Harry N. Antoniades

Platelet-derived growth factor in idiopathic pulmonary fibrosis.

Fibrosis is a complex process involving an inflammatory reaction, fibroblast proliferation, and abnormal accumulation of interstitial collagens. Mononuclear cells are usually present in lung fibrosis. Activated monocytes and macrophages in culture have been shown to produce several growth factors including platelet-derived growth factor (PDGF). PDGF is a potent mitogen and chemoattractant for fibroblasts and smooth muscle cells and a stimulator of collagen synthesis. We have studied the expression of c-sis/PDGF-2 mRNA in lung tissues derived from five patients with idiopathic pulmonary fibrosis (IPF) and from four control individuals without IPF. Northern blot analysis of specimens obtained from four patients with IPF revealed the expression of the c-sis/PDGF-2 protooncogene. A control lung tissue without IPF did not express the c-sis protooncogene. In situ hybridization extended these studies demonstrating the expression of the c-sis mRNA in the five specimens with IPF but not in the four control specimens without IPF. The expression of c-sis mRNA was localized primarily in the epithelial cells. Invading alveolar macrophages also expressed c-sis mRNA. The expression of c-sis mRNA was accompanied by the expression of PDGF-like proteins in lung specimens with IPF but not in control lung specimens. These findings demonstrate the in vivo expression of the c-sis/PDGF-2 protooncogene and the production of PDGF-like proteins in the epithelial cells and macrophages of the fibrotic tissue. This localized and sustained production of PDGF-like mitogen may constitute an important contributing factor in the abnormal fibroblast proliferation and collagen production, events associated with pulmonary fibrosis.

Coexpression of platelet-derived growth factor (PDGF) and PDGF-receptor genes by primary human astrocytomas may contribute to their development and maintenance.

The present studies investigated the expression of the two PDGF genes (c-sis/PDGF-2 and PDGF-1) and the PDGF-receptor b gene (PDGF-R) in 34 primary human astrocytomas. Northern blot analysis demonstrated the coexpression of the c-sis/PDGF-2 protooncogene and the PDGF-R gene in all astrocytomas examined. The majority of the tumors also expressed the PDGF-1 gene. There was no correlation between the expression of the two PDGF genes. Nonmalignant human brain tissue expressed the PDGF-R and PDGF-1 genes but not the c-sis/PDGF-2 protooncogene. In situ hybridization of astrocytoma tissue localized the expression of the c-sis and PDGF-R mRNAs in tumor cells. Capillary endothelial cells also expressed c-sis mRNA. In contrast, nonmalignant human brain tissue expressed only PDGF-R mRNA but not c-sis/PDGF-2 mRNA. The coexpression of a potent mitogenic growth factor protooncogene (c-sis) and its receptor gene in astrocytoma tumor cells suggests the presence of an autocrine mechanism that may contribute to the development and maintenance of astrocytomas. The expression of c-sis mRNA in tumor cells but not in nonmalignant brain cells may serve as an additional diagnostic criterion for the detection of astrocytomas in small tissue specimen using in situ hybridization for the detection of c-sis mRNA and/or immunostaining for the recognition of its protein product.

Gel Matrix Vehicles for Growth Factor Application in Nerve Gap Injuries Repaired with Tubes: A Comparison of Biomatrix, Collagen, and Methylcellulose

The repair of nerve gap injuries with tubular nerve guides has been used extensively as an in vivo test model in identifying substances which may enhance nerve regeneration. The model has also been used clinical nerve repair. The objective of this study was to compare three different gel matrix-forming materials as potential vehicles for growth factors in this system. The vehicles included a laminin containing extracellular matrix preparation (Biomatrix), collagen, and a 2% methylcellulose gel. The growth factor test substance consisted of a combination of platelet-derived growth factor BB (PDGF-BB) and insulin-like growth factor I (IGF-I). An 8-mm gap in rat sciatic nerve was repaired with a silicone tube containing each of the vehicles alone or with a combination of each vehicle plus PDGF-BB and IGF-I. At 4 weeks after injury, the application of the growth factor combination significantly stimulated axonal regeneration when applied in methylcellulose or collagen, but not in Biomatrix. A similar trend was present between the vehicle control groups. By 8 weeks after injury, nerves repaired with methylcellulose as a vehicle had significantly greater conduction velocity than either collagen or Biomatrix. It was concluded that a 2% methylcellulose gel was the best of the three matrices tested, both in its effects on nerve regeneration and flexibility of formulation.

p53 expression during normal tissue regeneration in response to acute cutaneous injury in swine.

