Quantitation of Platelet-Derived Growth Factor Receptors in Human Arterial Smooth Muscle Cells In Vitro
Abstract Platelet-derived growth factor (PDGF) is suggested to play an important role in the development of atherosclerosis as a migratory and mitogenic stimulus to arterial smooth muscle cells (ASMCs). Stimulated and unstimulated ASMCs were studied with respect to PDGF receptor (PDGF-R) mRNA and protein expression. Quantitative RT-PCR was developed for simultaneous evaluation of both PDGF-Rα and -Rβ mRNA expression and a quantitative ELISA for estimation of corresponding PDGF-R subunits. On the mRNA level, the overall PDGF-Rβ expression was approximately 100 times lower than that of PDGF-Rα. Furthermore, although PDGF-Rα mRNA levels were high irrespective of hASMC phenotype, PDGF-Rβ mRNA was influenced by serum stimulation with lower copy numbers in proliferating and confluent cells compared with quiescent cells. On the protein level, quiescent hASMCs expressed 10 times more PDGF-Rβ than PDGF-Rα. Serum stimulation decreased cell surface PDGF-Rs, with most prominent loss of PDGF-Rα (ELISA and immunohistochemistry). Our results suggest a differential regulatory pattern for PDGF-Rα and -Rβ and are compatible with the usage of alternative promoters for regulation of -Rα expression. Further, it seems that the number of available receptor subunits is not the only determinant of variations in cell stimulation with different PDGF isoforms.
- Received October 29, 1996.
- Accepted March 17, 1997.
Migration and proliferation of medial SMCs and their subsequent proliferation within the intima contribute both to intimal thickening of atherosclerotic vessels and to restenosis after reconstructive vascular surgery. During migration to the intima, SMCs change the differentiated, contractile phenotype to the dedifferentiated, synthetic state that is associated with proliferation. PDGF is a potent mitogen and chemoattractant for mesenchymal cells and is suggested to play an important role in atherosclerosis. The administration of an anti-PDGF antibody inhibited neointimal SMC accumulation after angioplasty in rats.1 Introduction of a gene construct encoding PDGF-B chain into intimal cells of normal porcine iliofemoral arteries resulted in vivo in vascular SMC proliferation and vascular stenosis.2
PDGF occurs as chimeras (AA, AB, or BB) of two distinct polypeptide chains denoted A and B. The A-chain occurs as two isoforms (long and short) as a result of alternative usage of exon 6. The longer isoform has a highly basic C-terminal sequence corresponding to exon 6. This sequence binds specifically with high affinity to heparin-like glycosaminoglycans3,4 and has cellular retention properties.5 PDGF stimulates cell proliferation and migration by binding to specific cell surface receptors (PDGF-Rs) composed of two subunits: α and β. The receptor subunits exist as free monomers in the absence of ligand.6 Dimerization of receptor subunits on binding to dimeric PDGF is a prerequisite for the mitogenic signal transduction. Together with the colony-stimulating factor-type 1 receptor (CSF-1R) and the cellular homolog of the oncogene product from Hardy-Zuckerman 4 feline sarcoma virus (c-kit protein), the PDGF-Rs form a subfamily among the protein tyrosine kinase receptors.7 The extracellular domain of the α subunit binds both PDGF-A and -B chains with high affinity, whereas the β subunit binds only the B-chain specifically. The binding sites in the extracellular domain of the two receptor subunits are structurally distinct.8 Mutational analyses have previously shown that loops 1 and 3 of PDGF are important for receptor binding.9,10 Recently it was shown that the loop 2 region is more important for binding to PDGF-Rβ than PDGF-Rα.11
PDGF-induced DNA synthesis and cell proliferation involves activation of the Ras proto-oncogene, mitogen-activating protein (MAP) kinase kinase, and MAP kinase.12 On the other hand, migration of cells is caused by an activation of the diacylglycerol-pathway and an elevation of intracellular calcium levels leading to disassembly of actin filaments. Furthermore, molecules involved in the intracellular signaling seem to be regulated differentially depending on the phenotypic state of the SMCs.12 However, the signal mechanism of PDGF-induced migration and/or proliferation of hASMC and the change from a contractile to a synthetic phenotype is still poorly understood.
Hosang and Rouge13 provided evidence for the existence of two different kinds of PDGF receptors in human umbilical SMC in vitro, one binding all three isoforms of PDGF and another binding only PDGF-BB. Further, different isolates of human thoracic aorta SMCs show a quantitative difference in the expression of PDGF-Rα and -Rβ subunits and a differential sensitivity to stimulation with different PDGF isoforms.14
It has been suggested that the responsiveness of cells to PDGF stimulation is regulated by the relative abundance of PDGF cell surface receptor subunits. Knowledge about the regulatory elements involved in transcription of these subunits has been limited so far. The PDGF-Rα gene was recently shown to contain at least two promoters in its genomic sequence that are responsible for different transcripts.15,16 Further, different forms of β-receptor mRNAs can also be produced, presumably because of alternative usage of two transcriptional promoters or posttranscriptional processing.17 The role of truncated proteins encoded by alternative PDGF-R transcripts is not known, since presently none of these proteins have been detected in vivo.
