World's Largest Resource for Cardiovascular Perfusion

Perfusion NewswireMobile ZonePlatelet-Derived Growth Factor: An Online Review

Platelet-Derived Growth Factor: An Online Review


Platelet-Derived Growth Factor
Contents:



  1. Overview


  2. Structural Information


  3. Binding Proteins


  4. Receptors


  5. Biological Activity


  6. References



Overview

Historically, it has been a goal of tissue culture
researchers to identify a substance(s) that was a universal growth or
maintenance factor for various cell lines and isolates (1). Early tissue
culture work demonstrated the superiority of serum over plasma in stimulating
the proliferation of fibroblasts in vitro (2). These observations
suggested that a factor released from platelets during degranulation was
probably responsible for this stimulatory activity. Subsequent investigations
clearly demonstrated that a factor released from platelets upon clotting was
capable of promoting the growth of various types of cells (3, 4). This factor
was subsequently purified from platelets and given the name platelet-derived
growth factor (PDGF) (1, 5). PDGF is now known to be produced by a number of
cell types besides platelets and it has been found to be a mitogen for almost
all mesenchymally-derived cells, i.e., blood, muscle, bone/cartilage,
and connective tissue cells (6).




Structural Information

Three forms of PDGF have been shown to
exist. Each form consists of a 30 kDa homo- or heterodimeric combination of two
genetically distinct, but structurally related, polypeptide chains designated A
and B. Although considerable work has been done on the primary structure of each
of the chains of human PDGF, the process has been complicated by the facts that
each is synthesized as a propeptide, that splice variants exist for the A chain,
and that C-terminal proteolytic processing apparently occurs for the B chain
(and possibly the A chain) (7-13).



The A chain is the product of a seven exon chromosomal 7 gene that gives
rise to one of two differentially-spliced variants. The “long”
variant, a prepropeptide of 211 amino acid (aa) residues, is synthesized with a
signal peptide of 20 aa residues, a propeptide sequence of 66 aa residues, and a
mature chain of 125 aa residues. In contrast, the “short” 196 aa
residue variant shows a 20 aa residue signal sequence, a 66 aa residue
propeptide, and a 16-18 kDa, 110 aa residue mature form (8, 9, 10). The
difference between the long and short forms is the result of alternative exon
usage, with the extended form utilizing exon 6 (18 aa residues), but not exon 7,
and the short form utilizing exon 7 (3 aa residues), but not exon 6 (8). The
difference between exon 6 utilization and exon 7 utilization is not, however,
limited to length. Within the 18 aa residues of exon 6 lies an approximately 10
aa residue sequence that signals cell retention. Failure to remove this
carboxyterminal peptide results in a failure to release freely circulating PDGF
(8, 14, 15). Retention under these circumstances implies binding to either
cell-surface glycosaminoglycans or intercellular matrix (16, 17). The short
version contains no retention sequence and is secreted into the circulation (16,
18). It is unclear whether any C-terminal processing of A chains occurs, but the
short variant’s 110 aa residue mature peptide ends with an arginine (8). This
suggests the possibility, as is the case for the B chain (see below), of a
carboxypeptidase-mediated C-terminal clipping to 109 aa residues with
equilization of A and B chain lengths for dimerization (8, 11, 13). No
definitive mechanism for C-terminus processing of the long form of the A chain
has been reported and it is not clear if this form is secreted (20, 21). One
potential N-linked glycosylation site exists in the mature A chain, but not the
B chain, and it is suggested to be utilized (10). In contrast to an earlier
suggestion that simultaneous expression of the long and short forms was a tumor
specific event, normal cells such as endothelial cell, macrophages, and
fibroblasts are now known to concurrently express both types of A chain, with
the short version clearly the most abundant (19).



