Human platelet-derived growth factor (PDGF) is a connective tissue cell mitogen comprising two related chains encoded by distinct genes. The B chain is the homolog of the v-sis oncogene product. Properties that distinguish these ligands include greater transforming potency of the B chain and more efficient secretion of the A chain. By a strategy involving the generation of PDGF A and B chimeras, these properties were mapped to distinct domains of the respective molecules. Increased transforming efficiency segregated with the ability to activate both alpha and beta PDGF receptors. These findings genetically map PDGF B residues 105 to 144 as responsible for conformational alterations critical to beta PDGF receptor interaction, and provide a mechanistic basis for the greater transforming potency of the PDGF B chain.

Human platelet-derived growth factor (PDGF) is a major mitogen for cells of connective tissue origin that is involved in development and wound healing (Ross et al. 1986). Abnormal expression of this growth factor has also been implicated in a variety of pathologic states including cancer (Eva et al. 1982; Gazit et al. 1984). PDGF is a disulfide linked dimer consisting of two related polypeptide chains, designated A and B, that are products of different genes. The gene encoding the human PDGF B chain is the normal counterpart of the v-sis oncogene (Waterfield et al. 1983; Doolittle, 1983; Devare et al. 1983). PDGF A and B chains are approximately 40% related (Betsholtz et al. 1986) and contain eight conserved cysteine residues (Giese et al. 1987; Sauer and Donoghue, 1988). PDGF A and B chains can form homodimers as well as the AB heterodimer, and there is evidence for the natural occurrence of all three isoforms (Johnsson et al. 1982).

Although homodimers of either PDGF A or B are mitogenic as well as chemotactic for cells possessing the appropriate PDGF receptor (Matsui et al. 19896), major differences in their biological properties have been observed. The PDGF B chain gene exhibits 10 to 100 fold greater transforming efficiency in the NIH/3T3 transfection assay (Beckman et al. 1988). Moreover, its product remains tightly cell associated (Robbins et al. 1985), whereas the PDGF A chain is efficiently secreted (Beckman et al. 1988). In addition, the two molecules differentially bind and activate the products of two distinct genes, encoding respectively the alpha and beta PDGF receptor (Matsui et al. 1989a; Yarden et al. 1986; Claesson-Welsh et al. 1989). While PDGF B interacts with either receptor, PDGF A binds and triggers only the alpha PDGF receptor (Matsui et al. 1989a,b;Hart et al. 1988; Heldin et al. 1988). In the present study, we constructed chimeras of PDGF A and B chains in an effort to map domains of each that potentially influence their normal functions and role in pathologic processes.

Strategy for construction of PDGF chimeric molecules

We initially constructed ten chimeric PDGF molecules to investigate structural regions of PDGF A or PDGF B associated with specific PDGF functions (LaRochelle et al. 1990). The chimeric constructs were developed utilizing preexisting, or engineered, common restriction endonuclease sites within the PDGF A- or PDGF B-coding sequences. All PDGF A- or PDGF B-coding sequences altered for this purpose by oligonucleotide-mediated, site-directed mutagenesis (Kunkel, 1985) were first shown to possess biological activities indistinguishable from those of their respective parental cDNAs. Each chimera was designated on the basis of the codon at which the recombination was performed. Four of the chimeric constructs, A97B99, B98A98, A177B179, and B178A178, were designed to maintain the functional integrity of the PDGF B minimal transforming domain (King et al. 1985; Hannink et al. 1986) or the analogous region of PDGF A (Fig. 1). An additional six constructs further dissected the minimal transforming domain. The A143B145 and B144A144 chimeras divided the transforming region roughly in half, while A104B106 and B105A105 chimeras as well as A153B155 and B154A154 chimeras further subdivided the minimal transforming domains (Fig. 1).

Fig. 1.

Transforming activity of PDGF chimeric constructs. Chimeric PDGF molecules were constructed by recombination of the PDGF A (hatched box) and B (open box) genes at common pre-existing or engineered (Kunkel, 1985) restriction endonuclease sites. The structure of each recombinant was verified by a combination of restriction endonuclease mapping and nucleotide sequence determination. NIH/3T3 cells were transfected with the recombinant plasmid DNA and 40, ug of carrier calf thymus DNA by the calcium phosphate precipitation technique (Wigler et al. 1977). Transfected cultures were either scored for colony formation in the presence of G418 (Southern and Berg, 1982) or for focus-forming activity as described (Giese et al. 1987; Beckman et al. 1988). Data shown represent the mean values of three experiments. Relative transforming efficiency was calculated by dividing the number of foci (FFU) by the number of colonies (CFU) per ng of DNA relative to that of PDGF A. Cysteine residues essential (C) or non-essential (c) for PDGF B transformation are shown.

