Platelet-derived growth factor
Platelet-derived growth factor (PDGF) is a family of dimeric glycoproteins composed of four polypeptide chains (A, B, C, and D) that assemble into five bioactive isoforms, acting as potent mitogens, chemoattractants, and regulators of cell survival for mesenchymal-origin cells such as fibroblasts, smooth muscle cells, and pericytes.[1][2] Discovered in 1974 by Russell Ross and colleagues at the University of Washington while studying the pathogenesis of atherosclerosis, PDGF was initially identified as a soluble factor released from the alpha granules of activated platelets that promotes the proliferation and migration of arterial smooth muscle cells, contributing to the formation of atherosclerotic lesions.[3] Subsequent research revealed its broader production by various cell types, including endothelial cells, macrophages, and fibroblasts, particularly at sites of injury or inflammation.[4] Structurally, all PDGF isoforms share a conserved C-terminal cystine-knot growth factor domain stabilized by disulfide bonds, with molecular weights around 25–30 kDa for the mature dimers; the A and B chains, encoded by genes on chromosomes 7 and 22 respectively, can form both homodimers (PDGF-AA, PDGF-BB) and the heterodimer PDGF-AB, whereas the more recently identified C and D chains (discovered in the late 1990s and early 2000s) primarily form homodimers (PDGF-CC, PDGF-DD) and include an N-terminal CUB domain that requires proteolytic cleavage by proteases like plasmin or furin for activation and secretion.[2][5] These structural variations influence ligand-receptor specificity and biological activity, with PDGF-A and -B showing about 50% sequence identity to each other, while PDGF-C and -D share roughly 25% identity with A/B but 50% with one another.[5] PDGF signals through two homologous cell-surface receptor tyrosine kinases, PDGFRα (encoded on chromosome 4q12) and PDGFRβ (encoded on chromosome 5q32), which undergo dimerization (αα, αβ, or ββ) upon ligand binding, leading to autophosphorylation and activation of downstream pathways including PI3K/AKT for cell survival and MAPK/ERK for proliferation and migration.[2][6][7] In normal physiology, PDGF is indispensable for embryonic development—such as vasculogenesis, organogenesis, and neural crest cell migration—and adult tissue homeostasis, including wound healing where it recruits pericytes to stabilize new vessels and stimulates extracellular matrix production.[1][5] Dysregulated PDGF signaling, however, drives pathological conditions like fibrosis (e.g., in kidney and lung), atherosclerosis via excessive smooth muscle proliferation, and cancers such as gliomas and sarcomas through autocrine loops that promote tumor growth and angiogenesis.[1][4] Therapeutically, PDGF inhibitors like imatinib target overactive receptors in certain malignancies, highlighting its clinical significance.[2]Discovery and History
Initial Discovery
Platelet-derived growth factor (PDGF) was first identified in the early 1970s as a key mitogen released from platelets, stimulating the proliferation of arterial smooth muscle cells (SMCs) in vitro. In 1974, Russell Ross and colleagues at the University of Washington demonstrated that dialyzed serum derived from clotted primate blood promoted robust SMC growth, whereas serum from recalcified platelet-poor plasma exhibited significantly reduced mitogenic activity. Experiments revealed that adding platelets or a platelet-free supernatant from thrombin-activated platelets to the plasma-derived serum restored its proliferative effects to levels comparable to whole blood serum, pinpointing the platelets as the primary source of the growth-promoting factor.[8] This discovery was motivated by investigations into the cellular basis of atherosclerosis, where SMC proliferation plays a central role.[3] Subsequent cell culture studies using platelet extracts further characterized the factor's potency. Extracts from human or primate platelets were applied to quiescent cultures of SMCs and fibroblasts, where they induced DNA synthesis and cell division, as measured by incorporation of tritiated thymidine. For instance, in BALB/c 3T3 fibroblasts, platelet-derived material at concentrations equivalent to 1% whole serum triggered a marked increase in DNA synthesis within 24 hours, highlighting its role as a competence factor for initiating the cell cycle. These assays established the term "platelet-derived growth factor" (PDGF) to describe this cationic protein, distinct from other serum components like plasma growth-promoting activity.[8][9] Between 1974 and 1978, initial purification efforts isolated PDGF from outdated human platelet-rich plasma, confirming its localization within platelet alpha granules—the dense storage organelles released upon activation. Using techniques such as heat treatment, ion-exchange chromatography, gel filtration, and sodium dodecyl sulfate-polyacrylamide gel electrophoresis, researchers achieved up to 800,000-fold purification. The native protein was identified as a disulfide-bonded dimer with a molecular weight of approximately 30,000 Da, comprising two distinct polypeptide chains of about 14,000 Da and 17,000 Da. Early biochemical assays on human glial cells and fibroblasts showed that purified PDGF at nanogram concentrations (e.g., 4 ng/mL) stimulated DNA synthesis equivalently to 1% human serum, underscoring its high specific activity.[3][10][9]Key Milestones