The B chain is the product of a six exon gene on chromosome 22. The B chain
gene is now known to be identical to the human c-sis gene, the normal
human cell counterpart to the monkey v-sis (simian sarcoma)
virus gene (13). The protein coded for by c-sis is a 27 kDa, 241 aa
residue prepropeptide with a 20 aa residue signal sequence, 61 aa residue
propeptide, and a 16 kDa, 160 aa residue “mature” polypeptide (10,
11). C-terminal cleavage of the mature B chain is believed to occur, resulting
in a final mature product of 12 kDa and 109 aa residues (13). This is proposed
to occur in two stages with a trypsin-like cleavage of residues 111 to 160,
followed by a carboxypeptidase cleavage of the remaining arginine at residue 110
(11, 12, 13, 20). As with the long form of chain A, a retention sequence
approximately 10 aa residues in length has also been identified in the B chain
C-terminus. Failure to remove this peptide also results in B chain
glycosaminoglycan retention (14, 15, 17).



Dimerization of the A and B chains involves two interchain disulfide bonds,
and each chain is staggered (or overlaps the other) with a 6 or 7 aa residue
extension at either end. Within the 103 overlapping aa residues, the two chains
exhibit 51% sequence identity (10, 11).



Cells known to express PDGF are many and varied. Those which are reported to
express the A chain protein (both long and short variants) include fibroblasts
(22, 23), endothelial cells (23, 24), osteoblasts (25), platelets (26), vascular
smooth muscle cells (27), macrophages and Langerhans cells (28), and fetal
fibroblasts (29). Cells producing B chain protein include fetal fibroblasts
(29), endothelial cells (23, 24), platelets (26), macrophages (30), neurons (31,
32) and breast ductal epithelium (33). A number of cell types have also been
shown to express mRNA for the PDGF chains. In particular, A chain mRNA has been
found in type I astrocytes (34), embryonic endodermal respiratory epithelium
(35), renal mesangial cells (36), and osteoclasts and chrondrocytes (37), while
B chain mRNA has been localized to embryonic endodermal respiratory epithelium
(38), renal mesangial cells (36) and osteoblasts (37).



As with many growth factors, PDGF is now considered to be a member of a
larger family of factors (39). In addition to PDGF, this family includes the
homodimeric factors VEGF (vascular endothelial growth factor) and PlGF
(placental growth factor) (40, 41), VEGF/PlGF heterodimers (42), and CTGF
(connective tissue growth factor), a PDGF-like factor secreted by human vascular
endothelial cells (43) and fibroblasts (44). Relative to the PDGF isoforms,
VEGF shows distant analogy to PDGF-BB while PlGF corresponds to PDGF-AA. CTGF
shows little amino acid identity with PDGF A or B, but reacts with anitsera
produced against PDGF (45). Recently, the status of PDGF has been re-evaluated
based on analysis of its 3-dimensional structure. Along with NGF, TGF-beta and
glycoprotein hormones (human chorionic gonadotrophic), PDGF is now classified as
a member of the cysteine-knot growth factor superfamily. Each member of this
group occurs as a dimer and is characterized by six cysteines which link
together to form a “molecular knot”. The existence of this knot is
only revealed by 3-D analysis, making the criteria for admission to this family
unique among superfamilies (46). A number of reviews on PDGF are available, and
can be found in references 1, 7, 47-50.