Fig. 1.

Transforming activity of PDGF chimeric constructs. Chimeric PDGF molecules were constructed by recombination of the PDGF A (hatched box) and B (open box) genes at common pre-existing or engineered (Kunkel, 1985) restriction endonuclease sites. The structure of each recombinant was verified by a combination of restriction endonuclease mapping and nucleotide sequence determination. NIH/3T3 cells were transfected with the recombinant plasmid DNA and 40, ug of carrier calf thymus DNA by the calcium phosphate precipitation technique (Wigler et al. 1977). Transfected cultures were either scored for colony formation in the presence of G418 (Southern and Berg, 1982) or for focus-forming activity as described (Giese et al. 1987; Beckman et al. 1988). Data shown represent the mean values of three experiments. Relative transforming efficiency was calculated by dividing the number of foci (FFU) by the number of colonies (CFU) per ng of DNA relative to that of PDGF A. Cysteine residues essential (C) or non-essential (c) for PDGF B transformation are shown.

Mapping of a PDGF B sub-domain responsible for its potent transforming activity

All wild type parental and recombinant PDGF constructs (LaRochelle et al. 1990) were transferred into a vector containing the metallothionein promoter (MMTneo) and analyzed for transforming activity by transfection of NIH/3T3 cells. Since the MMTneo vector also contained a dominant, selectable neomycin marker gene, it was possible to score neomycin-resistant colony formation for each plasmid as well. Thus, we were able to compare precisely the specific transforming efficiencies of each construct.

As shown in Fig. 1, the PDGF B expression vector showed around twenty-fold higher transforming efficiency than that of PDGF A, as previously reported (Beckman et al. 1988). A97B99 and B178A178 chimeric constructs, which both contained the minimal transforming domain of the PDGF B gene product, demonstrated high transforming efficiency, indistinguishable from that of the wild type PDGF B construct. In contrast, B98A98 and A17’B179 chimeric constructs, which possessed the analogous domain of PDGF A, showed ten to twenty fold lower specific transforming efficiency, equivalent to that of PDGF A (Fig. 1). Chimeras A104B106, B144A144 and B154A154 also possessed high specific transforming efficiency, whereas the reciprocal chimeras B105A105, A143B145 and A153B155 respectively, were only weakly transforming. All of these findings suggested that amino acid residues 105-144 of PDGF B were responsible for its more potent transforming activity (Fig. 1).

To test this hypothesis, we substituted only the minimal regions mapped above that were responsible for differences in transforming activities of the native PDGF A and B molecules. As shown in Fig. 1, the AB105-144 chimera possessed high specific transforming activity, indistinguishable from that of PDGF B. Conversely, substitution of PDGF A codons 104–144 for those of PDGF B reduced transforming activity of the resulting chimera to that of the PDGF A molecule. All of these results conclusively demonstrated that the subdomain encompassed by PDGF B amino acid residues 105–144 was responsible for its more potent transforming properties.

Immunochemical characterization of PDGF chimeric proteins

We next sought to verify the chimeric constructs by analysis of their translational products. Thus, each NIH/3T3 transfectant was metabolically labeled and subjected to immunoprecipitation analysis using antisera specific to PDGF A or PDGF B amino or carboxy termini, respectively. As shown in Fig. 2A, the primary PDGF A translational product, p42, as well as its 38×103 and 32×103Mr processed forms, were detected in lysates of PDGF A transfected cells. Similarly, the primary PDGF B translational product, p54, was processed at amino and carboxy termini to 40×103Mr, 34×103Mr and 24×103Mr species (Fig. 2B).

Fig. 2.

