Difference In Vascular Endothelial Growth Factor Levels Research Articles

VEGF is increased in serum but not in spinal cord from patients with amyotrophic lateral sclerosis.

Homogenates of postmortem spinal cord from seven patients with amyotrophic lateral sclerosis (ALS) and six controls together with serum from 13 patients with ALS and 13 controls were analysed for vascular endothelial growth factor (VEGF) using an immunoassay (ELISA). There was no significant difference in VEGF levels in the spinal cord between the ALS patients and the controls. In serum the VEGF levels were significantly higher in the ALS group than in the control group. There was a moderate inverse relation between the duration of the disorder and the serum VEGF levels. The findings indicate that the capacity to synthesize VEGF is preserved even in the late stages of ALS. The results might also be consistent with a transient hypoxic component during the course of ALS, but not with a persistant spinal hypoxia in the late stages of the disorder.

Insulin and insulin-like growth factors (IGFs) promote vascular endothelial growth factor (VEGF) secretion by macaque granulosa cells from preovulatory follicles.

Objective: To investigate the role of insulin−related factors in the regulation of VEGF production by nonluteinized granulosa cells from preovulatory follicles in monkeys. Design: Prospective, comparative, experimental study. Materials/Methods: Beginning at menses, female rhesus monkeys received recombinant human gonadotropins [30 IU FSH twice daily for 6 days, followed by 30 IU FSH and 30 IU LH twice daily for 2-3 days; Ares Advanced Technologies, Inc]. Granulosa cells were obtained by follicle aspiration from anesthetized animals the morning after the last FSH + LH treatment (nonluteinized granulosa cells; NLGC). Enriched preparations of NLGCs from individual monkeys were plated on fibronectin at 5 × 104 cells/well in 48-well plates in DMEM-Ham’s F-12 medium containing 5 μg/ml transferrin, 5 ng/ml H2SeO3, 10 μg/ml aprotinin and 25 μg/ml human low-density lipoprotein. Cells were incubated at 37°C in 5% CO2 -95% air environment for 3 days in the presence and absence of 100 ng/ml hCG (CR 123) and 1, 5, 10, 20, 50, 100 ng/ml of r-h IGF-1 or IGF-2 (R&D Systems, Inc.) or r-h Insulin (ICN Biomedicals Inc.). Media were collected daily and assayed for VEGF (Quantikine VEGF ELISA, R&D Systems; validated for macaque, which is identical to human, VEGF). Cells were cultured in triplicate for each treatment group, and each experiment was replicated 3–5 times. Differences in VEGF levels were determined by one-way ANOVA. Results: NLGCs cultured in chemically-defined conditions for 24 hours secreted low, but detectable, levels of VEGF (3–61 pg/ml) into the media. There was a dose-dependent increase in VEGF levels with addition of 1 to 100 ng/ml IGF-1 (48 to 238 pg/ml; p < 0.001), IGF-2 (69 to 140 pg/ml; p < 0.001), or insulin (30 to 135 pg/ml; p < 0.001). Maximal stimulation was observed with ≥ 20 ng/ml IGF-1 or IGF-2 or 100 ng/ml insulin. Exposure to surge levels of gonadotropin (100 ng/ml hCG) also increased (p < 0.001) VEGF levels compared to controls. However, hCG plus 50 ng/ml IGF-1, IGF-2 or insulin further increased VEGF levels compared to hCG alone (449 vs 285; 503 vs 435; 651 vs 278 pg/ml, respectively; p < 0.001 for IGF-1 and insulin). VEGF levels became nondetectable by 2-3 days in the absence of insulin/IGF or hCG. However, the presence both hCG and IGF-1/-2 or insulin, but neither factor alone, sustained VEGF levels at 530 ± 152, 779 ± 369, 509 ± 210 pg/ml respectively. Conclusions: We reported previously that, unlike in other tissues, gonadotropins (LH/CG) and not hypoxia are major regulators of VEGF secretion by macaque NLGCs. The current data are consistent with the concept that IGFs/insulin act in concert with the midcycle gonadotropin surge to promote VEGF production by granulosa cells in the preovulatory follicle. Supported By: SCCPRR HD18185, RR00163 and Fogarty TWO 00668.

