Induction of DNA synthesis in BALB/c 3T3 cells by serum components: Reevaluation of the commitment process

Serum contains a growth factor derived from platelets and also growth factors derived from platelet-poor plasma. Extracts of heated (100 degrees ) human platelets function synergistically with platelet-poor plasma to induce DNA synthesis in quiescent, density-inhibited BALB/c 3T3 cells. Platelet-poor plasma alone did not induce DNA synthesis. Cells exposed to platelet extracts became competent to enter the cell cycle, but the rate of entry into the S phase depended upon the concentration of platelet-poor plasma. The time required for the induction of this competent state was a function of the concentration of the platelet extract. A 2-hr exposure to 100 mug of the platelet extract at 37 degrees caused the entire cell population to become competent to enter the S phase. At 4 degrees or 25 degrees the cells did not become competent to synthesize DNA. The platelet extract-induced competent state was stable for at least 13 hr after removal of the platelet extract; however, in the absence of platelet-poor plasma, these competent cells did not progress through the cell cycle. The addition of an optimal concentration of platelet-poor plasma (5%) to these competent cells initiated cell cycle traverse with a rapid, first-order entry of cells into the S phase beginning 12 hr after addition of the plasma. The addition of a suboptimal concentration of the plasma (0.25%) did not increase the rate of cell entry into the S phase. Thus, the induction of DNA synthesis in quiescent BALB/c 3T3 cells can be resolved into at least two phases, controlled by different serum components: (i) competence, induced by the platelet-derived growth factor; and (ii) progression of competent cells into the cell cycle, mediated by factors in platelet-poor plasma.     

Angiogenesis in rat aortic rings stimulated by very low concentrations of serum and plasma

Arterial ring cultures are evaluated for assay of angiogenic activity in physiological fluids. A simplified image analysis algorithm involving public domain software is used to measure microvessels sprouting from rat aortic rings embedded in fibrin. Low concentrations (>1% v/v) of serum and plasma stimulate angiogenesis primarily from the adventitia in a dose-dependent manner. Half-percent serum is more potent than optimal concentrations of some of the important known growth factors, including platelet derived growth factor, vascular endothelial growth factor and basic fibroblast growth factor. Sera and plasmas obtained from rodent, bovine, porcine or human sources have comparable angiogenic activities. The angiogenic activity is found to be heat stable, non-dialyzable and on gel chromatography is co-eluted with albumin. However, purified albumin and the essential fatty acids linoleic and linolenic acids have no effect on angiogenesis. It is concluded that rat aortic rings sprout microvessels, primarily from the adventitia, when cultured with serum or plasma in concentrations below those that support cell culture. Accordingly, the aortic ring assay is a promising system to support isolation of angiogenic factors from blood and other physiological fluids.     

Serum contains a platelet-derived transforming growth factor.

High concentrations of fetal bovine serum induced colony formation in soft agar by anchorage-dependent, nontransformed mouse AKR-2B and rat NRK cells. The colony-stimulating activity in fetal bovine serum was precipitated by 45% saturated ammonium sulfate and migrated in molecular sieve chromatography as a single peak of activity in the 10,000-15,000 molecular weight range. The colony-stimulating activity was heat and acid stable and was destroyed by trypsin and dithiothreitol, indicating the activity is due to a polypeptide that requires disulfide bonds for biological activity. No competition for binding to the epidermal growth factor receptor was associated with the colony-stimulating activity. Isoelectric focusing revealed activity in the pI 4-5 range. The colony-stimulating activity in serum appeared to be of platelet origin because platelet-poor plasma and platelet-poor plasma-derived serum contained little activity, whereas acid/ethanol extracts of bovine and human platelets had potent colony-stimulating activity. Chromatography of platelet extracts on Bio-Gel P-60 revealed peaks of AKR-2B colony-stimulating activity in the 12,000 and 20,000 molecular weight ranges. The other biological and chemical properties of the platelet colony-stimulating activity were the same as those for the serum activity. The data indicate the presence in serum of a platelet-derived growth factor(s) with properties similar to those of the transforming growth factors.     

Platelets contain a peptide inhibitor of endothelial cell replication and growth.