Immunochemical characterization of PDGF chimeric proteins. Mass populations of 107 cells transfected with the MMT neo vector containing either PDGF A (panel A), PDGF B (Panel B), PDGF A143B14S (panel C), or PDGF B144A144 (panel D), were preincubated overnight in DMEM containing 10% calf serum and 25, MM ZnCl2. The medium was then replaced for three hours with cysteine- and methionine-free DMEM containing [35S]methionine and [35S]cysteine at 125, μCi mF1 and 25 μM ZnCl2. Crude membrane fractions were immunoprecipitated with PDGF A amino terminal antibody, PDGF A carboxy terminal antibody, PDGF B amino terminal antibody, PDGF B carboxy terminal antibody, or PDGF antibody. In some cases (1anes 2 and 4), antibodies were preincubated with excess homologous peptide. Immune complexes were analyzed by SDS-PAGE under non-reducing conditions and results were visualized by fluorography for ten days.

Fig. 2.

Immunochemical characterization of PDGF chimeric proteins. Mass populations of 107 cells transfected with the MMT neo vector containing either PDGF A (panel A), PDGF B (Panel B), PDGF A143B14S (panel C), or PDGF B144A144 (panel D), were preincubated overnight in DMEM containing 10% calf serum and 25, MM ZnCl2. The medium was then replaced for three hours with cysteine- and methionine-free DMEM containing [35S]methionine and [35S]cysteine at 125, μCi mF1 and 25 μM ZnCl2. Crude membrane fractions were immunoprecipitated with PDGF A amino terminal antibody, PDGF A carboxy terminal antibody, PDGF B amino terminal antibody, PDGF B carboxy terminal antibody, or PDGF antibody. In some cases (1anes 2 and 4), antibodies were preincubated with excess homologous peptide. Immune complexes were analyzed by SDS-PAGE under non-reducing conditions and results were visualized by fluorography for ten days.

A representative chimera, A143B145, encoded a p45 primary translational product, and was processed at amino and carboxy termini to 40 and 38×103Mr forms, as well as a major 26×103Mr species (Fig. 2C). The reciprocal chimeric construct directed synthesis of a p46 translational product, which was amino terminally processed to a major p32 species (Fig. 2D). Thus, each of these chimeras possessed the correct amino and carboxy terminal epitopes of their respective parental molecules and was of an intermediate molecular weight relative to PDGF A or PDGF B homodimers. Immunoprecipitation analysis of the remaining transfectants revealed the expected PDGF A or PDGF B antigenic determinants and predicted intermediate sizes (data not shown).

Mapping of a domain responsible for differences in PDGF A and B secretion

Previous studies of the compartmentalization of PDGF A and PDGF B in transformed NIH/3T3 fibroblasts have shown that PDGF B remains tightly membrane-associated, whereas PDGF A is efficiently secreted into culture fluids (Beckman et al. 1988). No obvious structural motif that might cause retention of PDGF B, such as hydrophobic stretches, has been observed. Thus, we sought to identify the domain(s) responsible for differences in secretion of the two molecules and whether such a domain could account for their different transforming potencies.

Following metabolic radiolabeling of cultures for 4h, medium conditioned by each transfectant, as well as the crude cell membrane fraction of each, were subjected to immunoprecipitation analysis with a panel of PDGF antibodies. Crude membrane preparations of each transfectant showed roughly comparable levels of PDGF immunoreactive protein. However, only those chimeras which contained PDGF A carboxy terminal amino acid residues 178-211, namely B98A98, B105A105, B144A144, B154A154 and B178A178, were found to be efficiently secreted. Fig. 3 shows that the B178A178 chimera, which contained only PDGF A amino acid residues 178–211, was efficiently released, whereas the reciprocal chimera, A177B179, remained more than 90% membrane associated.

Fig. 3.

Compartmentalization of PDGF chimeric constructs. NIH/3T3 cells transfected with A177B179 or B178A178 were metabolically labeled as described in the text. Crude cellular membrane or conditioned media were examined by immunoprecipitation using the antibodies indicated. In some cases, antibodies were preincubated with excess homologous peptide (1ane 2). Immune complexes were analyzed as indicated above and the results were visualized by fluorography for twelve days (Cell membrane) or 36 h (Conditioned medium).

Fig. 3.

Compartmentalization of PDGF chimeric constructs. NIH/3T3 cells transfected with A177B179 or B178A178 were metabolically labeled as described in the text. Crude cellular membrane or conditioned media were examined by immunoprecipitation using the antibodies indicated. In some cases, antibodies were preincubated with excess homologous peptide (1ane 2). Immune complexes were analyzed as indicated above and the results were visualized by fluorography for twelve days (Cell membrane) or 36 h (Conditioned medium).