Binding Proteins

An association exists between alpha-2
macroglobulin (alpha-2M) and the B chain-containing PDGF forms, AB and BB (51). alpha-2M
is a circulating 720 kDa homotetrameric glycoprotein produced by hepatocytes,
macrophages and astrocytes (52) whose most widely reported function is that of a
scavenger of proteases. Although PDGF does not interact with the region
associated with protease entrapment, it does bind to other alpha-2M sites not
influenced by activation (i.e., conformational change). PDGF-BB has been
noted to bind to both fast and slow alpha-2M and does so principally in a noncovalent
manner (53, 54). Significantly, the binding is reversible, and PDGF dissociation
is suggested to occur at either low pH or when equilibrium kinetics favor
dissociation, such as might be the case when PDGF is removed from the
circulation by binding to its own receptors (54). Binding studies for PDGF-AB
show the Kd to be (5 mM, while the Kd for PDGF-BB is reported to be 1 mM (51).
Functionally, it is unclear what the role is for B chain binding to alpha-2M. PDGF
binding to the slow form seems to result in its storage, as the alpha-2M receptor
binding motif(s) are not exposed, and the PDGF-alpha-2M complex simply circulates
(55). On the other hand, binding to fast or activated alpha-2M results in its rapid
clearance via alpha-2M receptors (half-life is 2-5 minutes), bringing the PDGF
molecule close to its own receptors and perhaps facilitating a secondary
PDGF-PDGFR interaction (56).




Receptors

Two distinct human PDGF receptor transmembrane binding
proteins have been identified, a 170 kDa, 1066 aa residue alpha-receptor (PDGFR alpha)
(55) and a 190 kDa, 1074 aa residue beta-receptor (PDGFR beta) (58). The two receptor
proteins are structurally related and consist of an extracellular portion
containing five immunoglobulin-like domains, a single transmembrane region, and
an intracellular portion with a protein-tyrosine kinase domain. A functional
PDGF receptor is formed when the two chains of a dimeric PDGF molecule each bind
one of the above receptor molecules, resulting in their approximation,
dimerization and activation. Between the two proteins, there is 44% overall
sequence identity. Within the extracellular domain, 30% of the aa residues are
identical (57). In addition, a 90 kDa soluble form of PDGFR alpha, consisting of the
extracellular segment of the alpha-receptor, has been found in cell culture medium
and in human plasma (59). The above two transmembrane receptors share
characteristics with other growth factor receptors, such as the M-CSF receptor,
c-kit, and the FGF receptor family (60). High-affinity binding of PDGF involves
dimerization of the receptors, forming either homodimers or heterodimers with
the alpha and beta receptors/chains (61-63). Although it appears that each subunit of
dimeric PDGF binds to one receptor monomer, it is unclear if these PDGF subunits
need to be covalently linked. Recent evidence suggests noncovalently linked B
chains are able to activate the PDGFR (64).



PDGFR alpha binds each of the three forms of PDGF dimers with high affinity. Its
Kd for PDGF-BB is 0.5 nM, for AB 0.1 nM, and for AA 0.2 nM (57). Although PDGFR beta
binds both PDGF-BB and PDGF-AB with high affinity (Kd = 0.5 pM and 1-2.5 nM
respectively), it has no reported binding to PDGF-AA (60, 65). The apparent
high-affinity binding of the AB dimer to the beta-receptor must be interpreted with
caution, however. Although PDGF-AB can bind to mutant 3T3 cells displaying only
beta-receptors, it requires 100-fold more PDGF-AB to dimerize the beta-receptors and
activate the cells than is required for cells also displaying alpha-receptors (65).
This required concentration is probably not physiologically relevant (65).



Cells known to express only alpha-receptors include oligodendroglial
progenitors, liver endothelial cells and mesothelium (60), plus platelets (66).
Cells expressing only beta-receptors include CNS capillary endothelium, neurons and
Ito (or fat storing) cells of the liver (60), plus monocytes/macrophages (67).
Cells showing coincident expression of alpha and beta receptors include smooth muscle
cells, fibroblasts (60), and Schwann cells (68).



Receptor binding by PDGF is known to activate intracellular tyrosine kinase,
leading to autophosphorylation of the cytoplasmic domain of the receptor as well
as phosphorylation of other intracellular substrates (69). This reaction is
described as one in trans, i.e., the two receptor molecules of
the receptor dimer phosphorylate each other. Specific substrates identified with
the beta-receptor include Src, GTPase Activating Protein (GAP), phospholypase Cg
(PLCg) and phosphotidylinositol 3-phosphate (69, 70). Both PLCg and GAP seem to
bind with different affinities to the a- and beta-receptors, suggesting that the
particular response of a cell depends on the type of receptor it expresses and
the type of PDGF dimer to which it is exposed (70, 71). In addition to the
above, a non-tyrosine phosphorylation-associated signal transduction pathway can
also be activated that involves the zinc finger protein erg-1 (early growth
response gene 1) (72). Recent reviews which deal with PDGF receptors and
receptor signalling can be found in references 60, 69, 73 and 742.