Platelets release specific growth factors that stimulate division of aortic endothelial cells. Acidification or heating to 56 degrees C of platelet extracts is required to detect these factors. Nonheated platelet extracts inhibit endothelial growth. To determine if this inhibitory effect is due to a discrete chemical substance, a crude extract obtained by freezing and thawing human platelets was incubated with an endothelial cell mitogen and found to inhibit the endothelial cell incorporation of [3H]thymidine by greater than 95%. The inhibitor was partially purified by ion-exchange and gel-filtration chromatography. The partially purified material inhibited endothelial cell division as well as DNA synthesis. Inhibition caused by the partially purified material was concentration dependent, reversible, and not due to cytotoxicity. Further purification by heparin-Sepharose chromatography resulted in removal of greater than 95% of the contaminating substances without a loss of inhibitory activity. Physical characterization revealed that heparin-Sepharose-purified factor was heat labile, had a molecular size estimated at 35-40 kDa, and was stable between pH 4.5 and 9.0. Treatment with proteolytic enzymes destroyed all biologic activity, suggesting a peptide composition. These data support the conclusion that platelets contain a potent inhibitor of endothelial replication and growth. Since platelet deposition and release occur after endothelial denudation, release of this platelet-derived growth inhibitor may be an important regulator of reendothelialization that occurs after vessel injury.     

Sequential addition of platelet factor and plasma to BALB/c 3T3 fibroblast cultures stimulates somatomedin-C binding early in cell cycle.

Recent studies have shown that confluent cultured fibroblasts must be rendered competent by a factor contained in platelets before they can respond to plasma components and progress from G0 to S phase of the cell cycle. Somatomedin-C is one of the components of platelet-poor plasma necessary for cell cycle progression, but other factors present in somatomedin-C-deficient plasma are also required. Because both competence and progression factors contained in somatomedin-C-deficient serum modulate somatomedin-C action in fibroblasts, this study was undertaken to examine their possible influence on the binding of 125I-labeled somatomedin-C. Exposure of BALB/c 3T3 fibroblasts to high concentrations of platelet-derived growth factor alone prevented the time-dependent decrease in somatomedin-C binding which occurs in serum-free medium. Lower concentrations of the growth factor or 5% (vol/vol) platelet-poor plasma reduced but did not abolish this decrease. In contrast, sequential addition of platelet-derived growth factor followed by platelet-poor plasma resulted in a 75% increase in 125I-labeled somatomedin-C binding over basal levels during the first 2 hr. Binding increased by 125% when somatomedin-C-deficient platelet-poor plasma was substituted for normal platelet-poor plasma. The hormonal induction of somatomedin-C receptors appears to be a mechanism whereby peptide growth factors such as platelet-derived growth factor may condition the cell to respond optimally to somatomedin-C.     

Potent stimulation of vascular endothelial cell growth by differentiated 3T3 adipocytes.

3T3 cells that have undergone adipose differentiation in vitro secrete into the culture medium a potent growth stimulatory activity for bovine aortic endothelial cells. When medium containing 2% fetal calf serum, which does not support significant endothelial cell growth, is conditioned by 3T3-F442A adipocytes, the endothelial cells grow rapidly (doubling time, 24 hr) at a rate equal to the growth rate in 20% fetal calf serum. The potency of the conditioned medium is further shown by the fact that it can be diluted 1:5 with little apparent loss of activity and shows a half-maximal stimulation at 10 microliter/ml. Serum is not required for either the secretion of this mitogen by the adipocytes or its action on the endothelial cells, as shown by the fact that the latter are stimulated to divide in serum-free medium conditioned by the adipocytes. The growth stimulatory activity appears to be specific for vascular endothelial cells in that no other cell type examined, including vascular smooth muscle cells and pericytes, are significantly stimulated by medium conditioned by 3T3-F442A cells. Similarly, medium conditioned by no other cell type examined has more than 10% of the activity of medium conditioned by the adipocytes. The specificity and potency of the adipocyte-derived factor suggest that it may play a role in the vascularization of this tissue during development. Preliminary biochemical analysis indicates that the adipocyte factor is nondialyzable and is not inactivated by heat or proteases. The protease insensitivity distinguishes the adipocyte growth stimulatory activity from the low levels of activity secreted by fibroblasts and preadipocytes, suggesting that the adipocyte mitogen is a product specifically related to the differentiation process.     