The aim of this paper was to develop methods for the quantitation of PDGF-R subunit expression on the mRNA (quantitative RT-PCR) and protein (quantitative ELISA) levels in hASMCs in vitro. The potential effects on receptor expression of changes in cell phenotype were studied. Further, we compared the PDGF-Rs expression levels with the responsiveness of hASMC to stimulation with different recombinant PDGF homodimers.
Cell Culture Conditions
Smooth muscle cells were isolated from the inner media of human uterine arteries by explantation and treated as described previously.18 Secondary (passage 5 to 8), untransformed, and mycoplasma-free hASMC from three donors were used. Cells were seeded in Waymouth’s MB 752/1 medium containing 100 U/mL penicillin, 100 μg/mL streptomycin, 2 mmol/L L-glutamine, and 2 mmol/L sodium pyruvate (basal medium, BM) supplemented with 10% (vol/vol) human serum and 10% fetal bovine serum (S-BM). Growth-arrest was induced in sparse (minimal cell-cell contact) cultures by incubation for 3 days in BM containing 1% (wt/vol) BSA (BSA-BM) instead of sera. Quiescent hASMC were harvested after another 3 days of incubation in BSA-BM while still in sparse cultures. Similar sparse cultures of proliferating hASMC were obtained by replacing the BSA-BM with S-BM for 3 days before harvest. Confluent (maximal cell-cell contact) hASMC were grown to confluence in multi-layered cultures before harvest.
Preparation of Total RNA
Total RNA was isolated as described.19 The purity of each RNA preparation was confirmed by measuring the absorbance at 260 nm (A260) and 280 nm (A280), respectively. All preparations with a ratio of A260/A280≥1.8 and showing intact electrophoretic bands corresponding to 18S and 28S ribosomal RNA were further used for RT-PCR quantitation.
Construction of Internal Standard for Quantitative RT-PCR
Primers for the α-receptor were cloned into the commercially available plasmid pAW108 (American Type Culture Collection, Rockville, Md) already containing specific primer pairs for PDGF-Rβ mRNA (Fig 1⇓). Both receptor primers corresponded to intracellular receptor domains. The complete cloned sequences are shown in Table 1A⇓. They contain sequences corresponding to primers for the PDGF-Rα shown in bold face, followed by sequences corresponding to primers for the VLDL (Very Low Density Lipoprotein) receptor. To facilitate selection of transformants, internal restriction sites for SmaI (5′-primer) and BamHI (3′-primer) were included (underlined in Table 1A⇓). The two complementary oligonucleotides corresponding to the 3′-primer (Oligo III and IV, Table 1A⇓) were annealed and ligated into the NarI site of pAW108. Similarly, the annealed 5′-primer oligonucleotides (Oligo I and II; Table 1A⇓) were cloned into the BstEII site of the plasmid (Fig 1⇓). After transformation, the new plasmid DNA containing both primers for the PDGF-Rα was isolated and denoted pAW108(αβ). The correct cloning was verified by PCR, restriction enzyme cleavage (BamHI or SmaI), and sequencing.
This plasmid was linearized with the restriction enzyme BamHI. A large-scale in vitro “run off” transcription was performed using a Transcription Kit as described (Promega, Madison, Wis). The concentration of the resulting pAW108(αβ) RNA was determined by quantitative RT-PCR against the commercially available pAW109 RNA of known concentration (Perkin-Elmer Cetus, Norwalk, Colo) using primer pairs for different target genes contained in both internal standards. The results of this comparison indicated that the target genes on both internal standards were amplified to the same extent, as shown by identical mRNA copy numbers obtained by quantitative RT-PCR.
Quantitative RT-PCR of PDGF-R mRNAs
The estimation was made in triplicate using quantitative RT-PCR,20 a Gene Amp RNA PCR Kit (Perkin-Elmer Cetus, Norwalk, Conn), and pAW108(αβ) RNA as internal standard. The oligonucleotide primers for the α-receptor were α3a/α3b (Table 1B⇓) and for the β-receptor βa/βb (20, Table 1B⇓). The reverse transcription (RT) mixture contained 1 to 10 ng of total cellular RNA from cultured hASMC, 104 to 106 molecules of pAW108(αβ) RNA, random hexamer primer, and SuperScriptTM reverse transcriptase (Life Technologies, Täby, Sweden). The RT mixture was incubated and processed for amplification as described21 using 5′-primers labeled with [γ-32P]ATP (specific activity 5000 Ci/mmole; Amersham Sweden). Annealing and extension steps were run at 64°C for 1 and 7 minutes, respectively. Incubations were made in a DNA Thermal Cycler (Perkin-Elmer Cetus, Norwalk, Conn). PCR products were separated on a 4% low melting Nusieve GTG agarose gel (FMC, BioProducts, Me) and visualized with ethidium bromide staining (Fig 2⇓). The appropriate bands were excised from the gel, melted, and the radioactivity counted. The amounts of radioactivity recovered from the excised gel bands were plotted against the known concentrations of the two different templates and the number of receptor mRNA copies/pg RNA calculated.20
Semiquantitative PCR of PDGF-R mRNAs
The primer pairs used for the semiquantitative RT-PCR of PDGF-Rα (α1a/α1b and α2a/α2b, extracellular receptor domain) and PDGF-Rβ (βa/βb, intracellular domain) mRNAs are shown in Table 1B⇑. PCR was performed without internal standard, and radioactivity of PCR products was determined as described.