Biological Activity

Because there are differences between cells
relative to the amounts of alpha- and beta-receptors that they express, and because of
the variability in PDGF isomer binding to receptors, there is a tremendous range
of possibilities for biological responses by PDGF. This is reflected in at least
four experimental systems where different isoforms of PDGF elicit different
results. Vascular smooth muscle cells (SMC) and fibroblasts are both known to
express both the alpha- and beta-receptors. On SMC, PDGF-AA initiates cellular
hypertrophy (increased protein synthesis), while BB induces hyperplasia
(mitosis) (75) On fibroblasts, the BB isoform initiates chemotaxis, while AA
inhibits chemotaxis (76). On dopaminergic neurons, PDGF-AA promotes embryonic
neuron fiber development, while BB serves only as a survival or maintenance
factor (77). Finally, within the developing lung, the BB isoform regulates the
growth and number of respiratory tubule epithelial cells, while the AA isoform
directs the actual formation of branches arising from the respiratory tubules
(35, 38). Clearly, the interpretation of any experimental study must take into
account the cell type used and isoform applied.



In general, PDGF isoforms are potent mitogens for connective tissue cells,
including dermal fibroblasts (78), arterial smooth muscle cells (79),
chondrocytes (80) and some epithelial and endothelial cells (81, 82). In
addition to its activity as a mitogen, PDGF is chemotactic for fibroblasts and
smooth muscle cells, cells which also respond mitogenically to PDGF (76, 79),
and for neutrophils and mononuclear cells, cells for which PDGF is not a mitogen
(83). There is a considerable body of evidence to indicate that PDGF derived
from macrophages, acting as a chemotactic and mitogenic agent for smooth muscle
cells, contributes to the myointimal thickening of arterial walls characteristic
of atherosclerosis (84). Other reported activities for PDGF include the
stimulation of granule release by neutrophils and monocytes (85), the
facilitation of steroid synthesis by Leydig cells (86), a stimulation of
neutrophil phagocytosis (87), a modulation of thrombospondin expression and
secretion (88), an upregulation of ICAM-1 in vascular smooth muscle cells (89),
and the transient induction of T cell IL-2 secretion, accompanied by a
down-regulation of IL-4 and IFN-gamma production which allow clonal expansion of
antigen-activated B and T helper lymphocytes prior to differentiation (90).
PDGF also appears to be ubiquitous in neurons throughout the CNS, where it is
suggested to play an important role in neuron survival and regeneration, and in
mediation of glial cell proliferation, differentiation and migration (77,
91-94).




References


 


  1. Ross, R. & A. Vogel (1978) Cell 14:203.

  2. Balk, S.D. et al. (1973) Proc. Natl. Acad. Sci. USA 70:675.

  3. Ross, R. et al. (1974) Proc. Natl. Acad. Sci. USA 71:1207.

  4. Kohler, N. & A. Lipton (1974) Exp. Cell Res. 87:297.

  5. Antoniades, H.N. et al. (1975) Proc. Natl. Acad. Sci. USA 72:2635.

  6. Raines, E.W. (1993) in Biology of Platelet-Derived Growth Factor,
    Westermark, B. and C. Sorg, eds., Basel, Karger, p. 74.