Migration of porcine endothelial and smooth muscle cells in response to platelet-associated factors

A NeuroProbe multiwell microchemotaxis assembly has been used to study the migration of porcine aortic endothelial and smooth muscle cells in vitro in response to serum factors. Whole pig serum was found to be chemokinetic for endothelial cells but not smooth muscle cells. Chemokinetic activity was higher in platelet-enriched, plasma-derived serum (PDS) than in platelet-free PDS. The supernatant from washed, lysed pig platelets also showed chemokinetic activity both for endothelial and smooth muscle cells. Chromatography on Sephadex G-200 showed that the lysate contained a number of activities that act independently on the two cell types. The chemokinetic activity that we have detected appears to be distinct from the chemotactic effect of platelet-derived growth factor and platelet factor 4 on the basis of insensitivity to protamine sulfate and molecular size. The role of these platelet factors in pathological changes in vascular structure is currently under study.     

Paracrine effect of human vascular endothelial cells on human vascular smooth muscle cell proliferation: Transmembrane co-culture method

Summary To investigate the role endothelial cells have on underlying smooth muscle cell proliferation, human aortic vascular smooth muscle cells were co-cultured with human aortic endothelial cells at different cell densities, using a transmembrane co-culture method. Subconfluent endothelial cells subseeded at low (0.5 × 10⁴ cells/well) and medium densities (2.0 × 10⁴ cells/well) stimulated smooth muscle cell proliferation by 43 ± 14% (P < 0.01) and 39 ± 8% (P < 0.02), respectively. However, this stimulatory effect on smooth muscle cell proliferation was not evident in confluent endothelial cells subseeded at high cell density (8.0 × 10⁴ cells/well). Treatment of smooth muscle cells with trapidil, at 10^–6M, for anti-platelet derived growth factor (PDGF) effect or with endothelin-1 receptor blocker FR 139317, at 10^–6M failed to inhibit this stimulatory effect. These results imply that subconfluent human endothelial cells are able to exert a stimulatory effect on human smooth muscle cell proliferation, and that this endothelial paracrine growth effect may not be mediated by endothelin or PDGF.     

Human bone marrow stromal cell colonies: response to hydrocortisone and dependence on platelet-derived growth factor

Long-term bone marrow cultures are dependent on the formation in vitro of an adherent cell layer that supports hematopoiesis. We have grown bone-marrow-adherent cells, termed stromal colony-forming units, or CFU-ST, as isolated adherent colonies, and examined some of their growth requirements. Bone marrow mononuclear cells separated from aspirates by density centrifugation and cultured in medium supplemented with fetal calf serum or human plasma gave rise to adherent colonies (CFU-ST). An average of 23.4 +/- 2.1 (mean +/- SEM, n = 19) CFU-ST were produced by 10(5) bone marrow mononuclear cells. CFU-ST could not be cultured from similarly prepared peripheral blood mononuclear cells. The colonies were composed of spindle cells, flat cells, and fat-containing cells, with all three types often present in the same colony, suggesting derivation from a common progenitor. Cells were negative for nonspecific esterase and factor VIII antigen. Hydrocortisone added to the cultures at concentrations of 10(-7) M induced the formation of adipose cells in the center of one-third to one-half of the colonies but did not affect CFU-ST number. Human platelet-poor plasma and platelet-rich plasma were substituted for fetal calf serum in the medium. When all determinations for four experiments were averaged, platelet-rich plasma gave 17.8 +/- 1.2 (mean +/- SEM, n = 16) colonies, whereas platelet-poor plasma gave only 0.2 +/- 0.1 colonies (n = 15). When purified platelet-derived growth factor (PDGF) was added to platelet-poor plasma, growth of CFU-ST was enhanced, and a dose-response relationship was found between size of colonies and concentration of added PDGF. Granulocyte-macrophage colony stimulating factor added to cultures had no effect on the growth of CFU-ST.     

