We have purified two growth factors from dogfish brain and retina by using their binding affinity for heparin-Sepharose. These growth factors were eluted at 1 and 2 M NaCl similarly to those purified from bovine brain or retina. Their mitogenic activity was assayed in vitro on the same mammalian cells: bovine lens epithelial cells or human fibroblasts. All these data seem to indicate that these growth factors belong to the families of other well defined mammalian growth factors: EDGF I, brain basic FGF, AGF II, on the one hand and EDGF II, brain acidic FGF, AGF I, ECGF, on the other. Thus, these growth factors have been widely distributed during evolution and retain at least a conservative sequence to stimulate cell proliferation, in mammalian cells.

Endothelial cell growth factor (ECGF) stimulates vascularization, however its relatively short half-life requires this angiogenic factor to be frequently administrated by non-specific and uncontrolled methods. This work describes the use of biocompatible chitosan, a polysaccharide having structural similarity to glycosaminoglycans, -albumin microspheres, as well as its fiber form, as a potential delivery system for the controlled and localized release of ECGF. Chitosan-albumin microspheres (400-600 microns) and fibers, formed in 0.5 M sodium hydroxide-methanol solution were incubated with ECGF. In vitro release was performed in PBS at 37 degrees C, under constant stirring. In vivo experiments were realized by implanting ECGF loaded matrices subcutaneously into rat groin fascia. After an initial ECGF burst of 1.32-1.62 mg (22-27%) within the first 2 hours, a daily release of 120-420 micrograms (2-7%) during the first, and 60-240 micrograms (1-4%) during the second week was observed from M(r) 70.000, 750.000, and 2,000.000 chitosan containing microspheres of 6 mg/ml loading. ECGF release rate of < 30 micrograms (0.5%)/day was maintained during the third week of experiments. By the increase in ECGF loading (12 mg/ml polymer), while the amount of release increased, percent release decreased. Chitosan-albumin fibers gave a ECGF release rate nearly similar to microspheres, and in vivo studies demonstrated a high degree of neovascularization for both types of implants, starting from 7 day-post implantation. Control animals that received ECGF injection did not show any significant neovascularization, after same period of time.

A two-site enzyme immunoassay for gliostatin (GLS)/platelet-derived endothelial cell growth factor (PD-ECGF) has been developed. The detection limit of gliostatin/PD-ECGF was 30 pg/well, and the optimal assay range was 0.1 to ng/well. This assay system enabled us to confirm the immunochemical identity of both factors and to detect immunoreactive gliostatin/PD-ECGF (IR-GLS/PD-ECGF) in human biological body fluids. The age-related analysis from newborn to 69 years revealed that the serum IR-GLS/PD-ECGF level was high in infants younger than 1 year old (1.8 ng/ml) and in the 20-year-old age group (1.8 ng/ml), and highest in the umbilical cord blood (2.1 ng/ml). Curiously high concentrations were detected in saliva with a significant sex difference (11.3 ng/ml for males and 48.7 ng/ml for females), and in synovial fluids (3.7 ng/ml). A number of human tumor cells, gastric cancer cells, MKN-74, neuroblastoma cells, GOTO, as well as epidermoid carcinoma cells, A431, were found to produce a significant amount of IR-GLS/PD-ECGF (0.2 to 21.8 ng/mg protein), and some of them secreted the IR-GLS/PD-ECGF in the conditioned medium (approximately 0.5 ng/ml). The enzyme immunoassay system is sufficiently sensitive for the basic and clinical study of gliostatin/PD-ECGF in human body fluids, tissues and organs.

We examined the expression of four angiogenesis factors (vascular endothelial growth factor (VEGF), interleukin-8 (IL-8), platelet-derived endothelial cell growth factor (PD-ECGF) and basic fibroblast growth factor (bFGF)) in five human ovarian cancer cell lines by Northern blot analysis and immunohistochemical staining. The cancer cells were grown as a subconfluent monolayer culture, a multicellular aggregate (spheroid) in a three-dimensional culture, and a nude mouse-transplanted subcutaneous tumor in order to simulate the cellular conditions of ovarian cancers in peritonitis carcinomatosa, i.e. floating single tumor cells, multicellular aggregates and peritoneally implanted tumors. In each cell line, the expression of VEGF was detected in a monolayer culture and obviously enhanced in a three-dimensional culture. IL-8 was expressed in two of five cultured cell lines, but neither PD-ECGF nor bFGF was detected. Each cell line-derived transplanted tumor expressed immunohistochemical products of the four angiogenesis factors examined. These observations were confirmed by surgical specimens and suggested that ovarian cancer cells expressed different kinds and/or doses of angiogenesis factors depending on the form of the changed tumor cells during peritoneal implant formation.