Long RT-PCR of PDGF-Rα mRNA
ExpandTM Long Template PCR System (Boehringer Mannheim, Mannheim, Germany) was used essentially as described.22 The amplification was performed in a Trio-Thermoblock (Biometra, Göttingen, Germany) with primers α4a/α3b (Table 1B⇑) as follows: denaturation at 94°C for 3 minutes, followed by 30 cycles of denaturation at 94°C for 1 minute, annealing at 65°C for 1 minute, and extension at 68°C for 2 minutes. The final 20 cycles were run with an elongated extension time of 20 seconds per cycle. A final extension was performed at 68°C for 7 minutes. The PCR products were visualized on a 1% agarose gel (Fig 3⇓).
Quiescent cells in semiconfluent or confluent cultures were harvested and dispensed in 96-well plates as described.23 105 cells/well for the α subunit and 0.5×105 cells/well for the β subunit were used. Throughout the experiments ELISA plates were washed as described.23 Nonspecific binding was blocked with PBS containing 0.1% Tween20 (PBS/Tween20) and 10% normal rabbit serum (Vector Laboratories, Burlingame, Calif) for 60 minutes at 37°C. Primary antibodies were mouse anti-human platelet-derived monoclonal antibodies directed specifically against PDGF-Rα and -Rβ subunits (Genzyme, Cambridge, Mass), respectively, and a negative control antibody (Mouse IgG1, Dako A/S, Glostrup, Denmark). Monoclonal antibodies of the IgG1 subclass against PDGF-R subunits were diluted (from 0.2 to 0.004 g/L for the α subunit and from 0.1 to 0.001 g/L for the β subunit) in PBS/Tween20 containing 10% normal rabbit serum and incubated with the cells for 2 hours at 37°C. Plates were then incubated for 60 minutes at 37°C with Biotinylated Rabbit Anti-Mouse antibody diluted according to the manufacturer′s instructions (Dako A/S, Glostrup, Denmark). Next, cells were incubated for 60 minutes at 37°C with ABComplex/AP (avidine-conjugated alkaline phosphatase complex) diluted according to Dako A/S. One g/L substrate p-nitrophenylphosphate (Sigma, St. Louis, Mo) in 10% diethanolamine, 0.5 mmol/L MgCl2 (pH 9.8) was added at room temperature and the kinetics of the reactions monitored by reading the absorbance at 405 nm and 650 nm at room temperature in a Vmax microplate reader (Molecular Devices, Menlo Park, Calif). This gives the rate (Vmax) of color development during a total runtime of 30 minutes with a reading interval of 15 seconds. Blanks without primary antibody were subtracted from all readings. A maximum rate of color development was obtained with 0.2 g/L for the α and with 0.1 g/L for the β subunit antibodies. This indicated that the availability of the corresponding receptor epitopes became rate-limiting at these concentrations, which consequently were used in subsequent quantitations.
The time course of downregulation of PDGF cell surface receptors was studied in confluent and semiconfluent cultures of hASMC. Quiescent cells were serum-stimulated for 0, 6, 12, 48, and 72 hours in confluent cultures and for 48 hours in semiconfluent cultures in triplicate wells before harvest. Cells were then harvested simultaneously, prepared, counted, and subjected to ELISA as described above.
Cells were seeded in 8-chamber glass Laboratory-Tek slides (Nunc, Naperville, Ill) and pretreated as described above. Proliferating (serum-stimulated) and quiescent (serum-deprived) hASMC were studied in sparse (about 500 cells/cm2) cultures and compared with confluent hASMC cultures in duplicates. All slides were washed 3 times with PBS-BSA. Cells were incubated for 30 minutes at 37°C in a humid chamber with antibodies against PDGF-Rα or -Rβ subunits. In stainings shown here, the dilutions in PBS-BSA giving the best signal to noice ratio in chess-board titrations (1:2 for α and 1:20 for β subunits) were used. The control antibody was diluted 1:50. Slides were washed 3 times in PBS-BSA and fixed with 4% formaldehyde in PBS for 15 minutes at 4°C. Cells were treated with 0.1% Triton X-100 in 0.15 mol/L NaCl, 1 mmol/L EDTA, and 50 mmol/L Tris (pH 7.2) for 5 minutes at room temperature. After washing, bound IgG was visualized by incubating the cells with FITC-rabbit anti-mouse IgG antibody (Southern Biotechnology Associates Inc, Birmingham, Ala) diluted 1:50. Slides were mounted with Vectashield mounting medium (Vector Laboratories, Burlingame, Calif) and studied in a Zeiss Axiophot microscope (Carl Zeiss, Oberkochen, Germany) equipped with epifluorescence and appropriate filters.
Stimulation of hASMC with different PDGF homodimers
Recombinant homodimers of the long (dimeric rA125) and short (dimeric rA109) PDGF A-chain variants were prepared as described earlier.4 Recombinant PDGF-BB homodimer was purchased from R&D Systems (Abingdon, UK). Human ASMC were seeded as above and growth-arrested. Recombinant PDGF homodimers were diluted in Waymouth′s MB 752/1 medium containing 10% human serum deprived of heparin-binding growth factors by Heparin-Sepharose affinity chromatography24 and added at different concentrations (0 to 40 nmol/L) to triplicate wells of quiescent cells. hASMC grown in 10% human and 10% fetal bovine sera were used as control. After 4 days, new test medium and 0.8μCi [3H]thymidine (specific activity 1 mCi/mL) (Amersham International, Buckinghamshire, UK) were added to each well. DNA synthesis was measured as thymidine index.18 The possibility that PDGF-AA and PDGF-BB had additive/synergistic effects was studied in experiments where PDGF-BB was added together with dimeric rA125 at 2.5 nmol/L each.