In 1979, key purification and partial characterization work by Carl-Henrik Heldin's group confirmed PDGF as a distinct mitogen released from alpha granules in platelets, demonstrating its role in stimulating connective tissue cell growth.[10] The cloning of PDGF genes marked a pivotal advancement in the 1980s. In 1984, the PDGF-B chain gene was cloned by Heldin and colleagues, revealing its sequence homology to the v-sis oncogene from simian sarcoma virus and confirming PDGF as a dimeric protein composed of A and B chains. This was followed in 1986 by the cloning of the PDGF-A chain by Christer Betsholtz et al., which established the existence of homodimeric and heterodimeric isoforms and localized the gene to chromosome 7.[11] Concurrently, in 1986, Yarden et al. cloned the PDGF receptor (PDGFR), identifying it as a transmembrane tyrosine kinase and elucidating its dimerization upon ligand binding, which laid the foundation for understanding PDGF signaling specificity. By the late 1980s, PDGF was recognized as the founding member of the PDGF/VEGF growth factor family, following the 1989 cloning of vascular endothelial growth factor (VEGF) by Leung et al., which highlighted shared structural features such as the cystine-knot motif and conserved receptor-binding domains across these dimeric proteins. In the early 1990s, structural elucidation advanced with the 1992 determination of the crystal structure of human PDGF-BB by Olofsson et al., revealing a novel antiparallel disulfide-linked homodimer with two protruding receptor-binding regions, which provided insights into its mitogenic potency and informed subsequent isoform comparisons.[12] Parallel animal model studies emerged around this time; the first PDGF knockout mice, reported in 1994 by Levéen et al. and Soriano, demonstrated lethal perinatal phenotypes in PDGF-B and PDGFR-β null mutants, including renal glomerulogenesis failure and cardiovascular defects, underscoring PDGF's essential role in vascular development and pericyte recruitment.[13][14] The PDGF family was completed with the discovery of two additional chains in the early 2000s. PDGF-C was identified in 2000 by three independent groups through homology-based searches and expression studies, revealing it as a novel ligand primarily activating PDGFRα.[15] PDGF-D was cloned in 2001, showing specificity for PDGFRβ and requiring proteolytic activation similar to PDGF-C.[16] These findings expanded the understanding of PDGF diversity and signaling specificity.Structure and Classification
Molecular Composition
Platelet-derived growth factor (PDGF) consists of disulfide-linked homo- or heterodimers formed by two of four possible polypeptide chains, designated A, B, C, or D, resulting in a mature protein with a molecular weight of approximately 28-31 kDa.[17] These dimers are stabilized by covalent disulfide bonds, which are essential for the structural integrity and bioactivity of the growth factor.[18] The four PDGF chains are encoded by separate genes located at distinct chromosomal positions in the human genome: PDGFA on chromosome 7p22.3, PDGFB on chromosome 22q13.1, PDGFC on chromosome 4q32.1, and PDGFD on chromosome 11q22.3.[19][20] The primary structure of each PDGF chain precursor comprises an N-terminal signal peptide of 18-22 amino acids, followed by a prodomain and a central growth factor domain of about 100-140 residues. For PDGF-C and PDGF-D, the prodomain includes an N-terminal CUB domain that maintains latency until proteolytic cleavage.[5] This central domain contains eight highly conserved cysteine residues that form a characteristic cystine-knot motif, including three intermolecular disulfide bonds linking the two chains and two intramolecular bonds within each chain to maintain the dimeric fold.[18] The signal peptide is cleaved co-translationally during translocation into the endoplasmic reticulum, yielding the pro-PDGF form.[21] Post-translational modifications are critical for PDGF maturation and include proteolytic processing to remove prodomains and, in some cases, N-linked glycosylation. For PDGF-A and PDGF-B chains, intracellular cleavage by furin-like proprotein convertases at dibasic motifs (e.g., RRKR for A-chain, RGRR for B-chain) generates active dimers stored in platelet alpha-granules.