The Hollow Fibre Assay (HFA) is usually applied as an early in vivo model for anti-cancer drug screening, but is potentially an excellent model for short-term in vivo pharmacodynamic studies. We used the model to study the in vivo role of thymidine phosphorylase/platelet-derived endothelial cell growth factor (TP/PD-ECGF) in the cytotoxicity and pharmacodynamics of TAS-102 in colon cancer cells. TAS-102 is a new oral drug formulation, which is composed of trifluorothymidine (TFT) and thymidine phosphorylase inhibitor (TPI), which prevents TFT degradation. We compared the activity with Xeloda (capecitabine), which is activated by TP into 5FU. Hollow fibres filled with human Colo320 or Colo320TP1 colorectal cancer cells with deficient or high TP expression, respectively, were implanted subcutaneously (s.c.) at both flanks of BALB/c mice. The mice were treated orally over 5 days with TAS-102, TFT alone, 5'DFUR+/-TPI or capecitabine at their maximum tolerated dose (MTD). The cells were retrieved from the fibres and assayed for growth (MTT assay), cell cycle distribution (flow cytometry) and apoptosis induction (FragEL method). TAS-102 induced considerable growth inhibition (50%, P<0.01) to both cell lines, which was completely abolished in the absence of TPI. Capecitabine and its metabolite 5'DFUR reduced proliferation of Colo320TP1 cells in the fibres significantly (down to 25-40%), but much less in Colo320 cells, whereas addition of TPI reduced the effect of 5'DFUR, although not completely. These differences in cytotoxic effects were reflected in the pharmacodynamic evaluation. TAS-102 induced a G2M-phase arrest (from 25 to 40%) and apoptosis (>8-fold), which was more pronounced in Colo320 than in Colo320TP1. Again, omission of TPI neutralised the effect of TAS-102. Similarly, 5'DFUR and capecitabine induced a significant G2M-phase arrest (up to 45%) in the Colo320TP1 cell line, but less pronounced in the parental Colo320. Addition of TPI to 5'DFUR reduced this effect to control levels. Also induction of apoptosis was reduced in the presence of TPI. The data demonstrated that the HFA is excellently suited for studying short-term pharmacodynamic effects of fluoropyrimidines in vivo. TAS-102 is only effective in inducing cytotoxicity when systemic TPI is present, but acts against both low and high TP expressing colon cancer cells, while 5'DFUR needs cellular TP to exert significant activity.

Thymidine phosphorylase (TP), also known as platelet-derived endothelial cell growth factor (PD-ECGF), has been implicated in bladder cancer angiogenesis and invasion. However, the molecular basis of its role in invasion remains unclear. We investigated the expression of TP and 10 invasion-related genes in bladder cancers from 72 randomly selected patients by real-time two-step RT-PCR assay. We found that the expression levels of TP, MMP-9, uPA, and MMP-2 were significantly higher in invasive tumors than those in superficial tumors. Also, the expression level of TP significantly correlated with that of uPA, MMP-1, MMP-9, PAI-1 and VEGF. KK47/TP cells, bladder cancer cells that overexpress TP, had a higher expression of MMP-7 and MMP-9 than KK/CV cells that express lower level of TP in hypoxic condition. PC/TP cells, prostate cancer cells that overexpress TP, also had a higher expression of MMP-1 and MMP-7 than PC/CV cells that express no detectable TP. Taken together these data indicate that TP enhances the invasion of tumor cells through the induction of invasion-related genes.