Two-way analyses of variance (ANOVA) was used to test the influence of donors and culture conditions. Descriptive statistics were used on untransformed data to calculate means and standard errors of means (SEM). P values of <0.05 (two-sided tests) were regarded as statistically significant.
Quantitative RT-PCR Analyses of PDGF-Rα and -Rβ mRNA Expression
The expression of PDGF-R mRNAs was quantitated by the RT-PCR method using the constructed internal standard pAW108(αβ) RNA and primers corresponding to intracellular parts of both α- and β-receptors (Table 1B⇑). All RT-PCR products obtained from amplification of both α- and β- receptors and internal standard were of expected sizes (Fig 2⇑). According to our results, human ASMC from all donors under all studied culture conditions (quiescent, proliferating, and confluent) expressed transcripts for both PDGF-R subunits (Fig 4⇓). The expression of PDGF-Rα mRNA seemed to be constantly high irrespective of cell phenotype (P=.564). Our results suggested variation between donors (P=.012, data not shown).
The expression of PDGF-Rβ mRNA was much lower than for the PDGF-Rα subunit. In contrast to α-mRNA, the expression of β-mRNA varied with different culture conditions (P=.002, Fig 4⇑). Thus, the expression of PDGF-Rβ transcripts were, on average, almost 10 times higher in mitogen-deprived (quiescent) hASMC than in serum-exposed (proliferating or confluent) cells (Fig 4⇑). Finally, the overall number of transcripts for the α subunit (1066±177 copies/pg RNA, mean±SEM) was at least 2 orders of magnitude higher (P<.01) than that for the β subunit (6±0.6 copies/pg RNA).
The relatively high levels of transcripts for the α subunit compared to the β subunit were unexpected since previous publications13,14 have indicated a higher concentration of β than α subunits on the protein level in hASMC. Therefore, the results were checked in several ways to ascertain that the obtained results were not the result of an artifact caused by the PCR amplification.
The specificity of the primers for the PDGF-Rα and the uniqueness of the amplified fragment were verified using the BLAST network service25 at the National Center for Biotechnology Information (NCBI). Sequences producing high-scoring segment pairs were those of the PDGF-Rα only. Further, the PCR fragment was sequenced with the 32P-end labeled 5′-primer α3a (Table 1B⇑) as sequencing primer using the fmol® DNA sequencing System (Promega, Madison, Wis). The obtained sequence conformed with the expected PDGF-Rα fragment.26 We concluded that the amplification obtained during quantitative RT-PCR was specific for the gene of interest.
To validate our comparisons between the expression of PDGF-Rα and -Rβ mRNAs, priming efficiencies of the α- and β- primers were compared on the internal standard pAW108(αβ) where the amplified sequences were almost identical (Fig 1⇑). The results showed that the priming efficiency was in favor of the amplification of the β-receptor gene (10 times better, data not shown) compared with the α-receptor gene. Consequently, the high ratio of PDGF-Rα/-Rβ mRNAs could not be explained simply by the difference in priming efficiency of the α and β-primers. The efficiency of PCR may decrease with increasing length and GC-content of amplified sequences. The PCR fragment of the α receptor (420 bp) was twice as long as that of the β receptor (230 bp), and both fragments contained similar amounts of GC-nucleotides (51% in the α and 61% in the β fragment) and comparatively high nucleotide homology (40%) based on sequence analysis with the GCG-program package.27 Thus, we concluded that our results showing much higher expression of α- than β-subunit mRNA in hASMC grossly reflects the actual situation.
Expand Long Template PCR system
Since quantitation of PDGF-Rα mRNA expression is based on amplification of the intracellular part of the receptor, we wanted to ascertain that hASMC expressed the full-length PDGF-Rα mRNA. We used the Expand Long Template PCR system, which allows amplification of longer fragments than can be amplified by conventional PCR methods. With primer pair α4a/α3b (Table 1B⇑), we could amplify almost the entire peptide-coding region of the PDGF-Rα mRNA in this system (Fig 3⇑). When RNA derived from different donors was tested, in all cases only one band was detected. This band was of the expected size, about 3264 bp, which is closely comparable to the entire peptide-coding sequence of 3395 bp. This confirmed that hASMC express the full-length mRNA for PDGF-Rα. These results also confirmed that a splice variant of the PDGF-Rα lacking exon 1428 does not exist in hASMC.
Quantitative ELISA of PDGF Cell Surface Receptor Subunits
In order to evaluate the relative abundance of PDGF cell surface receptor subunits, a quantitative ELISA was developed. Saturating concentrations of antibodies were used for both PDGF-Rs providing Vmax values, which allowed a comparison of the expression of both subunits. In quiescent semiconfluent cultures, there was, on average, a 13-fold excess of β compared with α subunits (P=.027, Fig 5⇓).