  7. Heldin, C-H. & B. Westermark (1989) Br. Med. Bull. 45:453.

  8. Bonthron, D.T. et al. (1988) Proc. Natl. Acad. Sci. USA 85:1492.

  9. Rorsman, F. et al. (1988) Mol. Cell. Biol. 8:571.

  10. Betsholtz, C. et al. (1986) Nature 320:695.

  11. Heldin, C-H. (1993) in Biology of Platelet-Derived Growth Factor,
    Westermark, B. and C. Sorg, eds., Basel, Karger, p. 1.

  12. Collins, T. et al. (1985) Nature 316:748.

  13. Johnsson, A. et al. (1984) EMBO J. 3:921.

  14. LaRochelle, W.J. et al. (1991) Genes Dev. 5:1191.

  15. Ostman, A. et al. (1991) Cell Reg. 2:503.

  16. Pollock, R.A. & W.D. Richardson (1992) Growth Factors 7:267.

  17. Kelly, J.L. et al. (1993) J. Cell Biol. 121:1153.

  18. Beckmann, M.P. et al. (1988) Science 241:1346.

  19. Matoskova, B. et al. (1989) Mol. Cell. Biol. 9:3148.

  20. Ostman, A. et al. (1992) J. Cell Biol. 118:509.

  21. Andersson, M. et al. (1994) J. Biol. Chem. 269:926.

  22. Paulsson, Y. et al. (1987) Nature 328:715.

  23. Pierce, G.F. et al. (1995) J. Clin. Invest. 96:1336.

  24. Lindner, V. & M.A. Reidy (1995) Am. J. Pathol. 146:1488.

  25. Rydziel, S. et al. (1994) Endocrinology 134:2541.

  26. Hart, C.E. et al. (1990) Biochemistry 29:166.

  27. Stouffer, G.A. & G.K. Owens (1994) J. Clin. Invest. 93:2048.

  28. Reuterdahl, C. et al. (1993) J. Clin. Invest. 91:2065.

  29. Buch, S. et al. (1994) Exp. Cell Res. 211:142.

  30. Lindmark, G. et al. (1993) Lab Invest. 69:682.

  31. Sasahara, M. et al. (1991) Cell 64:217.

  32. Sasahara, M. et al. (1995) Mol. Brain Res. 32:63.

  33. Coltrera, M.D. et al. (1995) Cancer Res. 55:2703.

  34. Richardson, W.D. et al. (1988) Cell 53:309.

  35. Souza, P. et al. (1995) Development 121:2559.

  36. Bhandari, B. et al. (1994) Biochem. J. 297:385.

  37. Andrew, J.G. et al. (1995) Bone 16:455.

  38. Souza, P. et al. (1994) Development 120:2163.

  39. Tischer, E. et al. (1989) Biochem. Biophys. Res. Commun. 165:1198.

  40. Neufeld, G. et al. (1994) Prog. Growth Factor Res. 5:89.

  41. Maglione, D. et al. (1991) Proc. Natl. Acad. Sci. USA 88:9267

  42. Cao, Y. et al. (1996) J. Biol. Chem. 271:3154.

  43. Igarishi, A. et al. (1992) J. Dermatol. 19:642.

  44. Igarishi, A. et al. (1995) J. Invest. Dermatol. 105:280.

  45. Bradham, D. M. et al. (1991) J. Cell Biol. 114:1285.

  46. Sun, P.D. & D.R. Davies (1995) Annu. Rev. Biophys. Biomol. Struct.
    24:269.

  47. Deuel, T.F. (1987) Annu. Rev. Cell Biol. 3:443.

  48. Ross, R. et al. (1986) Cell 46:155.

  49. Meyer-Ingold, W. & W. Eichner (1995) Cell Biol. Int. 19:389.

  50. Westermark, B. & C-H. Heldin (1993) Acta Oncol. 32:101.

  51. Bonner, J.C. & A.R. Osornio-Vargas (1995) J. Biol. Chem. 270:16236.

