaction with the receptor in a similar manner as Gln¹⁷. Interestingly, Leu²⁴ is located in a Leu-rich region of the E5 transmembrane domain (residues 18-26) in which 8 of the 9 residues are leucines. Thus, it is possible that one of these Leu residues and Ile⁵⁰⁶ of the receptor can participate in hydrophobic interactions that resemble interactions between leucine zipper proteins. Evidence for leucine/isoleucine zipper interactions occurring between transmembrane domains has been implicated in pentamer formation of the 52-amino acid transmembrane phosphoprotein phospholamban B (52-54).

In summary, we have presented compelling data in support of the model that the transmembrane -helix of the PDGFR makes multiple direct contacts with the E5 transmembrane domain along a single interface. Further investigation is required to determine whether the required receptor amino acids actually participate in direct intermolecular contacts or provide structural stability to the complex. Moreover, characterizing the structure of the E5-PDGFR complex should lead to a better understanding of what constitutes an active dimer conformation of the receptor. Thus, the E5/PDGFR interaction provides an interesting and useful model for studying structure-function relationships in receptor tyrosine kinase signaling.

	ACKNOWLEDGEMENTS

We thank Paul Black, Daniel DiMaio, John Lehman, Jeff Banas, Tom Friedrich, and Jim McSharry for useful discussions. We also thank Dawn Mattoon for confirming the sequence of our truncated receptor and Daniel DiMaio for useful reagents.

	FOOTNOTES

^* This work was supported by National Institutes of Health Grant CA73682.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

To whom correspondence should be addressed. Tel.: 518-262-6285; Fax: 518-262-5748; E-mail: pettil@mail.amc.edu.

Published, JBC Papers in Press, September 25, 2002, DOI 10.1074/jbc.M209582200

² D. Mattoon and D. DiMaio, personal communication.

	ABBREVIATIONS

The abbreviations used are: BPV, bovine papillomavirus type 1; PDGF, platelet-derived growth factor; PDGFR, PDGF  receptor; IL-3, interleukin-3; TBS, Tris-buffered saline; WGL, wheat germ lectin.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				L. M. Petti, E. C. Ricciardi, H. J. Page, and K. A. Porter Transforming signals resulting from sustained activation of the PDGF{beta} receptor in mortal human fibroblasts J. Cell Sci., April 15, 2008; 121(8): 1172 - 1182. [Abstract] [Full Text] [PDF]

				C.-C. Lai, A. P. B. Edwards, and D. DiMaio Productive Interaction between Transmembrane Mutants of the Bovine Papillomavirus E5 Protein and the Platelet-Derived Growth Factor {beta} Receptor J. Virol., February 1, 2005; 79(3): 1924 - 1929. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 277/49/47149    most recent M209582200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Nappi, V. M. Articles by Petti, L. M. Search for Related Content PubMed PubMed Citation Articles by Nappi, V. M. Articles by Petti, L. M. Social Bookmarking                      What's this?