To determine the potential of growth, invasion and metastasis of uterine endometrial cancer cells associated with neovascularization, the expressions of platelet-derived endothelial cell growth factor (PD-ECGF) and its mRNA in uterine endometrial cancers and in normal uterine endometria as controls were determined and the relationship between their expressions and histological grades, grades of myometrial invasion and clinical stages of uterine endometrial cancers was analyzed. The levels of PD-ECGF were significantly higher in uterine endometrial cancers of well-differentiated grade (G1) with invasion to < or =1/2 myometrium (B) and of stage 1 than in those of moderately and poorly differentiated grades (G2 and G3, respectively) limited to endometrium (A) and with invasion to >1/2 myometrium (C) and of stages II and III/IV and in normal uterine endometria. There was no significant difference in the levels between uterine endometrial cancers of G2 and G3, A and C, or stages II and III/IV and normal uterine endometria. Therefore, the active availability of PD-ECGF might contribute to the acceleration of angiogenic activity in the early process of invasion of well-differentiated uterine endometrial cancers.

The enzyme, thymidine phosphorylase (dThdPase) is identical to platelet-derived endothelial cell growth factor (PD-ECGF), which acts as a potent angiogenic factor. The present study immunohistochemically examined the expression of dThdPase in human colorectal mucosa, adenomas and carcinomas, as well as six cultured colorectal carcinoma cell lines, in terms of intratumoral microvessel density (IMVD) and P53 expression. Thymidine phosphorylase was observed in lymphocytes, fibroblasts and macrophages, as well as smooth muscle cells and Schwann cells in the peripheral nerve fibers. The dThdPase-positive stromal cells apparently outnumbered the normal epithelial cells, adenoma and carcinoma cells with dThdPase. Weak but obvious cytoplasmic immunoreactivity was noted in a few normal colonic epithelia, predominantly the upper surface area, while a few adenoma cells showed weak nuclear immunostaining for dThdPase in six (24%) of the 25 colonic adenomas. Expression of dThdPase was noted in 33 (73.3%) of the 45 Dukes A and B, 14 (51.9%) of the 27 Dukes C and 14 (56.0%) of the 25 Dukes D carcinomas. The mean IMVD was 84.0 +/- 26.2 in the 36 dThdPase-negative carcinomas and 97.9 +/- 31.6 in the 61 dThdPase-positive carcinomas, the value being significantly higher in the latter group (P < 0.05). The frequency of dThdPase expression was significantly lower in the P53-negative carcinomas than in the positive carcinomas (P < 0.05). Western blot analysis showed the highest expression of dThdPase in LoVo carrying the wild-type p53 gene, followed by Colo201, Colo320, DLD-11 and WiDr carrying the mutated gene. These results indicate that: (i) the main source of dThdPase is stromal cells, including lymphocytes and macrophages in both colorectal normal and carcinoma tissues; (ii) dThdPase may take part in the induction of intratumoral microvessels, regardless of tumor stage; and (iii) expression might be modulated by not only P53 but also other molecules.

The overall survival of patients with pancreatic ductal adenocarcinoma is extremely low. Although gemcitabine is the standard used chemotherapy for this disease, clinical outcomes do not reflect significant improvements, not even when combined with adjuvant treatments. There is an urgent need for prognosis markers to be found. The aim of this study was to analyze the potential value of serum cytokines to find a profile that can predict the clinical outcome in patients with pancreatic cancer and to establish a practical prognosis index that significantly predicts patients' outcomes. We have conducted an extensive analysis of serum prognosis biomarkers using an antibody array comprising 507 human cytokines. Overall survival was estimated using the Kaplan-Meier method. Univariate and multivariate Cox's proportional hazard models were used to analyze prognosis factors. To determine the extent that survival could be predicted based on this index, we used the leave-one-out cross-validation model. The multivariate model showed a better performance and it could represent a novel panel of serum cytokines that correlates to poor prognosis in pancreatic cancer. B7-1/CD80, EG-VEGF/PK1, IL-29, NRG1-beta1/HRG1-beta1, and PD-ECGF expressions portend a poor prognosis for patients with pancreatic cancer and these cytokines could represent novel therapeutic targets for this disease.