The presence and release of alpha 2-antiplasmin from human platelets

An antigen immunochemically indistinguishable from plasma alpha 2- antiplasmin, the primary plasmin inhibitor, was detected in human platelets. By radioimmunoassay, 33-114 ng alpha 2-antiplasmin antigen was quantitated in the detergent-soluble extract of 10(9) washed human platelets from 10 normal donors with a mean level of 62 +/- 24 ng/10(9) platelets. Plasma alpha 2-antiplasmin, either in the platelet suspending medium or on the surface of the platelets, could account for less than 8% of the antigen present in the platelet extracts. When stimulated with thrombin, the platelets released alpha 2-antiplasmin antigen without cell lysis, and greater than 85% of the alpha 2- antiplasmin antigen was released at a high thrombin dose. At a lower dose of thrombin, alpha 2-antiplasmin and platelet factor 4 were partially released without concomitant secretion of serotonin. No alpha 2-antiplasmin antigen was detected in extracts or red blood cells, polymorphonuclear leukocytes, and adherent and nonadherent mononuclear cells. Thus, the platelet is the only peripheral blood cell containing significant amounts of alpha 2-antiplasmin.     

Stimulation of fibroblast collagen and total protein formation by an endothelial cell-derived factor

We investigated the effect of medium conditioned by bovine aortic endothelial cells on collagen accumulation and total protein formation by human embryonic fibroblasts or bovine smooth muscle cells in cultures. The conditioned medium at a 1:10 dilution induced a twofold increase in collagen and total protein accumulation in fibroblast cultures. At low concentration (1:50 dilution), the conditioned medium stimulated collagen accumulation preferentially; at high concentration (1:10 dilution), overall protein synthesis also was increased. The increase in type I collagen accumulation was associated with an increase in the steady-state level of alpha 1 (I) mRNA for collagen. The conditioned medium increased the production of types I and III collagen without affecting the proportion of collagen types in both fibroblast and smooth muscle cell cultures. Partial purification of the endothelial cell-derived factor disclosed it to be a heat-stable protein with an apparent molecular weight of 8-10 kDa. The stimulation of protein formation by this substance was not inhibited by antibodies against transforming growth factor-beta or the insulinlike growth factor I receptor. The partially purified factor stimulated protein production without affecting fibroblast proliferation. This endothelial cell-derived protein may play a role in the remodeling of vascular connective tissue by stimulating collagen synthesis.     

Release of endothelium-derived relaxing factor from freshly harvested porcine endothelial cells

Vascular relaxation in rabbit aortic preparations induced by acetylcholine is endothelium-dependent. The nature of the endothelium-derived relaxing factor (EDRF) has not been ascertained because it is very labile (reported half-life 6-50 seconds). To obtain a stable source of EDRF, a system was developed in which the relaxing factor was continuously produced by freshly harvested porcine endothelial cells. Endothelial cells were collected from aortas by exposing the endothelial lining to collagenase 0.1%. Cells were washed and concentrated by repeated centrifugation to obtain a high cell count (7.2 X 10(6) cells/ml). Endothelium-deprived aortic strips from rabbits were incubated in these cells suspended in tissue culture medium and fetal calf serum. The strips were precontracted with histamine. Acetylcholine was added to induce EDRF release. Significant relaxation of endothelium-deprived aortic strips was observed. Superoxide dismutase, an enzyme known to protect EDRF against inactivation, caused further relaxation, which was inhibited by the addition of hemoglobin, an agent known to inhibit the relaxing action of EDRF. Even without the addition of acetylcholine, hemoglobin caused contraction of the denuded aortic strips in suspension of porcine endothelial cells, demonstrating spontaneous EDRF release. Hemoglobin had no effect in cell-free medium. Endothelial-cell-dependent relaxation occurred without attachment of endothelial cells to the endothelium-deprived aortic strips: when the cell suspension was replaced by cell-free medium, relaxation did not occur after acetylcholine. Scanning electron microscopy showed no attachment of endothelial cells to the subendothelial layer. It can be concluded that freshly harvested endothelial cells produce endothelium-derived relaxing factor with an without stimulation by acetylcholine.     

Coordinate control of 3T3 cell proliferation by platelet-derived growth factor and plasma components.