To confirm our immunological findings, we analyzed mitogenic activities associated with culture fluids and crude membrane preparations of transfectants containing parental or chimeric PDGF constructs. Comparable mitogenic activity was detected in each crude membrane fraction. However, only in the case of PDGF A and those chimeras containing at least the carboxy terminal thirty-four amino acid residues of PDGF A, was mitogenic activity detectable in culture fluids (data not shown). In each case, the mitogenic activity was specifically inhibited by neutralizing PDGF antibody, establishing the PDGF-related nature of the secreted mitogen. As shown above, potent transforming activity mapped to PDGF B amino acid residues 105–144 (Fig. 1). Thus, localization of the domain responsible for differences in PDGF A and B secretory properties to their carboxy terminal regions excluded this property from being responsible for their different transforming activities.

Activation of alpha and beta PDGF receptors in NIH/3T3 cells expressing PDGF chimeras

We next investigated whether PDGF receptor binding and/or activation might be responsible for the differences in oncogenic potency of PDGF A and B. Thus, we examined the steady state level of tyrosine phosphorylation of alpha and beta receptors expressed in NIH/3T3 transfectants containing either wild type PDGF A or B constructs, as well as each of the chimeras. To do so, cell lysates were enriched for each receptor by immunoprecipitation with alpha or beta PDGF receptor-specific peptide antisera, followed by immunoblotting with anti-phosphotyrosine antibody. As shown in Fig. 4, NIH/3T3 cells showed no detectable alpha or beta PDGF receptor tyrosine phosphorylation. As expected from known receptor binding properties of each ligand (Matsui et al. 1989a,b;Hart et al. 1988; Heldin et al. 1988), NIH/3T3 cells expressing PDGF A demonstrated tyrosine phosphorylated 180×103Mr alpha but not beta receptor species. In contrast, both 180×103Mr alpha and beta receptor species were tyrosine phosphorylated in PDGF B-producing cells (Fig. 4). The specificity of the antibody was demonstrated by the ability of phosphotyrosine to compete for immunodetection of these proteins (data not shown).

Fig. 4.

Tyrosine phosphorylation of alpha and beta PDGF receptors in NIH/3T3 cells expressing PDGF chimeras. NIH/3T3 cells (1ane 1) or transfectants expressing PDGF A (1ane 2), PDGF B (1ane 3), or the chimeric PDGF constructs (1ane 4-15) were incubated overnight in DMEM, 25 μM ZnCl2. After 16 h, the cells were washed with PBS/1.0mM sodium orthovanadate and lysed as described by Matsui et al. (1989a,b). Protein extracts were immunoprecipitated with anti-peptide antibodies specific for the alpha (panel A) or beta (panel B) PDGF receptors (Matsui et al. 1989a,6). Immunoprecipitated proteins were blotted to Immobilon-P and probed with anti-phosphotyrosine specific antibody. Filters were treated with 125I-labeled protein A and subjected to autoradiography. The electrophoretic mobility of the ppl80 alpha or beta PDGF receptors are shown. In some cases immature forms of PDGF receptors were recognized as well.

Fig. 4.

Tyrosine phosphorylation of alpha and beta PDGF receptors in NIH/3T3 cells expressing PDGF chimeras. NIH/3T3 cells (1ane 1) or transfectants expressing PDGF A (1ane 2), PDGF B (1ane 3), or the chimeric PDGF constructs (1ane 4-15) were incubated overnight in DMEM, 25 μM ZnCl2. After 16 h, the cells were washed with PBS/1.0mM sodium orthovanadate and lysed as described by Matsui et al. (1989a,b). Protein extracts were immunoprecipitated with anti-peptide antibodies specific for the alpha (panel A) or beta (panel B) PDGF receptors (Matsui et al. 1989a,6). Immunoprecipitated proteins were blotted to Immobilon-P and probed with anti-phosphotyrosine specific antibody. Filters were treated with 125I-labeled protein A and subjected to autoradiography. The electrophoretic mobility of the ppl80 alpha or beta PDGF receptors are shown. In some cases immature forms of PDGF receptors were recognized as well.

Fig. 5.

Summary of biological properties mapped by PDGF chimeras. PDGF A (hatched box) and PDGF B (open box) cDNAs are represented. Cysteine residues essential (C) or nonessential (c) for PDGF B transformation are shown. Percent secretion was determined by comparing the quantity of PDGF protein immunologically detected in cell membrane preparations with that secreted into the culture fluid. Autophosphorylation was utilized as a measure of receptor activation. Relative transforming efficiency is defined in Fig. 1.