Thymidine phosphorylase (TP, EC 2.4.2.4), an enzyme involved in pyrimidine salvage pathway, is identical to platelet-derived endothelial cell growth factor (PD-ECGF) and gliostatin. It is extremely upregulated in a variety of solid tumours. The TP amplification is associated with concomitant overexpression of many angiogenic factors such as matrix metalloproteases (MMPs), interleukins (ILs), vascular endothelial growth factor (VEGF) etc., resulting in promotion of angiogenesis and cancer metastasis. In addition, overshooting TP level protects tumour cells from apoptosis and helps cell survival. Thus, TP is identified as a prime target for developing novel anticancer therapies. Pioneering research activities investigated a large number of TP inhibitors, most of which are pyrimidine or purine analogues. Recently, an array of structurally diverse non-nucleobase derivatives was designed, synthesized and established as promising TP inhibitors. This review, following an outline on the TP structure and functions, gives an overview of the recent advancement of various non-nucleobase TP inhibitors as novel anti-cancer agents.

Gliostatin/platelet-derived endothelial cell growth factor (GLS/PD-ECGF) has a potential for arthritogenic action. The aim of this study was to examine whether measurement of serum GLS can be used to evaluate symptomatic improvements after surgery (arthroplasty or synovectomy) as well as the aggressiveness of disease activity in rheumatoid arthritis (RA). Serum GLS levels were determined by enzyme immunoassay in rheumatoid factor (RF)-positive and -negative RA patients. In those undergoing surgery, levels were measured 3 months before and after the operations. Both RF-positive and -negative RA sera showed higher GLS levels than normal and osteoarthritis sera. Patients undergoing arthroplasty demonstrated a decrease in serum GLS levels after the operations, but patients undergoing synovectomy did not, reflecting the extent of remaining or reproliferating synovial tissues rich in GLS production. These findings suggest that the serum GLS level is a useful indicator for evaluation of synovitis and the systemic efficacy of surgical treatment.

Serum components in which embryos are cultured in vitro are very important for normal embryonic development. In this study, rat serum was fractionated using Macrosep filters to study the effect of a single growth factor. The fractionated serum, both that containing only material greater than 30 kDa molecular weight >> 30 kDa) and that from which material between 30 kDa and 50 kDa had been removed (< 30 kDa+>> 50 kDa), caused significant embryonic growth retardation. Addition of different concentrations of basic fibroblast growth factor (bFGF, 18 kDa), vascular endothelial growth factor (VEGF, 45 kDa) and platelet-derived endothelial growth factor (PD-ECGF, 45 kDa), to fractionated serum (bFGF to>> 30 kDa serum and VEGF or PD-ECGF to < 30 kDa+>> 50 kDa serum) partially restored embryonic growth and development according to a morphological scoring system and protein assay. This restoration was clear by all criteria, as well as in yolk sac vascularisation and heart development. The growth promoting effects of all 3 factors were significant but did not reach the level seen in embryos grown in whole rat serum. The effect of these growth factors was also investigated on anembryonic yolk sac development using a concentration for which maximum whole embryonic growth was seen (128 ng/ml bFGF, 1.6 ng/ml VEGF and 4 ng/ml PD-ECGF), and significant anembryonic yolk sac development was found. These findings suggest that the angiogenic factors may have a growth promoting effect on total embryonic development and vascularisation.

The expression of platelet-derived endothelial cell growth factor (PD-ECGF) was determined immunohistochemically in 143 non-small cell lung carcinomas. Staining was observed in 48% of the cases. A relationship between histology, stage, erbB-1, erbB-2, ras and PD-ECGF expression was not found. A relationship of borderline significance was observed between PD-ECGF and p53 expression. There was also no relationship between PD-ECGF expression and proliferative activity (G1 phases, S phases, cyclin A). In contrast, a correlation between PD-ECGF- and VEGF-expression was detectable (p=0.009). Furthermore, PD-ECGF expression was related to the response of lung carcinomas to doxorubicin (p=0.0004). Of 35 sensitive tumors, 26 carcinomas were PD-ECGF-positive (74%) while of 108 resistant carcinomas only 43 tumors (40%) exhibited PD-ECGF expression.