It has been suggested that serum-stimulated cells express fewer cell surface receptors because of internalization of receptor-ligand complexes, and in this paper we show that PDGF-Rβ mRNAs are down-regulated in confluent hASMC cultures (Fig 4⇑). The temporal changes in expression of receptor subunit proteins on the cell surface were studied with ELISA as a function of time after serum stimulation of quiescent confluent and semiconfluent hASMC cultures. Both PDGF-Rs were partially downregulated in confluent cultures within 72 hours of growth stimulation (Fig 6⇓). In these cells the amount of α-subunit was reduced to 60% whereas that of β subunit was reduced to 80% of controls, approximately. In semiconfluent cultures, the amount of α-subunit was reduced to 25% of controls, whereas the expression of β subunit was unchanged during the 48-hour time span. Thus, our quantitative experiments with ELISA suggested that it was mainly the α subunit of the PDGF-R, which was downregulated on the cell surface in response to ligand stimulation, even in cultures with a considerable degree of cell-cell contact.
Immunofluorescent Staining of PDGF Cell Surface Receptors
Immunofluorescent staining of α-receptor subunits showed qualitatively the presence of very sparse clusters of immunofluorescence on the surface of hASMC in sparse quiescent cultures (Fig 7A⇓). In contrast, staining of the β-receptors in these cultures showed a much higher number of clusters (Fig 7B⇓). This was consistent with the quantitative ELISA. In controls with nonimmune mouse IgG, no clusters of immunofluorescence were found (Fig 7C⇓). Almost no clusters of fluorescence were found on the surface of serum-stimulated sparse hASMC, suggesting that very low amounts, if any, of PDGF-R subunits were exposed on the cell surface (Figs 7D⇓ and 7E⇓). Confluent cells expressed both PDGF-R subunits irrespective of whether they were deprived of or stimulated with serum, as shown for PDGF-Rα in Fig 7F⇓ and 7G⇓.
The localization of the immunofluorescent staining to the cell surface was confirmed using a confocal LSM microscope (Axiovert 135 mol/L, Microsystems, Carl Zeiss, Oberkochen, Germany). Stained cells were analyzed in 10 cross-sections, and a 3-dimensional computer representation was created. The staining was restricted to the cell surface only and followed the shape of the cell. No distinct immunofluorescence was seen within the cytoplasm with the technique used in these experiments (results not shown).
Semiquantitative RT-PCR Analyses
According to the results described above, the large discrepancy observed for PDGF-Rα and -Rβ between the mRNA data (β/α ratio approximately 1/100) and the protein data (β/α approximately 10/1) suggests the possibility of a differential regulation of PDGF-R expression. Since truncation of the extracellular domain of the PDGF-Rα caused by alternative usage of two promoters during transcription has been reported previously,15,16 a semiquantitative RT-PCR was established with primer pairs amplifying sequences in the extracellular domain of the α subunit mRNA.
Amplification with anyone of the extracellular domain primer pairs (α1a/α1b and α2a/α2b, Table 1B⇑) produced less PCR product than the intracellular primer pair (α3a/α3b) (data not shown). The expression of β- (primer pair βa/βb, Table 1B⇑) and α-mRNAs was compared using primers labeled with the same specific activity, the PCR amplifications run simultaneously, and the products separated on the same gel. Assuming equal amplification efficiency, the β/α ratio was 1.5/1 (Table 2⇑), which was much closer to the reported relative approximate abundance of the receptor subunits on the cell surface (β/α=10/1) than the β/α ratio obtained by amplification of α-receptor mRNA with intracellular primer pair (β/α ratio approximately 1/100). The β/α ratio was stable under different cell culture conditions (P=.899). These results are compatible with the possibility of the expression of PDGF-Rα mRNAs truncated in their extracellular domain in addition to the full-length variant in hASMC.
Stimulation of hASMC with PDGF Homodimers
Quiescent hASMC from all donors incorporated radiolabeled thymidine into DNA in response to stimulation with exogenous dimeric rA125, rA109 or PDGF-BB (Figs 8A⇓-8C). In the absence of exogenous PDGF, the thymidine indices varied between 10 and 30% of serum-stimulated references. This may be partly caused by endogenous production of PDGF by hASMC (A. Krettek, unpublished data, 1997). When the recombinant homodimers were added to hASMC, they repeatedly induced a dose-dependent increase in thymidine incorporation into DNA up to about 5 to 10 nmol/L concentration (Fig 8⇓). However, above optimal homodimer concentrations, the degree of stimulation decreased slightly but consistently. From 0 to 5 nmol/L of exogenous PDGF, the increase in thymidine indices was independent of donor (P=.694). On average, the maximal increase above the baseline (no added PDGF) to 5 nmol/L PDGF BB was 38±3% (mean±SEM) of reference cultures. This was insignificantly higher (P=.091) than the stimulation of thymidine indices observed with dimeric rA125 (19±3%) and dimeric rA109 (21±6%), which were almost equally effective.
The possibility of additive and/or synergistic effects between dimeric rA125 and PDGF-BB was tested on hASMC from one donor. When PDGF-BB was combined with dimeric rA125, the combination did not increase thymidine incorporation above the stimulation seen with PDGF-BB alone, and the pattern of dose-dependence was similar to that of PDGF-BB (data not shown). This suggested that PDGF-BB and PDGF-AA had simple additive but not synergistic effects.