Fig. 5.

Summary of biological properties mapped by PDGF chimeras. PDGF A (hatched box) and PDGF B (open box) cDNAs are represented. Cysteine residues essential (C) or nonessential (c) for PDGF B transformation are shown. Percent secretion was determined by comparing the quantity of PDGF protein immunologically detected in cell membrane preparations with that secreted into the culture fluid. Autophosphorylation was utilized as a measure of receptor activation. Relative transforming efficiency is defined in Fig. 1.

When the steady state level of PDGF receptor tyrosine phosphorylation was examined in transfectants containing the PDGF chimeras, readily detectable levels of the activated 180×103Mr alpha PDGF receptor were observed in each case (Fig. 4). However, there was chronic activation of the 180 × 103Mr beta PDGF receptor species of cells expressing A9’B99, A104B106, B144A144 and B154A154, as well as B178A178 chimeras, all of which contained at least PDGF B amino acid residues 105–144. As shown in Fig. 4, the AB105-144 chimera which substituted only PDGF B amino acid residues 105–144 into the analogous region of PDGF A, demonstrated the same pattern of receptor tyrosine phosphorylation as observed with PDGF B. Conversely, the switch of analogous PDGF A amino acid residues into PDGF B led to a pattern of receptor tyrosine phosphorylation indistinguishable from that of PDGF A. These results demonstrate that amino acid residues 105-144 are responsible for the ability of PDGF B to bind and activate preferentially the beta PDGF receptor. Since increased transforming activity of PDGF B molecules mapped to these same residues, we conclude that the ability to activate beta receptors in addition to alpha receptors provides the basis for the greater transforming potency of the PDGF B chain gene for NIH/3T3 cells (LaRochelle et al. 1990).

The minimal transforming domain of PDGF B has been mapped and encompasses 84 amino acids (King et al. 1985; Hannink et al. 1986). Biochemical studies of cysteine residues involved in disulfide linkages (Giese et al. 1987; Vogel and Hoppe, 1989) suggest that intrachain disulfide bonds, essential to the transforming function of PDGF B, fold this minimal transforming domain into two major loops, according to all models which have been proposed (Giese et al. 1987 ; Vogel and Hoppe, 1989). PDGF A shares functional and structural similarities with PDGF B, including the ability to activate the PDGF alpha receptor (Matsui et al. 1989a,b;Hart et al. 1988; Heldin et al. 1988) as well as conservation of all cysteine residues and spacing between these residues. Less is known about the structure and disulfide linkages of PDGF A. In the present studies, chimera PDGF molecules were generated between PDGF A and B. The analysis of the biological activity of each of the twelve reciprocal chimeras generated argues strongly that very similar disulfide linkages must be formed by the PDGF A chain to allow for such interchangeability of portions of these two related molecules.

Analysis of transforming potency of PDGF B single codon deletion mutants (Giese et al. 1990) has localized a domain which is much more sensitive to mutagenesis than other domains. Moreover, monoclonal antibody sis 1 neutralization of PDGF B (LaRochelle et al. 1989) complimented by epitope mapping (Giese et al. 1990; LaRochelle et al. 1989), suggests that this same region contains a surface epitope that is critical for PDGF receptor binding and activation. Analysis of these PDGF chimeras directly demonstrated that amino acid residues 105-144 of PDGF B, within this same critical domain, are responsible for the more potent transforming properties of the PDGF B chains. PDGF A activated only alpha PDGF receptors whereas PDGF B triggered both alpha and beta PDGF receptors in an autocrine fashion, and the differences in transforming potency of these molecules directly correlated with those constructs containing PDGF B codons 105-144 and their activation of beta PDGF receptors. Thus, this domain is not only critical for ligand-receptor interaction but is the major determinant of the subtle amino acid changes that must determine differences in the ability of the PDGF molecule to interact with other receptors. Moreover, the direct correlation between the ability to activate beta as well as alpha PDGF receptors with greater transforming potency of the chimeras containing PDGF B codons 105-144, provides a mechanistic basis for greater transforming efficiency of the PDGF B chain.