Differences in the secretion of the vascular regulators endothelin-1 (ET-1, prostacyclin (PGI2) and thromboxane A2 (TXA2) associated with ageing were investigated in cultured endothelial cells isolated from normal human umbilical veins (HUVECS). HUVECS at different population doubling levels (PDLs) were cultured in medium MCDB-104 supplemented with FBS and ECGF. Cell saturation density was determined by a Coulter counter, and concentrations of ET-1, PGI2 and TXA2 in the media were determined by radioimmunoassay. The cellular and nuclear size of HUVECS increased with advancing age, and the dividing ability decreased. Cell saturation density of HUVECS decreased 5-fold between PDLs 7 and 67 (P < 0.01). The secretion of ET-1 by HUVECS at a young stage of growth (PDL 7.6) increased linearly between 0 and 36 h of incubation (P < 0.01). ET-1 secretion increased approximately 3-fold between PDLs and 67 (P < 0.01). PGI2 secretion increased 6-fold between PDLs 7 and 67 (P < 0.01), and TXA2 secretion increased 18-fold between PDLs 7 and 67 (P < 0.01). The ratio of PGI2 to TXA2 secretion decreased 3-fold between PDLs 7 and 40 (P < 0.01), and remained at the lower ratio between PDLs 40 and 67. This data indicates that the anti-thrombotic or anti-vasoconstrictive role of endothelial cells may decrease during in vitro ageing.

Recent investigations on regeneration of the periodontium have attempted to define factors involved in the formation of a new connective tissue attachment. One essential biological event involved in tissue regeneration is directed cell migration (chemotaxis). Extracellular matrix proteins have been shown to influence chemotaxis, cell proliferation and differentiation. Recently, the extracellular matrix proteins, fibronectin (FN) and laminin (LM), and the polypeptide, endothelial cell growth factor (ECGF), have been shown to stimulate a variety of biological processes. Current assay systems which attempt to define cell migration are the Boyden chamber assay and a random cell migration assay. Neither assay system adequately defines in vivo cell migration. Here we present a new in vitro assay system that tests the capacity of several biological response modifiers applied on dentin to stimulate a chemotactic and proliferative response from various cell types. The assay system consists of two types of assays. Assay I measures the chemotactic activity of test substances bound to dentin. In this assay cells must actively move through a filter (Nuclepore) towards a factor bound to dentin. Assay II examines the ability of dentin-bound biological response modifiers to stimulate directed movement and proliferation of cells on dentin surfaces. We report that periodontal ligament (PDL) cells migrate towards FN and ECGF; that PDL cell migration is enhanced when dentin is preconditioned with tetracycline HCl; that PDL cells have an increased proliferative response when dentin is conditioned with both FN and ECGF; that gingival epithelial cells have increased migratory and proliferative responses when LM is used to condition dentin; and that there is a reciprocal utilization of biological response modifiers by gingival epithelial cells and PDL cells.

The purpose of this study was to examine how chondrocytes are involved in the molecular mechanism of inflammation in rheumatoid arthritis (RA). A chondrosarcoma cell line (OUMS-27) was cultured and treated with interleukin-1beta (IL-1beta). Changes in the expression levels of matrix metalloproteinase-1 (MMP-1), metalloproteinase-13 (MMP-13), and gliostatin/platelet-derived endothelial cell growth factor (GLS/PD-ECGF) were assessed by reverse transcription polymerase chain reaction (RT-PCR) and enzyme-linked immunosorbent assays. IL-1beta induced the expressions of MMP-1, MMP-13, and GLS mRNAs and proteins in a dose-dependent manner. Selective inhibition of the p38 mitogen-activated protein kinase (p38 MAPK) pathway with SB 203580 and SB 202190 blocked the expression of MMP-1, MMP-13, and GLS more strongly than selective in hibition of the extracellular signal-regulated kinase 1 and 2 (ERK1/2) pathway by PD 98059. These findings suggest that chondrocytes may intensify cartilage destruction and inflammation in RA by the induction of MMP-1, MMP-13, and GLS by IL-1beta and that the p38 MAPK pathway plays an important role in these inductions.