During the development of atherosclerotic vascular disease and restenosis, the movement of ASMC from the media to the intima in the arterial wall of injured vessels is accompanied by a mitogen-induced change in cell phenotype. The mitogenic stimulation of cells is a complicated process, wherein the concentration of mitogen has to be balanced by the relative abundance of appropriate receptors. Because PDGF is considered to be one of the main mitogens and chemoattractants for SMC and PDGF signaling requires expression of PDGF-Rs, we decided to investigate the effect of cell phenotype on the expression of PDGF-R mRNA and protein in hASMC. For this purpose, we have developed quantitative methods for simultaneous evaluation of PDGF α- and β-receptor expression both on mRNA (quantitative RT-PCR) and on protein levels (quantitative ELISA).
Quantitative RT-PCR with primers amplifying the intracellular domain of the receptors showed that quiescent, proliferating, and confluent hASMC expressed mRNAs for both receptor subunits. However, whereas the abundance of mRNAs for the α subunit was constitutively high irrespective of cell phenotype, that for the β subunit was two orders of magnitude lower and varied with cell phenotype (quiescent > proliferating ≈ confluent). This suggests that ligand stimulation of sparse cells and/or cell-cell interactions among confluent cells may induce growth arrest at least partly by lowering endogenous production of PDGF-Rβ. Due to the fact that mRNA for PDGF-Rα was highly expressed irrespective of cell phenotype, the ratio of β/α subunit mRNA was lowest in confluent and proliferating and highest in quiescent cells. A phenotype-dependent expression of PDGF-Rs has been previously reported from in vitro studies on rat arterial SMC.29
Using quantitative ELISA, we have shown that quiescent semiconfluent cells expressed approximately 10-fold excess of β subunits compared with α subunits on the cell surface. On serum stimulation, mainly the α subunit of the PDGF-R was downregulated on the cell surface, even in cultures with considerable degree of cell-cell contact.
Our results suggest that the expression of PDGF-Rα and PDGF-Rβ are regulated differentially both at the protein and the mRNA levels. For the α subunit, the mRNA containing the intracellular domain was high but largely unregulated, whereas the extracellular peptide domain was low but highly regulated. For the β subunit, the corresponding mRNA was low but regulated, whereas the analogous protein was high but less regulated. Taken together, these results suggest an independent control of PDGF-Rα and PDGF-Rβ expression, which may also be accomplished by different cytokines secreted from cells of the atherosclerotic lesion.30,31 The reason for such a mechanism may be explained by the different functions of PDGF receptors. The binding of the ligand to any of the PDGF receptors stimulates SMC proliferation.32,33 However, it has been suggested that activation of PDGF-Rβ by PDGF-BB primarily stimulates chemotaxis, whereas binding of the same ligand to PDGF-Rα inhibits SMC chemotaxis.34 One can assume that cells on serum stimulation stop to migrate at least partly because of a downregulation of the PDGF-Rβ message.
There are two theoretical explanations for the discrepancies observed between the mRNA (ratio of β/α of about 1/100) and protein levels (ratio of β/α of about 10/1) of PDGF-R subunits. One possibility is the occurrence of PDGF-Rβ mRNAs and proteins truncated in their intracellular domains. Such species would consequently be detected by ELISA with antibodies directed against the extracellular domains of the receptor, but not by our RT-PCR assay amplifying sequences coding for the intracellular part of the PDGF-Rβ. However, until now only PDGF-Rβ mRNAs with deleted extracellular domains have been detected.17 Another explanation might be the presence of full-length PDGF-Rα mRNA as well as truncated variants lacking a part corresponding to the extracellular domain of the PDGF-Rα subunit. Our qualitative long-template PCR system showed the presence of a transcript corresponding to the full-length α receptor sequence. Variants of PDGF-Rα arising from alternative promoter usage have been described previously in teratocarcinoma cells.16,28 Even if translated, such protein variants, lacking the extracellular domain, would not be recognized by the antibodies used in this study. If this were the case, it could explain our observation that using primer pairs for the mRNA sequence corresponding to the extracellular α-receptor domain, we found a ratio of β/α mRNA close to 1/1, which was much higher than the ratio of about 1/100 found with the intracellular domain-primers and closer to the ratio of about 10/1 seen on the protein level. Thus, our results are compatible with the possibility of alternative promoter usage during the transcription of the PDGF-Rα gene in hASMC.