The present studies investigated the in vivo expression of the p53 suppressor gene and protein product in response to acute cutaneous injury in swine, along with the parallel expression of the c-sis/PDGF-B mitogen and its receptor beta (PDGF-R beta). p53 expression was shown to be suppressed during the period of active cellular proliferation in the injured tissue and to reemerge during the stages of healing. In contrast, c-sis/PDGF-B and PDGF-R beta were expressed during the early phase of active cellular proliferation and they were suppressed upon healing. This inverse relationship between mitogenic growth factors and p53 suggests the presence of well-controlled physiologic mechanisms that regulate in vivo the processes of normal tissue repair in response to injury. At the stages of tissue regeneration, these mechanisms include both the expression of growth factors that promote cell proliferation and the suppression of p53 that downregulates proliferation. At the stages of healing, the expression of the mitogenic growth factors is suppressed and that of p53 reemerges, reaching its peak at the time of complete epithelialization and healing of the injured tissue. These studies are the first to link the response of p53 protein to physiologic processes of tissue regeneration in vivo.

Cultured endothelial cells derived from the human iliac arteries.

SummaryCells derived from the endothelium of human iliac arteries were cultured in vivo. The cells were isolated, grown, and subcultured in HEPES buffered Medium 199 supplemented with 20% heat inactivated human whole blood serum, human alpha-thrombin, and commercial endothelial cell growth supplement derived from bovine brain. The cells were viable in culture for 8 to 10 passages at a split ratio of 1:3. After the 10th passage, the cells began to enlarge and their growth rate was reduced. No cultures were viable after the 12th passage. The cells were determined to be of endothelial origin by their morphology at confluence; their ultrastructural characteristics, including the presence of Weibel-Palade bodies; the production and release of factor VIII-related antigen; and by their maintenance of a surface that prevented platelet attachment. The cultured arterial endothelial cells released prostacyclin in response to challenge with thrombin and protamine sulfate but not in response to bradykinin or the platelet-derived growth factor. Although the cultures described in this report were derived from patients with varying degrees of atherosclerotic disease, there were no significant differences in morphological or physiological parameters among these cultures or in comparison with commonly studied cells derived from human umbilical veins.

Expression of platelet-derived growth factor (PDGF)-related transcripts and synthesis of biologically active PDGF-like proteins by human malignant epithelial cell lines.

Human malignant epithelial cell lines were analyzed for expression of platelet-derived growth factor (PDGF) genes. Of the 12 cell lines tested, 9, derived from breast, lung, gastric, and ovarian carcinomas, were found to express both PDGF-1 and PDGF-2 genes. The levels of both PDGF-1 and PDGF-2 transcripts were superinduced when these cells were treated with cycloheximide, an inhibitor of protein synthesis. These cells also released an activity that in studies with BALB-c/3T3 cells, inhibited binding of 125I-labeled PDGF and stimulated incorporation of [3H]thymidine. This stimulating activity was inhibited after reduction of the conditioned media by mercaptoethanol or after preincubation with antibodies to PDGF. Moreover, this activity was not affected by heat treatment. Immunoprecipitation studies revealed that breast, lung, and gastric carcinoma cells produced PDGF-like proteins that migrated as 30- and 32-kD species under nonreducing conditions and as 15- and 16-kD species under reducing conditions. In contrast, malignant cells of ovarian origin produced 14-16-kD PDGF-like proteins that were unchanged in mobility after reduction. As PDGF receptors were not detected on these malignant epithelial cells, the production of PDGF-like proteins may affect other cells in the microenvironment by paracrine mechanisms and may contribute to excessive cell proliferation, inflammatory reactions, and connective tissue remodeling seen in certain carcinomas.

The combination of platelet-derived growth factor-BB and insulin-like growth factor-I stimulates bone repair in adult Yucatan miniature pigs.

The combination of insulin‐like growth factor‐I and platelet‐derived growth factor‐BB has previously been shown to stimulate healing of soft tissue wounds and the formation of bone and ligament around teeth. The purpose of the present study was to evaluate the effects of platelet‐derived growth factor‐BB and insulin‐like growth factor‐I individually and in combination on the healing of osseous wounds. Four standardized cortical wounds were created in each tibia of 11 adult Yucatan miniature pigs. The wounds in one tibia per animal were treated with either purified recombinant human insulin‐like growth factor‐I, platelet‐derived growth factor‐BB, or both in a methylcellulose gel. The wounds in each contralateral tibia received placebo gel alone. Coded serial sections of each wound were evaluated by computer‐aided histomorphometry 21 days after surgery. The area and perimeter of the newly formed mineralized callus, the thickness of the total callus, and the percentage of mineralized tissue within the callus were significantly increased compared with the values of matched controls only in wounds treated with a combination of insulin‐like growth factor‐I and platelet‐derived growth factor‐BB. No significant differences in the measured parameters of callus formation were found in wounds treated with either insulin‐like growth factor‐I or platelet‐derived growth factor‐BB alone. Cartilage was present only in sites treated with insulin‐like growth factor‐I alone. These results suggest that the combination of platelet‐derived growth factor‐BB and insulin‐like growth factor‐I stimulates bone formation in wounds in long bones of adult animals and that these growth factors act via different pathways during the repair process.