The quantitative differences in transforming activities of PDGF A and B chain genes for NIH/3T3 cells in vitro correlate with in vivo findings that a PDGF B encoding retrovirus induces fibrosarcomas in nude mice (Pech et al. 1989), while an analogous retrovirus encoding PDGF A has as yet not produced detectable tumors (P. Arnstein and S. A. Aaronson, unpublished observations). Nonetheless, we have shown that the alpha PDGF receptors can couple with mitogenic signalling pathways as efficiently as the beta PDGF receptor when either is independently expressed in a null hematopoietic cell (Matsui et al. 1989a,b). Thus, it is possible that in fibroblasts which express both alpha and beta receptors, the alpha receptor may be somewhat less efficient than the beta receptor in stimulating this pathway. Alternatively, non-saturating ligand concentrations expressed in autocrine transformation of NIH/3T3 must be quantitatively more effective in mitogenic signalling when both receptors are stimulated, as opposed to the alpha receptor alone.

We also mapped an additional domain of the two PDGF molecules that was responsible for the differences in their PDGF secretory properties. This domain localized to the carboxy terminal amino acid residues of PDGF, and was demonstrated not to be a determinant of the differences in transforming potency of the two molecules. These differences in the secondary properties could be explained either by a tendency of the PDGF B carboxy terminus to cause association with the cell membrane or an effect of the analogous domain of the PDGF A to promote secretion. Further studies will be necessary to resolve this question.

There is increasing evidence that functional domains of protein correspond to individual exons (Gilbert, 1978; 1985). In the case of the PDGF A molecule, the signal peptides and proteolytic processing sites correspond to specific exons (Heldin and Westermark, 1989). Amino acid residues 105–144 which specify beta PDGF receptor activation are included exclusively within exon four of PDGF B. In contrast, the secretory properties of the long form of PDGF A or the membrane retention properties of PDGF B correspond to amino acids predominantly encoded by exon 6 of either molecule. Thus, the functional differences which we mapped correspond to individual exons specific for either of the PDGFs.

Comparison of the PDGF B 105–144 region with PDGF A reveals two subdomains (residues 106-114 and 135-143) which contain little amino acid homology, separated by an almost entirely conserved stretch of twenty amino acids (residues 115–134). Thus, combinations of PDGF B amino acid substitutions within these two subdomains of PDGF A will determine which PDGF B amino acids bind and activate the beta PDGF receptor. Amino acids of PDGF B directly involved in beta PDGF receptor binding and activation may then be altered to develop competitive antagonists.

W.J.L. is supported by an American Cancer Society Postdoctoral Fellowship (PF no. 3030). We thank C. Betsholtz, B. Westermark and C.-H. Heldin for providing the PDGF A cDNA.