Further, we wanted to investigate if variations in growth stimulation with different PDGF isoforms can be explained by the relative distribution of PDGF-Rα and -Rβ subunits on the cell surface. According to our results, PDGF-BB consistently stimulated thymidine incorporation into hASMC DNA more efficiently than the two variants of dimeric rA. This difference was consistent with observations by others.35,36 The degree of stimulation of cell growth by PDGF variants can depend either on variation in signaling activity of one of the receptor subunits or/and on the difference in their relative abundance on the cell surface. Our results showed that the abundance of immunoreactive PDGF-Rβ subunits was much higher (approximately 10 times) than that of -Rα subunits on the surface of hASMC in vitro, in concordance with other reports.13,14 However, since PDGF-BB stimulates through both receptor subunits, whereas PDGF-AA stimulates the cells only through the less abundant PDGF-Rα, larger differences in stimulatory capacity between PDGF-BB and PDGF-AA variants were expected. Our results suggest that the numbers of available receptor subunits are not the only determinants of maximal thymidine incorporation, and other possibilities must be taken into consideration. One possibility would be that because of higher endogenous production of PDGF-AA than of PDGF-BB (A. Krettek, unpublished data, 1997), a more efficient internalization of PDGF-AA/PDGF-Rα complexes may result in more efficient signal transduction. Indeed, the α subunit was down-regulated more than the β subunit on stimulation with serum. This was most prominent for PDGF-Rα in semiconfluent cultures with reduction to 25% compared with unstimulated controls. The β subunit seemed to be either slightly affected (quiescent confluent cells) or not affected at all (quiescent semiconfluent) by serum stimulation. In sparse cultures, down-regulation of both receptor subunits was pronounced as judged from immunofluorescent staining. We have shown previously that hASMC became quiescent in confluent multilayered cultures, even in the presence of serum mitogens, indicating that they became unresponsive to serum mitogens, including PDGF.18 Our current results showed that confluent cells expressed both PDGF-R subunits on the cell surface also in the presence of PDGF. This suggested that the unresponsiveness of confluent cultures to stimulation with serum or PDGF was not caused by complete lack of receptors. It has been suggested previously that mechanical interactions between cells and their surrounding matrix provide other regulatory signals that modulate autophosphorylation of growth factor receptors and cell proliferation.37
The thymidine incorporation on stimulation of hASMC with all three PDGF isoforms showed a bimodal dose dependence with increase in thymidine indices up to a maximum stimulation at about 5 nmol/L and consistent dose-dependent decrease in stimulation above this concentration. Recently, a bimodal response was described for bFGF,38 and similar results were obtained in studies of the effect of PDGF variants on the migration of baboon vascular SMC.32 The similarities between the bimodal mitogenic and migratory responses of hASMC to stimulation with PDGF-AA and -BB suggest that both may obey a similar basic mechanism. The possible explanation for these observations is that the dimerization of receptor subunits required for optimal signal transduction is maximal at a molar ratio of ligand to receptor of 1:2 and that above this level, additional ligand will block signaling by monosaturating the receptor subunits.
In conclusion, our results are compatible with a complex transcriptional regulation of PDGF-R expression in hASMC and with the possibility that a substantial part of the PDGF-Rα transcripts may be truncated. They are also compatible with a regulation caused by internalization of ligand-receptor complexes, cell phenotype, or cell-cell interactions. Given the low abundance of divalent PDGFR-α subunits and the high abundance of monovalent PDGF-Rβ subunits on the cell surface, the small difference in stimulatory capacity between PDGF-BB and dimeric rA variants suggested that the numbers of available receptor subunits are not the only determinants of maximal thymidine incorporation.
Selected Abbreviations and Acronyms
|ASMC||=||arterial smooth muscle cell|
|ELISA||=||enzyme-linked immunosorbent assay|
|hASMC||=||human arterial smooth muscle cell|
|PCR||=||polymerase chain reaction|
|PDGF||=||platelet-derived growth factor|
|SMC||=||smooth muscle cell|
This study was supported by grants from the Swedish Medical Research Council (grant No. 4531), the Swedish Heart Lung Foundation (grant No. 51044), and the Wilhelm & Martina Lundgren′s Science Foundation. Alexandra Krettek is the recipient of a PhD grant from the Swedish Heart Lung Foundation (grant No. 55018). We thank Dr. Johan Hoebeke at CJF INSERM 93-09, Université Francois Rabelais, Tours, France, for helpful advice on the ELISA measurements.
Ferns GAA, Raines EW, Sprugel KH, Motani AS, Reidy MA, Ross R. Inhibition of neointimal smooth muscle accumulation after angioplasty by an antibody to PDGF. Science. 1991;253:1129-1132.
Nabel EG, Yang Z, Liptay S, San H, Gordon D, Haudenschild CC, Nabel GJ. Recombinant platelet-derived growth factor B gene expression in porcine arteries induces intimal hyperplasia in vivo. J Clin Invest. 1993;91:1822-1829.
Fager G, Camejo G, Bondjers G. Heparin-like glycosaminoglycans influence growth and phenotype of human arterial smooth muscle cells in vitro. I. Evidence for reversible binding and inactivation of the platelet-derived growth factor by heparin. In Vitro Cell Dev Biol. 1992;28A:168-175.
LaRochelle WJ, May-Siroff M, Robbins KC, Aaronson SA. A novel mechanism regulating growth factor association with the cell surface: identification of a PDGF retention domain. Genes Dev. 1991;5:1191-1199.
Herren B, Rooney B, Weyer KA, Iberg N, Schmid G, Pech M. Dimerization of extracellular domains of platelet-derived growth factor receptors: a revised model of receptor-ligand interaction. J Biol Chem. 1993;268:15088-15095.
Hanks SK, Quinn AM, Hunter T. The protein kinase family: conserved features and deduced phylogeny of the catalytic domains. Science. 1988;241:42-52.
Heidaran MA, Yu J-C, Jensen RA, Pierce JH, Aaronson SA. A deletion in the extracellular domain of the α platelet-derived growth factor (PDGF) receptor differentially impairs PDGF-AA and PDGF-BB binding affinities. J Biol Chem. 1992;267:2884-2887.