Platelet-derived growth factor and malignant transformation

Investigations of human platelet-derived growth factor (PDGF), a potent mitogen for mesenchymal derived cells in culture [l-3], have provided a rational basis for the understanding of at least one mechanism involved in malignant transformation. PDGF is a heat-stable (loo”), cationic (isolectric point 9.8) polypeptide [4] that circulates in blood stored in the a-granules of platelets [5]. It is released from platelets into the serum during blood clotting, constituting the major polypeptide growth factor of serum. It is suggested that in uiuo PDGF is delivered during platelet degranulation at the site of injury where it participates in the process of wound healing by stimulating the proliferation and migration of connective tissue cells. We have shown recently that PDGF and the transforming protein of the simian sarcoma virus (SSV). an acute transforming retrovirus of primate origin, derive from the same or closely related cellular genes [6]. This conclusion is based on the demonstration that PDGF and the SSV transforming protein [7,8] share extensive amino acid sequence homology 16, 9, 101, have common antigenic determinants and structural conformation [ 111, and exert identical biological functions [12, 131. These findings suggest that the ability of the simian sarcoma virus to induce transformation derives from the incorporation of the PDGF gene within the retroviral genome. The resulting transforming one gene (v-sis) region within the retrovirus genome codes for a PDGF-like mitogen and is capable of inducing neoplastic transformation by the continuous production of this potent mitogen causing sustained cell proliferation. Consistent with the findings described above is the detection of v-sis-related messenger RNAs in human tumors of mesenchymal origin, such as glioblastomas, fibrosarcomas, and osteosarcomas [14]. Production of PDGF-like mitogen by these human malignant cells in culture has been reported [ 15-171. More recent studies have demonstrated that these cells synthesize, process, and release PDGF-like polvpeptides which are recognized by specific PDGFante the current understanding of PDGF structure and function; its role in the regulation of normal cell growth; and its link to malignant transformation.

Platelet-derived growth factor increases prostaglandin production and decreases epidermal growth factor receptors in human osteosarcoma cells.

Human platelet-derived growth factor (PDGF) stimulated the production of prostaglandin E2 (PGE2) by G-292 cells, a clonal line of human osteosarcoma cells. Half-maximal stimulation occurred with 9 ng/ml PDGF and maximal stimulation, 3-fold above control values, occurred with 40 ng/ml of the protein. Treatment of G-292 cells with 40 ng/ml PDGF also reduced the binding of iodinated epidermal growth factor (EGF) to the EGF receptor on G-292 cells. The effect was time-dependent, and EGF binding was reduced to 60% of control by 24-48 h. PDGF did not, however, compete directly for binding to the EGF receptor. The effects of PDGF and EGF on increased PGE2 production appeared to be additive at all concentrations tested, indicating that they may act through a common pathway, but not via the same membrane receptors.

Inhibitory Effects of “Bound” Insulin on Insulin Uptake by Isolated Tissues

“Bound” insulin preparations obtained from pooled human sera inhibited the uptake of crystalline insulin by isolated rat muscle (hemidiaphragm) and epididymal adipose tissue. When crystalline insulin alone was incubated with isolated tissues, its concentration in the incubating medium declined gradually, as shown by immunoassay of samples obtained at various intervals during the incubation. Addition of “bound” insulin preparations into the incubation medium greatly reduced the rate of disappearance of the crystalline insulin, “bound” insulin inhibiting the uptake of insulin by the tissues. The inhibition of insulin uptake by muscle, caused by “bound” insulin, was accompanied by inhibition of the biologic activity of insulin on this tissue. Preincubation of the isolated tissues with “bound” insulin, with whole fasting human sera from maturity-onset diabetics, or with crystalline insulin, prior to the addition of crystalline insulin, also resulted in significant inhibition of insulin uptake by these tissues. Synalbumin preparations obtained from fasting human sera by acid-ethanol extraction, as described by Vallance-Owen and collaborators, also inhibited the uptake of crystalline insulin by isolated muscle and adipose tissue and the biologic activity of insulin on muscle. Like “bound” insulin, the synalbumin extracts were inactive on isolated muscle and unreactive with anti-insulin antisera but exerted insulin-like activity on isolated adipose tissue. Their action when injected intraperitoneally into intact rats was similar to that of “bound” and crystalline insulin on the muscle and the adipose tissue. It is suggested that the synalbumin preparations obtained from sera by acid-ethanol extraction may contain the bulk of the serum “bound” insulin which contributes, at least in part, to the insulin inhibitory properties of these preparations.