Beckman
,
M. P.
,
Betsholtz
,
C.
,
Heldin
,
C.-H.
,
Westermark
,
B.
,
Dimarco
,
E.
,
Difiore
,
P. P.
,
Robbins
,
K. C.
and
Aaronson
,
S. A.
(
1988
).
Comparison of the biological properties and transforming potential of human PDGF-A and PDGF-B
.
Science
241
,
1346
1348
.
Betsholtz
,
C.
,
Johnsson
,
A.
,
Heldin
,
C.-H.
,
Westermark
,
B.
,
Lind
,
P.
,
Urdea
,
M. S.
,
Eddy
,
R.
,
Shows
,
T. B.
,
Philipott
,
K.
,
Mellor
,
A. L.
,
Knott
,
T. J.
and
Scott
,
J.
(
1986
).
cDNA sequence and chromosomal localization of human platelet-derived growth factor A-chain and its expression in tumor cell lines
.
Nature
320
,
695
699
.
Claesson-Welsh
,
L.
,
Eriksson
,
A.
,
Westermark
,
B.
and
Heldin
,
C.-H.
(
1989
).
cDNA cloning and expression of the A-type platelet-derived growth factor (PDGF) receptor establishes structural similarity to the B-type PDGF receptor
.
Proc. natn. Acad. Sci. U.S.A.
86
,
4917
4921
.
Dev are
,
S.
G.,
Reddy
,
E. P.
,
Law
,
D. J.
,
Robbins
,
K. C.
and
Aaronson
,
S. A.
(
1983
).
Nucleotide sequence of the simian sarcoma virus genome: demonstration that its acquired cellular sequences encode the transforming gene product p28sls
.
Proc. natn‥ Acad. Sci. U.S.A.
80
,
731
735
.
Doolittle
,
R. F.
,
Hunkapiller
,
M. W.
,
Hood
,
L. E.
,
Devare
,
S. G.
,
Robbins
,
K. C.
,
Aaronson
,
S. A.
and
Antoniades
,
H. N.
(
1983
).
Simian sarcoma virus one gene, v-sis, is derived from the gene (or genes) encoding a platelet-derived growth factor
.
Science
221
,
275
277
.
Eva
,
A.
,
Robbins
,
K. C.
,
Andersen
,
P. R.
,
Srinivasan
,
A.
,
Tronick
,
S. R.
,
Reddy
,
E. P.
,
Ellmore
,
N. W.
,
Galen
,
A. T.
,
Lautenberger
,
J. A.
,
Papas
,
T. S.
,
Westin
,
E. H.
,
Wong-Staal
,
F.
,
Gallo
,
R. C.
and
Aaronson
,
S. A.
(
1982
).
Cellular genes analogous to retroviral one genes are transcribed in human tumour cells
.
Nature
295
,
116
119
.
Gazit
,
A.
,
Igarashi
,
H.
,
Chiu
,
I.-M.
,
Srivivasan
,
A.
,
Yaniv
,
A.
,
Tronick
,
S. R.
,
Robbins
,
K. C.
and
Aaronson
,
S. A.
(
1984
).
Expression of the normal human sis/PDGF-2 coding sequence induces cellular transformation
.
Cell
39
,
89
97
.
Giese
,
N.
,
Larochelle
,
W. J.
,
May-Siroff
,
M.
,
Robbins
,
K. C.
and
Aaronson
,
S. A.
(
1990
).
A small v-sis/PDGF B protein domain in which subtle conformational changes abrogate PDGF receptor interaction and transforming activity
.
Molec. cell. Biol.
10
, in press.
Giese
,
N.
,
Robbins
,
K. C.
and
Aaronson
,
S. A.
(
1987
).
The role of individual cysteine residues in the structure and function of the v-sis gene product
.
Science
236
,
1315
1318
.
Gilbert
,
W.
(
1978
).
Why genes in pieces?
Nature
271
,
501
.
Gilbert
,
W.
(
1985
).
Genes in pieces revisited
.
Science
228
,
823
824
.
Hannink
,
M.
,
Sauer
,
M. K.
and
Donoghue
,
D. J.
(
1986
).
Deletions in the C-terminal coding region of the v-sis gene: dimerization is required for transformation
.
Molec. cell. Biol.
6
,
1304
1314
.
Hart
,
C. E.
,
Forstrom
,
J. W.
,
Kelly
,
J. D.
,
Seifert
,
R. A.
,
Smith
,
R. A.
,
Ross
,
R.
,
Murray
,
M. J.
and
Bowen-Pope
,
D. F.
(
1988
).
Binding of different dimeric forms of PDGF to human fibroblasts: evidence for two separate receptor types
.
Science
240
,
1529
1531
.
Heldin
,
C.-H.
,
Backstrom
,
G.
,
Ostman
,
A.
,
Hammacher
,
A.
,
Ronnstrand
,
L.
,
Rubin
,
K.
,
Nister
,
M.
and
Westermark
,
B.
(
1988
).
Binding of different dimeric forms of PDGF to human fibroblasts: evidence for two separate receptor types
.
EMBO J.
7
,
1387
1394
.
Heldin
,
C.-H.
and
Westermark
,
B.
(
1989
).
Platelet-derived growth factor: three isoforms and two receptor types
.
Trends in Genet.
5
,
108
111
.
Johnsson
,
A.
,
Heldin
,
C.-H.
,
Westermark
,
B.
and
Wasteson
,
A.
(
1982
).