Clements JM, Bawden LJ, Bloxidge RE, Catlin G, Cook AL, Craig S, Drummond AH, Edwards RM, Fallon A, Green DR, Hellewell PG, Kirwin PM, Nayee PD, Richardson SJ, Brown D, Chahwala SB, Snarey M, Winslow D. Two PDGF-B chain residues, arginine 27 and isoleucine 30, mediate receptor binding and activation. EMBO J. 1991;10:4113-4120.
Hosang M, and Rouge M. Human vascular smooth muscle cells have at least two distinct PDGF receptors and can secrete PDGF-AA. J Cardiovasc Pharmacol. 1989;14:S22-S26.
Kraft HJ, Mosselman S, Smits HA, Hohenstein P, Piek E, Chen Q, Artzt K, Van Zoelen EJJ. Oct-4 regulates alternative platelet-derived growth factor α receptor gene promoter in human embryonal carcinoma cells. J Biol Chem. 1996;271:12873-12878.
Vu TH, Martin GR, Lee P, Mark D, Wang A, Williams LT. Developmentally regulated use of alternative promoters creates a novel platelet-derived growth factor receptor transcript in mouse teratocarcinoma and embryonic stem cells. Mol Cell Biol. 1989;9:4563-4567.
Wang AM, Doyle MV, Mark DF. Quantitation of mRNA by the polymerase chain reaction. Proc Natl Acad Sci U S A. 1989;86:9717-9721.
Mattson L, Johansson H, Ottosson M, Bondjers G, Wiklund O. Expression of lipoprotein lipase mRNA and secretion in macrophages isolated from human atherosclerotic aorta. J Clin Invest. 1993;92:1759-1765.
Kübler G, and Frey B. Amplification of fragments up to 7.7 kb from human dystrophin RNA by RT-PCR with AMV reverse transcriptase and the expandTM long template PCR system. Biochemica. 1995;3:15-16.
Cleasson-Welsh L, Hammacher A, Westermark B, Heldin C-H, Nistér M. Identification and structural analysis of the A type receptor for platelet-derived growth factor. J Biol Chem. 1989;264:1742-1747.
Devereux J, Haeberli P, Smithies O. A comprehensive set of sequence analysis programs for the VAX. Nucleic Acid Res. 1984;12:387-396.
Mosselman S, Cleasson-Welsh L, Kamphuis S, Van Zoelen EJJ. Developmentally regulated expression of two novel platelet-derived growth factor α-receptor transcripts in human teratocarcinoma cells. Cancer Res. 1994;54:220-225.
Schöllman C, Grugel R, Tatje D, Hoppe J, Folkman J, Marmé D, Weich HA. Basic fibroblast growth factor modulates the mitogenic potency of the platelet-derived growth factor (PDGF) isoforms by specific up-regulation of the PDGF α receptor in vascular smooth muscle cells. J Biol Chem. 1992;267:18032-18039.
Gronwald RGK, Seifert RA, Bowen-Pope DF. Differential regulation of expression of two platelet-derived growth factor receptor subunits by transforming growth factor-β. J Biol Chem. 1989;264:8120-8125.
Koyama N, Hart CE, Clowes AW. Different functions of the platelet-derived growth factor-α and -β receptors for the migration and proliferation of cultured baboon smooth muscle cells. Circulation Res. 1994;75:682-691.
Koyama N, Morisaki N, Saito Y, Yoshida S. Regulatory effects of platelet-derived growth factor-AA homodimer on migration of vascular smooth muscle cells. J Biol Chem. 1992;267:22806-22812.
Vassbotn FS, Östman A, Siegbahn A, Holmsen H, Heldin C-H. Neomycin is a platelet-derived growth factor (PDGF) antagonist that allows discrimination of PDGF α- and β-receptor signals in cells expressing both receptor types. J Biol Chem. 1992;267:15635-15641.
Heidaran MA, Beeler JF, Yu J-C, Ishibashi T, La Rochelle WJ, Pierce JH, Aaronson SA. Differences in substrate specificities of α and β platelet-derived growth factor (PDGF) receptors: correlation with their ability to mediate PDGF transforming functions. J Biol Chem. 1993;268:9287-9295.
Beckmann MP, Betsholtz C, Heldin C-H, Westermark B, Di Marco E, Di Fiore PP, Robbins KC, Aaronson SA. Comparison of biological properties and transforming potential of human PDGF-A and PDGF-B chains. Science. 1988;241:1346-1349.
Lin Y-C, and Grinnell F. Decreased level of PDGF-stimulated receptor autophosphorylation by fibroblasts in mechanically relaxed collagen matrices. J Cell Biol. 1993;122:663-672.
Bono F, Rigon P, Lamarche I, Lormeau JC, Herbert JM. Heparin inhibits the binding of 125I-bFGF to cultured human aortic smooth muscle cells. European Vascular Biology Association Meeting, Göteborg, Sweden, May 30-June 2, 1996. Abstract
Gronwald RGK, Grant FJ, Haldeman BA, Hart CE, O′Hara PJ, Hagen FS, Ross R, Bowen-Pope DF, Murray MJ. Cloning and expression of a cDNA coding for the human platelet-derived growth factor receptor: evidence for more than one receptor class. Proc Natl Acad Sci U S A. 1988;85:3435-3439.