[21] In contrast, PDGF-C and PDGF-D are secreted as latent complexes with their prodomains intact and require extracellular proteolytic activation by serine proteases such as plasmin, tissue plasminogen activator (tPA), or urokinase plasminogen activator (uPA).[21] Glycosylation occurs at specific asparagine residues in certain chains, such as three N-linked sites (Asn25, Asn55, Asn254) in PDGF-C, which may influence stability and processing, though it is absent or minimal in PDGF-A.[22][21] Dimerization is represented simply as the covalent linkage of two monomeric chains, such as A + B → PDGF-AB, facilitated by the conserved cysteines in the growth factor domains.[18] These structural elements form the foundational building blocks for the various PDGF isoforms.Isoforms and Family Members
Platelet-derived growth factor (PDGF) exists in five principal dimeric isoforms, formed by disulfide-linked combinations of four distinct polypeptide chains: PDGF-A, PDGF-B, PDGF-C, and PDGF-D. These isoforms include the homodimers PDGF-AA, PDGF-BB, PDGF-CC, and PDGF-DD, as well as the heterodimer PDGF-AB. The chains share structural homology, particularly in their conserved C-terminal growth factor domains, but exhibit varied tissue-specific expression patterns that contribute to their functional diversity. For instance, PDGF-BB is the predominant isoform stored in platelet alpha-granules and released upon activation, while PDGF-AA is widely expressed in epithelial and mesenchymal cells during development and repair processes.[23][24][23] The isoforms display distinct ligand specificities for the two PDGF receptor tyrosine kinases, PDGFRα and PDGFRβ, which can form homodimers (αα, ββ) or heterodimers (αβ). PDGF-AA binds exclusively to PDGFRαα, PDGF-AB binds to both PDGFRαα and αβ, PDGF-BB binds to all three receptor dimers (αα, αβ, ββ), PDGF-CC binds to PDGFRαα and αβ, and PDGF-DD binds solely to PDGFRββ. This differential binding enables isoform-specific activation of signaling pathways tailored to cellular contexts.[21][25][21] PDGF belongs to the PDGF/VEGF superfamily, sharing a characteristic cystine-knot fold in its growth factor domain—a structural motif involving three intramolecular disulfide bonds that stabilizes the dimeric ligand for receptor interaction. This evolutionary relationship with vascular endothelial growth factors (VEGFs), including VEGF-A, VEGF-B, VEGF-C, VEGF-D, VEGF-E, and placental growth factor (PlGF), underscores their common ancestry, despite VEGFs primarily promoting angiogenesis and vascular permeability through distinct receptor interactions.[21][23][23] Homologs of PDGF are conserved across the animal kingdom, with PDGF/VEGF-like factors identified in invertebrates such as Drosophila melanogaster (e.g., PVF1, PVF2, PVF3) and Caenorhabditis elegans, where they regulate cell migration and patterning during development. These non-mammalian variants retain the cystine-knot structure and similar receptor-binding properties, highlighting the ancient origins of the superfamily.[23][26][27]Biological Mechanisms
Receptor Interactions
Platelet-derived growth factor (PDGF) ligands exert their effects by binding to two closely related cell-surface receptor tyrosine kinases: platelet-derived growth factor receptor alpha (PDGFRα) and platelet-derived growth factor receptor beta (PDGFRβ). These receptors belong to the class III family of receptor tyrosine kinases and possess an extracellular ligand-binding domain composed of five immunoglobulin-like domains, a single transmembrane helix, and an intracellular tyrosine kinase domain. Upon ligand binding, PDGFRα and PDGFRβ can assemble into three possible dimeric complexes—homodimers αα and ββ, or the heterodimer αβ—each capable of transducing distinct signals depending on the ligand involved.[21] The binding specificities of the five dimeric PDGF isoforms (AA, BB, AB, CC, DD) to these receptor dimers vary, with PDGF-BB exhibiting the broadest and highest affinity for all three combinations, while other isoforms show more restricted preferences. For instance, PDGF-AA and PDGF-CC bind exclusively to PDGFRα homodimers and αβ heterodimers, PDGF-AB binds to αα and αβ, and PDGF-DD binds only to αβ and ββ. These affinities have been quantified through binding assays, with dissociation constants (K_d) typically in the low nanomolar range for high-affinity interactions, such as K_d ≈ 0.1–1 nM for PDGF-BB to PDGFRββ. The following table summarizes the key binding specificities:| PDGF Isoform | PDGFRαα | PDGFRαβ | PDGFRββ |
|---|---|---|---|
| AA | High | High | None |
| BB | High | High | High |
| AB | High | High | Low/None |
| CC | High | High | None |
| DD | None | High | High |