Platelet-derived growth factor: identification of constituent polypeptide chains
.
Biochem. biophys. Res. Commun.
104
,
66
74
.
King
,
C. R.
,
Giese
,
N.
,
Robbins
,
K. C.
and
Aaronson
,
S. A.
(
1985
).
In vitro mutagenesis of the v- sis transforming gene defines functional domains of its growth factor-related product
.
Proc, natn. Acad. Sci. U.S.A.
82
,
5295
5299
.
Kunkel
,
T. A.
(
1985
).
Rapid and efficient site-specific mutagenesis without phenotypic selection
.
Proc. natn. Acad. Sci. U.S.A.
82
,
488
492
.
Larochelle
,
W. J.
,
Giese
,
N.
,
May-Siroff
,
M.
,
Robbins
,
K. C.
and
Aaronson
,
S. A.
(
1990
).
Molecular localization of the transforming and secretory properties of PDGF A and PDGF B
.
Science
248
,
1541
1544
.
Larochelle
,
W. J.
,
Robbins
,
K. C.
and
Aaronson
,
S. A.
(
1989
).
Immunochemical localization of the epitope for a monoclonal antibody that neutralizes human platelet-derived growth factor mitogenic activity
.
Molec. cell. Biol.
9
,
3538
3542
.
Matsui
,
T.
,
Heidaran
,
M.
,
Miki
,
T.
,
Popescu
,
N.
,
Larochelle
,
W. J.
,
Kraus
,
M.
,
Pierce
,
J.
and
Aaronson
,
S. A.
(
1989a
).
Isolation of a novel receptor cDNA establishes the existence of two PDGF receptor genes
.
Science
243
,
800
804
.
Matsui
,
T.
,
Pierce
,
J.
,
Fleming
,
T. P.
,
Greenberger
,
J. S.
,
Larochelle
,
W. J.
,
Ruggiero
,
M.
and
Aaronson
,
S. A.
(
1986b
).
Independent expression of human alpha or beta platelet-derived growth factor receptor cDNAs in a naive hematopoietic cell leads to functional coupling with mitogenic and chemotactic signaling pathways
.
Proc. natn. Acad. Sci. U.S.A.
86
,
8314
8318
.
Pech
,
M.
,
Gazit
,
A.
,
Arnstein
,
P.
and
Aaronson
,
S. A.
(
1989
).
Generation of fibrosarcoma in vivo by a retrovirus that expresses the normal B chain of platelet-derived growth factor and mimics the alternative splice pattern of the v-sis oncogene
.
Proc. natn. Acad. Sci. U.S.A.
86
,
2693
2697
.
Robbins
,
K. C.
,
Leal
,
F.
,
Pierce
,
J. H.
and
Aaronson
,
S. A.
(
1985
).
The v-sis/PDGF-2 transforming gene product localizes to cell membranes but is not a secretory protein
.
EMBO J.
4
,
1783
1792
.
Ross
,
R.
,
Raines
,
E. W.
and
Bowen-Pope
,
D. F.
(
1986
).
The biology of platelet-derived growth factor
.
Cell
46
,
155
169
.
Sauer
,
M. K.
and
Donoghue
,
D. J.
(
1988
).
Identification of the nonessential disulfide bonds and altered conformations in the v-sis protein, a homolog of the B chain of platelet-derived growth factor
.
Molec. cell. Biol.
8
,
1011
1018
.
Southern
,
P. J.
and
Berg
,
P.
(
1982
).
Transformation of mammalian cells to antibiotic resistance with a bacterial gene under control of the SV40 early promoter
.
J. Molec. Appl. Genet.
1
,
327
341
.
Vogel
,
S.
and
Hoppe
,
J.
(
1989
).
Binding domains and epitopes in platelet-derived growth factor
.
Biochemistry
28
,
2961
2966
.
Waterfield
,
M. D.
,
Scrace
,
G. T.
,
Whittle
,
N.
,
Stroobant
,
P.
,
Johnsson
,
A.
,
Wasteson
,
A.
,
Westermark
,
B.
,
Heldin
,
C.-H.
,
Huang
,
J. S.
and
Deuel
,
T. F.
(
1983
).
Platelet-derived growth factor is structurally related to the putative transforming protein p28sis of simian sarcoma virus
.
Nature
304
,
35
39
.
Wigler
,
M.
,
Silverstein
,
S.
,
Lee
,
L.-S.
,
Pellicer
,
A.
,
Cheng
,
Y.-C.
and
Axel
,
R.
(
1977
).
Transfer of purified herpes virus thymidine kinase gene to cultured mouse cells
.
Cell
11
,
223
232
.
Yarden
,
Y.
,
Escobedo
,
J. A.
,
Kuang
,
W.-J.
,
Yang-Feng
,
T. L.
,
Daniel
,
T. O.
,
Tremble
,
P. M.
,
Chen
,
E. Y.
,
Ando
,
M. E.
,
Harkins
,
R. N.
,
Francke
,
U.
,
Fried
,
V. A.
,
Ullrich
,
A.
and
Williams
,
L. T.
(
1986
).
Structure of the receptor for platelet-derived growth factor helps define a family of closely related growth factor receptors
.
Nature
323
,
226
232
.