DNA synthesis and cell division were measured in Swiss mouse 3T3 cells cultured in different concentrations of cell-free plasma-derived serum and increasing amounts of a platelet-derived growth factor. In plasma-derived serum alone, the cells were quiescent and they were arrested in the Go/G1 phase of the cell cycle. Addition of a platelet-derived growth factor to quiescent cells maintained in plasma-derived serum stimulated both DNA synthesis and cell division. When plasma components were present at high concentration (5%, vol/vol), the amount of platelet factor added to the cultures determined the number of cell doublings. Plasma-derived molecules were required for the platelet factor to stimulate DNA synthesis and cell division in the maximal number of cells. In addition, plasma components had to be present for recently divided cells to respond to the platelet factor. When 3T3 cells were cultured in excess platelet factor and limiting amounts of plasma-derived serum (0.5%, vol/vol), the cells underwent one doubling and then ceased to proliferate. Addition of fresh plasma-derived serum to these cells induced a second round of cell division. Plasma components and the platelet-derived growth factor acted in a coordinate fashion to regulate the proliferation of Swiss 3T3 cells.     

Exposure to platelet-derived growth factor modulates the porcine aortic smooth muscle cell response to somatomedin-C

Platelet-derived growth factor (PDGF) is a potent mitogen for smooth muscle cells, but to detect maximal stimulation by PDGF, the cells must be incubated with plasma. Somatomedin-C (Sm-C), a peptide growth factor that is present in plasma, has been shown to interact with PDGF synergistically to stimulate DNA synthesis in cultured fibroblasts. These studies were designed to test the hypothesis that PDGF interacted with Sm-C to stimulate smooth muscle cell replication and to compare the response of this cell type to that of fibroblasts. When PDGF or Sm-C was added individually to smooth muscle cell cultures, each growth factor induced only minimal increases in [3H]thymidine incorporation into DNA (i.e. PDGF, 3,400 +/- 1,120 to 54,900 +/- 1,550 cpm; Sm-C, 3,400 +/- 1,120 to 10,950 +/- 980 cpm). In contrast, addition of increasing concentrations of Sm-C to cultures that were continuously exposed to PDGF in the presence of Sm-C-deficient plasma resulted in a synergistic increase in [3H]thymidine incorporation (3,400 +/- 1,120 to 54,500 +/- 1,800 cpm; P less than 0.001). To determine if Sm-C was required for smooth muscle cell replication, cultures were sequentially exposed to PDGF, followed by Sm-C-deficient plasma. The rate at which the non-Sm-C-exposed cells synthesized DNA was retarded compared to that of cells exposed to Sm-C-containing plasma; however, 68% nuclear labeling was present after 44 h of incubation. To exclude the possibility that some cellular secretory product was substituting for Sm-C, the medium was changed every 2 h and replaced by fresh Sm-C-deficient medium. Using these test conditions, exposure to PDGF and Sm-C-deficient plasma induced only 11% labelling. Readdition of pure Sm-C to this medium restored nuclear labeling to 82% at 44 h. Other variables that appeared to modulate the cellular response to Sm-C were culture density and simultaneous PDGF exposure. Sm-C and PDGF both appear to be potent mitogens for porcine aortic smooth muscle cells, and when added together to quiescent cultures, their effects are synergistic. Smooth muscle cells appear to require Sm-C to initiate DNA synthesis, and in its absence produce a Sm-like factor that can partially compensate for Sm-C deficiency and allow replication.     

Receptors for insulin-like growth factors and growth effects of multiplication-stimulating activity (rat insulin-like growth factor II) in rat embryo fibroblasts