  52. Sottrup-Jensen, L. (1989) J. Biol. Chem. 264:11539.

  53. Bonner, J.C. et al. (1995) J. Biol. Chem. 270:6389.

  54. Bonner, J.C. et al. (1992) J. Biol. Chem. 267:12837.

  55. Bonner, J.C. (1994) Ann. N.Y. Acad. Sci. 737:324.

  56. Crookston, K.P. et al. (1994) J. Biol. Chem. 269:1533.

  57. Claesson-Welsh, L. et al. (1989) Proc. Natl. Acad. Sci. USA 86:4917.

  58. Gronwald, R.G.K. et al. (1988) Proc. Natl. Acad. Sci. USA 85:3435.

  59. Tiesman, J. & C.E. Hart (1993) J. Biol. Chem. 268:9621.

  60. Claesson-Welsh, L. (1993) in Biology of Platelet-Derived Growth
    Westermark, B. and C. Sorg, eds., Basel, Karger, p. 31.

  61. Heldin, C-H. et al. (1989) J. Biol. Chem. 264:8905.

  62. Bishayee, S. et al. (1989) J. Biol. Chem. 264:11699.

  63. Seifert, R.A. et al. (1989) J. Biol. Chem. 264:8771.

  64. Kenney, W.C. et al. (1994) J. Biol. Chem. 269:12351.

  65. Seifert, R.A. et al. (1993) J. Biol. Chem. 268:4473.

  66. Vassbotn, F.S. et al. (1994) J. Biol. Chem. 269:13874.

  67. Inaba, T. et al. (1993) J. Biol. Chem. 268:24353.

  68. Eccleston, P.A. et al. (1993) Dev. Biol. 155:459.

  69. Claesson-Welsh, L. (1994) Prog. Growth Factor Res. 5:37.

  70. Heidaran, M.A. et al. (1993) J. Biol. Chem. 268:9287.

  71. Eriksson, A. et al. (1995) J. Biol. Chem. 270:7773.

  72. Mundschau, L.J. et al. (1994) J. Biol. Chem. 269:16137.

  73. Claesson-Welsh, L. (1994) J. Biol. Chem. 269:32023.

  74. Hart, C.E. & D.F. Bowen-Pope (1990) J. Invest. Dermatol. 94:53S.

  75. Inui, H. et al. (1994) J. Biol. Chem. 269:30546.

  76. Siegbahn, A. et al. (1990) J. Clin. Invest. 85:916

  77. Gaicobini, M.M.J. et al. (1993) Neuroscience 57:923.

  78. Lubinus, M. et al. (1994) J. Biol. Chem. 269:9822.

  79. Bornfeldt, K.E. et al. (1994) J. Clin. Invest. 93:1266.

  80. Guerne, P-A. et al. (1994) J. Cell. Physiol. 158:476.

  81. Brewitt, B. & J.I. Clark (1988) Science 242:777

  82. Kuwabara, K. et al. (1995) Proc. Natl. Acad. Sci. USA 92:4606.

  83. Deuel, T.F. et al. (1982) J. Clin. Invest. 69:1046.

  84. Ross, R. et al. (1990) Science 248:1009.

  85. Tzeng, D.Y. et al. (1985) Blood 66:179.

  86. Risbridger, G.P. (1993) Mol. Cell. Endocrinol. 97:125.

  87. Wilson, E. et al. (1987) Proc. Natl. Acad. Sci. USA 84:2213.

  88. Majack, R.A. et al. (1987) J. Biol. Chem. 262:8821.

  89. Morisaki, N. et al. (1994) Biochem. Biophys. Res. Commun. 200:612.

  90. Wiedmeier, S.E. et al. (1994) J. Immunol. 152:3417.

  91. Smits, A. et al. (1993) Eur. J. Neurosci. 5:986.

  92. Yeh, H-J. et al. (1991) Cell 64:209.

  93. Gogate, N. et al. (1994) J. Neurosci. 14:4571.

  94. Gard, A.L. et al. (1995) Development 121:2187.

 









Leave a Reply