REPLY TO GUNSILIUS ET AL

The letter from Dr Gunsilius and colleagues offers us the opportunity to comment further on the prognostic value of angiogenic growth factors in non-Hodgkin's lymphoma (NHL). Our study ( Bertolini et al, 1999 ) indicated that high pretreatment circulating levels of vascular endothelial growth factor (VEGF) and basic fibroblast growth factor (b-FGF) are predictive of a poor outcome in NHL. It must be noted that Salven et al (1997 ) reported on 82 NHL patients (including high- and low-grade NHL) in whom serum VEGF levels above the median value at the time of diagnosis were highly predictive of a poor survival after the 5-year follow-up. Moreover, the same group has recently confirmed our observation that high pretreatment circulating b-FGF level predicts poor prognosis in NHL ( Salven et al, 1999a ). Two recent papers gave further insight into this issue. First, Vacca et al (1999 ) evaluated angiogenesis in 71 B-cell NHL patients and suggested that the frequency of NHL tissue microvessels, most probably influenced by angiogenic growth factors, increases simultaneously with pathological progression. Second, Bellamy et al (1999 ) have demonstrated that a number of NHL cell lines produce very relevant amounts of VEGF. Taken together, these data suggest that in NHL patients high levels of circulating VEGF (and possibly b-FGF) may be generated by tumours with an angiogenic and aggressive phenotype. We agree with Dr Gunsilius and colleagues that studies on VEGF kinetics after response to treatment are now warranted. The prospective data we have collected so far (unpublished observations) show that VEGF levels decrease in NHL patients responding to therapy. Dr Gunsilius and colleagues also suggest that our VEGF measurements in EDTA–plasma samples may in part reflect VEGF release from activated platelets. Again, this issue has been discussed in recent papers. Wynendaele et al (1999 ) have reported that, although EDTA is known to influence platelet shape, no statistically significant difference in VEGF levels was observed in plasma samples collected with EDTA compared with plasma samples collected with four different anticoagulants (sodium citrate, theophylline, adenosine and dipyridamole) to obtain maximal platelet stabilization. As indicated in the same paper ( Wynendaele et al, 1999 ), there is increasing evidence that in cancer patients serum VEGF levels encompass both platelet-delivered VEGF and VEGF from other sources, most probably angiogenic tumour tissues. In this context, Salven et al (1999b ) have demonstrated that leucocytes and platelets from cancer patients contain higher levels of VEGF than those from healthy controls, possibly acting as scavengers of circulating VEGF produced by the tumour tissue. Accordingly, in cancer patients, the VEGF reservoir circulating in leucocytes and platelets may play a role in tumour angiogenesis and metastasis formation, and it may be clinically useful to evaluate it. Along this line of investigation, we are following the suggestion of Salven et al (1999b ) and prospectively evaluating VEGF levels in both whole blood and plasma from NHL patients.

A comparison of the expression of the malignant phenotype in two androgen-independent human prostate cancer cell lines after orthotopic implantation in nude mice

Expression of some features of the malignant phenotype were compared in the DU145 and PC-3M human prostate cancer cell lines after their intraprostatic growth in nude mice. At necropsy, 27/74 (36%) mice injected with DU145 cells and 41/75 (55%) injected with PC-3M cells had either invasive macroscopic tumors, or microscopic intraprostatic tumor cell nests (p = 0.02). Para-aortic lymph node metastases had occurred in 19% of the DU145 cell, and 44% of the PC-3M cell tumor-bearing animals (p = 0.033). Immunohistochemical staining showed high mutant p53 and vascular endothelial growth factor (VEGF) expression by DU145 cells; PC-3M cells did not express detectable p53, and had relatively low VEGF immunohistochemical reactivity. Assays by ELISA established a statistically significant difference in VEGF levels between the cell lines (p < 0.001). Urokinase-like plasminogen activator (uPA) levels, determined by ELISA, were ten-fold higher in the PC-3M cell tumors, as were matrix metalloproteinase-9 (MMP-9) activities assessed by zymography. These findings of high expression of uPA and MMP-9, two key proteolytic enzymes in the invasive/metastatic process, by PC-3M cell prostatic tumors are consistent with their aggressive behavior; the low VEGF levels compared with those in the poorly metastatic DU145 cell tumors suggest that other angiogenic factors may be critical for prostate cancer cell progression in this model.