Levels of multiplication-stimulating activity (MSA) in fetal rat serum are high (2-4 micrograms/ml), suggesting that MSA may have a role in fetal growth. We now demonstrate that fibroblasts derived from rat embryos (REFs) have specific MSA receptors and respond to MSA with increased DNA synthesis. Two types of insulin-like growth factor (IGF) receptors were demonstrated by competitive binding of radioiodinated MSA, IGF-I, and IGF-II and by chemical cross-linking of [125I]iodo-MSA and [125I]iodo-IGF-I to REFs. One type of receptor (mol wt, 260,000 under reducing conditions) did not interact with insulin, and another type of receptor (mol wt, 130,000, under reducing conditions) was recognized by insulin. Scatchard analysis of [125I]iodo-MSA binding data was consistent with one class of noninteracting binding sites. A biological response of MSA, increased DNA synthesis, was demonstrated with autoradiography in REFs. During a 16-hr incubation, DNA synthesis was stimulated by normal rat serum, and platelet-poor plasma plus platelet-derived growth factor (PDGF), but not by serum from hypophysectomized (hypox) rats or hypophysectomized (hypox) platelet-poor plasma plus PDGF. However, when MSA was added to hypox serum or to hypox platelet-poor plasma plus PDGF, DNA synthesis was stimulated to the level achieved by normal rat serum. By contrast, during a longer cell multiplication experiment, REFs grew equally well in normal or hypox rat serum, raising the possibility that REFs may produce a MSA-like factor.     

Effect of glucose and insulin on immunoreactive endothelin-1 release from cultured porcine aortic endothelial cells

We measured the release of immunoreactive endothelin-1 (IR-ET-1) by cultured porcine aortic endothelial cells under normoglycemic (5.5 mmol/L) and hyperglycemic (27.5 and 55 mmol/L) conditions. Compared with cells incubated in the presence of a normal glucose concentration, cells incubated in 27.5 mmol/L glucose medium released 52% less IR-ET-1, and those incubated in 55 mmol/L glucose medium released 54% less IR-ET-1. The observed effects of elevated glucose on IR-ET-1 release were both sugar-specific and not due to increased osmolarity. Fetal calf serum (FCS)-stimulated IR-ET-1 release in the presence of elevated glucose was also less than that in the presence of a normal glucose concentration. In addition, the effects of two hormones, insulin and insulin-like growth factor 1 (IGF-1), on IR-ET-1 release were examined. Both insulin and IGF-1 dose-dependently stimulated IR-ET-1 release. Twenty micrograms/mL insulin and 10(-8) mol/L IGF-1 increased IR-ET-1 release by 38% and by 44%, respectively. These results indicate that hyperglycemic condition results in reduction of IR-ET-1 release from cultured porcine aortic endothelial cells and that insulin and IGF-1 stimulate its release. The possible relevance of these observations to physiological regulation of ET-1 release in vivo and pathological processes in diabetes remains to be established.     

Effect of serum from patients with idiopathic thrombocytopenic purpura on cultured endothelial cell growth

This study examined the effects of platelet-derived serum factors from patients with idiopathic thrombocytopenic purpura (ITP) on human umbilical cord vascular endothelial cell growth and DNA synthesis. Vascular endothelial cell growth factors were identified in the platelet extracts, and platelet-derived vascular endothelial cell growth factors with molecular weights of less than 10,000 were also identified in the sera. These growth factors were stable even after 30 minutes of heat treatment at 56 degrees C and acid treatment. When ITP patient serum was added to vascular endothelial cells, their growth capacity and DNA synthesis capacity were lower than when serum from normal subjects or patients with ITP in remission was added. This indicated that in ITP, impairment of vascular integrity and abnormal repair of the vascular walls that accompany thrombocytopenia may be causes of the predisposition to hemorrhage. The decrease in ITP of platelet-derived vascular endothelial cell growth factors with molecular weights of less than 10,000 was hypothesized to be a contributory factor.     