The aim of this study was to investigate the expression of platelet-derived endothelial growth factor (PD-ECGF) in human gallbladder carcinomas to elucidate its role in angiogenesis and tumour progression. To this end, 56 archival surgical specimens of gallbladder lesions were examined for PD-ECGF/thymidine phosphorylase (TP) expression by immunohistochemistry and the PD-ECGF/TP protein level was assessed in five fresh specimens of gallbladder carcinoma by enzyme-linked immunosorbent assay (ELISA). Hyperplastic epithelial cells and adenoma cells showed no or faint staining with PD-ECGF/TP. Out of 43 gallbladder carcinomas, 27 (63%) showed moderate to strong immunoreactivity in the cytoplasm and nuclei of the tumour cells. PD-ECGF/TP immunoreactivity in stromal infiltrating cells was detected in 43% (3/7) hyperplasias, 17% (1/6) adenomas and 86% (37/43) carcinomas. PD-ECGF/TP protein levels in carcinoma tissues were higher than those in corresponding normal mucosa. PD-ECGF/TP expression did not correlate with angiogenesis, but significantly correlated with depth of invasion, lymph node metastasis, and tumour stage. These results overall suggest that PD-ECGF/TP produced by both cancer cells and infiltrating cells is associated with tumour progression in human gallbladder carcinoma.

The mechanisms of cancer cachexia have not been elucidated. We previously reported that cyclic plasma perfusion using non-coated charcoal was effective in cancer cachexia. In the present study we investigated the angiogenic effect of cyclic plasma perfusion on adipose tissue. Twenty rabbits were divided into two groups, i.e., tumor-bearing rabbits subjected to cyclic plasma perfusion (n=10, PP group), and tumor-bearing rabbits subjected to sham-perfusion (n=10, SP group). The changes in body weight, total body fat, and expression of angiogenic factors were investigated. Loss of body weight and total body fat was significantly suppressed in the PP group. Apoptosis of adipocytes was seen in both groups only in the early stage of tumor bearing. Lipolytic activity in the PP group showed a lower ratio than that in the SP group. In the PP group, increased microvessel density and expression of vascular endothelial growth factor (VEGF) and platelet-derived endothelial cell growth factor (PD-ECGF) were seen at 40 days after transplantation. Similar findings were not seen in the SP group. These results suggest that cyclic plasma perfusion not only decreased lipolytic activity but induced angiogenesis in the adipose tissue.

BACKGROUND

Platelet-derived endothelial cell growth factor (PD-ECGF) is an angiogenic factor that is expressed in various cancer tissues. Little is known regarding plasma PD-ECGF levels in patients with chronic liver disease such as chronic hepatitis (CH), cirrhosis, and hepatocellular carcinoma (HCC) with cirrhosis. The expression of PD-ECGF in HCC tissues also remains to be clarified.

METHODS

Plasma PD-ECGF levels in patients with chronic liver disease were determined with an enzyme-linked immunoadsorbent assay system using the mouse monoclonal antibodies specific to PD-ECGF. These were cross-sectionally compared among groups of normal persons, CH, cirrhosis, and HCC patients. The HCC patients were classified into two groups based on TNM stage: early and advanced stage disease groups. PD-ECGF expressions in HCC tissues were immunohistologically examined.

RESULTS

The plasma PD-ECGF levels from the normal individuals and those with CH, cirrhosis, and HCC specimens were 4.2+/-0.5, 4.3+/-0.6, 4.6+/-1.1, and 6.0 +/-2.5 U/mL, respectively. The plasma PD-ECGF concentration was highest in HCC (P < 0.05). No significant difference was found among the normal subjects, CH, and cirrhosis specimens. Plasma PD-ECGF concentrations were significantly higher in the advanced stage disease HCC group compared with the early stage disease group (6.75+/-2.62 U/mL vs. 4.19+/-0.34 U/mL) (P < 0.05). Immunohistochemical expression of PD-ECGF in HCC cells increased significantly compared with normal liver cells (P < 0.05).

CONCLUSIONS

Circulating PD-ECGF plasma level might be a new tumor marker for progression in patients with HCC. Immunohistological findings correspond to elevation of the plasma PD-ECGF in HCC patients. It is possible that increased production of PD-ECGF in HCC cells causes abundant neovascularization.

This study was designed to comprehensively analyze the differential expression of proteins from human umbilical vein endothelial cells (HUVECs) exposed to tumor conditioned medium (TCM) and to identify the key regulator in the cell cycle progression. The HUVECs were exposed to TCM from breast carcinoma cell line MDA-MB-231, then their cell cycle distribution was measured by flow cytometer (FCM). The role of protein in cell cycle progression was detected via two-dimensional polyacrylamide gel electrophoresis (2-DE) and western blotting. Following the stimulation of TCM, HUVECs showed a more cells in the S phase than did the negative control group (ECGF-free medium with 20% FBS), but the HUVECs' level was similar to the positive control group (medium with 25 micrograms/ml ECGF and 20% FBS). Increased expression of cyclin D1/E and some changes in other related proteins occurred after incubation with TCM. From our results, we can conclude that breast carcinoma cell line MDA-MB-231 may secrete soluble pro-angiogenic factors that induce the HUVEC angiogenic switch, during which the expression of cell cycle regulator cyclin D1/E increases and related proteins play an important role in this process.