Cultured endothelial cells produce a platelet-derived growth factor-like protein

The platelet-derived growth factor (PDGF) binds specifically to high-affinity receptors on the surface of bovine aortic smooth muscle cells and 3T3 cells. Conditioned medium from cultured bovine aortic endothelial cells (EC) prevents PDGF binding to these receptors in a dose-dependent manner at 4°C. The ¹²⁵I-labeled PDGF that is displaced by the conditioned medium shows no increase in trichloroacetic acid solubility or decrease in binding capability to fresh cells. The competitor activity was identified as a protein by ammonium sulfate precipitability and sensitivity to trypsin. The competitor protein also is found in the serum-free conditioned media from porcine aortic EC and human umbilical vein EC but not in media from bovine aortic smooth muscle cells, human neonatal foreskin fibroblasts, or the interleukin-producing thyoma cell line EL-4. The competitor protein, like PDGF, has no effect on the specific 4°C binding of either ¹²⁵I-labeled insulin to 3T3 cells or ¹²⁵I-labeled epidermal growth factor to human epidermoid A431 cells. Saturation curves of PDGF binding to smooth muscle cells that had been preincubated in the presence and absence of competitor indicate that the concentration for half-maximal binding of ¹²⁵I-labeled PDGF to its receptor (≈30 pM) is unchanged by the competitor, whereas the apparent number of available receptor sites or maximal level of binding is greatly diminished. The competitor activity produced by cultured human umbilical vein EC is completely inhibited by antiserum against pure human PDGF, whereas the same PDGF antiserum only partially inhibits the mitogenic activity of the conditioned media. In addition, ≈7-fold more crude endothelium-derived growth factor is required for half-maximal inhibition of ¹²⁵I-labeled PDGF binding as is required for half-maximal stimulation of DNA synthesis. These results suggest that EC secrete a PDGF-like protein that is biochemically distinct from the majority of EC-derived mitogenic activity.     

Effects of platelet-derived growth factor and somatomedin-C/insulin-like growth factor I on the deoxyribonucleic acid replication of fetal rat islets of Langerhans in tissue culture

The effects of platelet-derived growth factor (PDGF) on DNA replication and release of insulin and somatomedin-C/insulin-like growth factor I (SM-C/IGF-I) from cultured fetal rat islets have been studied. In medium containing 1% fetal calf serum and 16.7 mM glucose, both PDGF (2-10 ng/ml) and SM-C/IGF-I (100 ng/ml) stimulated DNA replication 2-fold. The growth stimulatory effects of the two peptides were additive. In the absence of serum, SM-C/IGF-I, but not PDGF, stimulated DNA replication. Under conditions of PDGF-stimulated DNA replication there was no increased release of either insulin or SM-C/IGF-I into the culture medium. An antibody against SM-C/IGF-I which inhibited SM-C/IGF-I-stimulated DNA replication did not affect PDGF-stimulated DNA replication. Similarly, an antibody against PDGF did not affect DNA replication stimulated by SM-C/IGF-I. It is concluded that PDGF stimulates islet cell DNA replication. This is the first demonstration of a tissue of nonmesodermal origin responding to PDGF. Stimulation of DNA replication appears to be independent of SM-C/IGF-I release, and furthermore, the results indicate that the islets do not produce PDGF-like substances themselves. It is suggested that PDGF is of importance for fetal islet development.     

Tumor necrosis factor type alpha stimulates human endothelial cells to produce granulocyte/macrophage colony-stimulating factor.

Tumor necrosis factor type alpha (TNF-alpha) is produced by monocytes and has been purified, sequenced, and cloned from the HL-60 cell line. Soluble products of monocytes stimulate endothelial cells to release multilineage hematopoietic colony-stimulating activity. To determine whether TNF-alpha could stimulate endothelial cells to produce these activities, we added recombinant human TNF-alpha to cultured human umbilical vein endothelial cells. Untreated endothelial cell conditioned medium and TNF-alpha-stimulated endothelial cell conditioned medium were tested for hematopoietic colony stimulating activity in colony-forming assays in methylcellulose. TNF-alpha stimulated growth factor production by endothelial cells. Fifth-passage human endothelial cells and multiply-passaged bovine aortic endothelial cells responded similarly to first-passage endothelial cells, indicating that the action of TNF-alpha on endothelial cells is direct and not due to contaminating lymphocytes or monocytes present in the first-passage cultures. To investigate the molecular basis for these findings, polyadenylylated RNA was prepared from the TNF-alpha-stimulated endothelial cells and probed for granulocyte-macrophage colony-stimulating factor and granulocyte colony-stimulating factor mRNA. Granulocyte-macrophage colony-stimulating factor, but not granulocyte colony-stimulating factor, message was detected. This finding suggests that at least some of the hematopoietic colony-stimulating activity released by the TNF-alpha-stimulated endothelial cells is granulocyte-macrophage colony-stimulating factor. These results demonstrate that a purified monocyte product can stimulate endothelial cells to produce the multilineage growth factor granulocyte-macrophage colony-stimulating factor and extend the role of this immunoregulatory protein to the regulation of hematopoiesis in vitro.