Crystals of recombinant platelet-derived endothelial cell growth factor (PD-ECGF) were obtained by the hanging drop vapour diffusion technique. The crystals belong to the space group P2(1)2(1)2(1) with unit cell dimensions a = 63.7 A, b = 70.4 A, c = 219.6, alpha = beta = gamma = 90 degrees, and probably contain a single dimer in the asymmetric unit. Diffraction to a minimum Bragg spacing of 3.5 A has been obtained using a synchrotron X-ray source.

Platelet-derived endothelial cell growth factor (PD-ECGF) stimulates chemotaxis of endothelial cells in vitro and has angiogenic activity in vivo. Recently PD-ECGF was shown to have thymidine phosphorylase activity. In order to study possible therapeutic applications of PD-ECGF we used a rat model to determine its pharmacokinetics and tissue distribution after intravenous injection. [125I]PD-ECGF disappeared from the plasma in a biphasic manner, with estimated distribution and elimination half-lives of 17 min and 7 h, respectively. PD-ECGF was metabolized in the liver, excreted via the bile, and not accumulated in any organ system. The stability and long half-life in the circulation, together with the specificity for endothelial cells, suggest that PD-ECGF may be useful as a therapeutic agent to stimulate re-endothelialization in vivo, or, in view of its thymidine phosphorylase activity, in chemotherapy, by decreasing the pool of available thymidine.

Heparin in combination with endothelial cell growth factor (ECGF) affects physiological responses and growth of human umbilical vein endothelial cells (HUVEC). We have examined the effect of heparin, crude ECGF (endothelial cell growth supplement [ECGS]), or both on the basal and thrombin challenged output of metabolites by HUVEC. The supernatant and/or cell lysate was assayed for released prostacyclin, von Willebrand factor, tissue plasminogen activator, plasminogen activator inhibitor and thrombospondin. Heparin modified release of all these metabolites when in combination with ECGS, and in general these responses were the opposite of those generated by inflammatory mediators such as interleukin-1. It has been postulated that heparin acts by potentiating the effect of ECGF, but heparin inhibited thrombospondin release and enhanced that of von Willebrand factor in the absence of ECGS, while ECGS alone inhibited release of plasminogen activator inhibitor. Thus, under our experimental conditions it would appear that heparin and crude ECGF can affect HUVEC independently of one another.

BACKGROUND

Thymidine phosphorylase (dThdPase) is identical to platelet-derived endothelial cell growth factor (PD-ECGF). dThdPase is known to promote the development of new blood vessels, which are fundamental to tumor growth and metastasis. Vascular endothelial growth factor (VEGF) is a 34-42 kilodalton (kD) protein that induces both angiogenesis and vascular permeability. Proliferating cell nuclear antigen (PCNA) is a 36 kD nuclear protein, and its expression is associated with DNA synthesis and cell proliferation.

METHODS

The authors investigated the correlations of dThdPase and VEGF with the growth of head and neck squamous cell carcinoma (HNSCC) in 95 patients by examining PCNA expression as a marker of tumor proliferation. They also retrospectively examined the expression of dThdPase in primary HNSCC and its association with angiogenesis and clinicopathologic findings.

RESULTS

Microvessel count was significantly correlated with the expression of VEGF (P = 0.046) but not with dThdPase expression. The expression of PCNA was significantly correlated with dThdPase (P < 0.001) but not VEGF expression. A significant correlation was found between VEGF and dThdPase expression (P = 0.003). Neither dThdPase nor VEGF correlated with clinicopathologic findings, except for the correlation between tumor location and VEGF expression (P 0.020).

CONCLUSIONS

These findings suggest that VEGF is involved in angiogenesis in HNSCC. dThdPase may have effects on tumor growth other than angiogenic activity in HNSCC.