Thrombospondin and apoptosis: molecular mechanisms and use for design of complementation treatments.

Thrombospondin-1 is the first and most studied naturally occurring protein inhibitor of angiogenesis. Its characteristic multi-domain structure determines thrombospondin-1 divergent functions, which include but are not limited to the regulation of angiogenesis. Below we overview the structural determinants and receptors expressed on the endothelial and other cell types, that are at the root of thrombospondin-1 striking ability to block neovascularization. We specifically emphasize thrombospondin-1 direct apoptotic action on the remodeling vascular endothelium and summarize current knowledge of its pro-apoptotic signaling and transcriptional networks. Further, we provide comprehensive survey of the thrombospondin-based anti-angiogenic strategies with special focus on the combination treatments. We convincingly illustrate how precise knowledge of the pro-apoptotic events and intermediates elicited by thrombospondin in the vascular endothelial cells facilitates the design of the most effective treatment combinations, where the efficacy of thrombospondin-derived compounds is maximized by the partner drug(s) ("complementation" strategies) and provide examples of such fine-tuning of the thrombospondin-based anti-angiogenic treatments.

Backbone modification alters the efficacy of antisense oligonucleotides directed against mRNA encoding either TGF-alpha or EGFR in the treatment of prostate cancer cell lines.

Antisense oligonucleotides (oligos) directed against mRNA-encoding, transforming growth factor-alpha (TGF-alpha) and the epidermal growth factor receptor (EGFR), have been shown to significantly inhibit in vitro and in vivo growth of prostate tumor models. Recently, second generation oligos have been employed with identical base sequences, but containing backbome modifications that enhance stability, solubility and circulatory patterns. Using relatively low concentrations of oligos, we compared the efficacy of the first generation phosphorothioated oligos against TGF-alpha (MR1) and EGFR (MR2) with second generation oligos containing completely phosphorothioated backbones and different patterns of 2'-methoxyethyl (2'-MOE) backbone modifications, while retaining the original designated base sequence using, the LNCaP and PC-3 prostate cancer cell lines, respectively. All experiments were conducted in vitro with lipofectin to enhance oligo entry. Under these conditions, using oligo concentrations between 0.83 and 3.32 microM for LNCaP cells treated with oligos directed against TGF-alpha only the first generation MR1 had inhibitory activity. When treated with oligos directed against EGFR, none of the oligos had inhibitory activity and they behaved similarly. Using the PC-3 cell line and treatment directed against TGF-alpha with oligo concentrations between 0.42 and 3.32 microM, first generation MR1 and second generation 5005 behaved similarly with no notable effect, while second generation 5007 produced dramatic growth stimulation. When PC-3 cells were treated with oligos directed against EGFR, second generation 5006 and 5008 had similar and apparently dose-dependent inhibition. We conclude that backbone modifications influence oligo efficacy and may result in either enhanced or diminished activity. Because of their activity against the hormone insensitive PC-3 cells, the 5006 and 5008 compounds warrant additional study at greater concentrations and also merit in vivo testing.

Relaxin decreases renal interstitial fibrosis and slows progression of renal disease.

BACKGROUND: Relaxin, a hormone of the insulin-growth factor family, promotes collagen remodeling. In rodent models of pulmonary and dermal fibrosis, relaxin reduced interstitial fibrosis. To study relaxin's effect in renal disease, we used the experimental bromoethylamine (BEA) model that leads to severe renal interstitial fibrosis, a decrease in glomerular filtration rate, and albuminuria at one month. METHODS: Rats were injected with BEA one week prior to implantation of an osmotic pump delivering relaxin (2 microg/hour) or vehicle continuously for 28 days. RESULTS: BEA caused a significant decrease in creatinine clearance, which was partially prevented by relaxin. In the relaxin-treated BEA rats, serum creatinine was normal, and albumin excretion was slightly decreased. By morphometric measurement, relaxin administration was associated with a significant decrease in interstitial fibrosis at the corticomedullary junction. This was accompanied by a decrease in the number of ED-1 positive cells (an index of macrophage infiltration) and in the intensity of immunohistochemical staining for transforming growth factor-beta. This antifibrotic effect of relaxin did not appear to be mediated by systemic hemodynamic changes since the mean arterial pressure was not significantly different among the groups. CONCLUSIONS: Relaxin may have a useful application in decreasing interstitial fibrosis and thereby slowing the progression of renal disease.

Differential effects of enalapril and irbesartan in experimental papillary necrosis.

This study was undertaken to determine whether angiotensin receptor blockers are as renoprotective as angiotensin-converting enzyme inhibitors in an experimental model of chronic interstitial renal disease. Groups of rats received one of the following treatments for 1 week: (1) enalapril, (2) diltiazem, (3) a cocktail of hydralazine, reserpine, and hydrochlorothiazide, or (4) irbesartan (an AT1 antagonist). The animals were injected with bromoethylamine (200 mg/kg), and antihypertensive treatment continued for 1 month. All drugs were effective in lowering the mean arterial pressure. The bromoethylamine-treated rats developed albuminuria and sustained a 40-50% decrease in creatinine clearance. Enalapril and irbesartan reduced albuminuria, but only enalapril partially prevented the decline in creatinine clearance and lowered the number of ED-1-positive cells. Diltiazem and cocktail had no effect on proteinuria, creatinine clearance, or ED-1 cells. In this experimental model, the effects of enalapril and irbesartan were not identical. Both drugs reduced proteinuria, but enalapril was more effective in protecting the renal function. The fact that the AT1 antagonist protected against albuminuria but did not affect the clearance of creatinine implies that the results seen with angiotensin-converting enzyme inhibition may be in part due to an effect on angiotensin II via AT2 receptor blockade or through an effect on bradykinin.

Treatment of the T98G glioblastoma cell line with antisense oligonucleotides directed toward mRNA encoding transforming growth factor-alpha and the epidermal growth factor receptor.

Antisense oligonucleotides (oligos) complementary to mRNA encoding transforming growth factor-alpha (TGF-alpha) and its target, the epidermal growth factor receptor (EGFR), are efficacious against human prostate and breast cancers carried in athymic nude mice. Glioblastomas, also regulated by EGFR expression, would appear to be similarly susceptible, and we now employ them against the T98G tumor model. T98G cells were distributed into wells and allowed to adhere prior to addition of oligos (12.5 microM) directed against TGF-alpha and/or EGFR for 6 d of treatment before thymidine radiolabeling. Supplemental media and oligos (25 microM final concentration) were added after d 3. Statistically significant inhibition by oligos directed against TGF-alpha, EGFR, and their combination was 13.8%, 26.3%, and 18.1%, respectively. In a subsequent experiment cells were incubated with increasing amounts of each oligo and their combination for 3 d prior to radiolabeling. Statistically significant inhibition of growth for either oligo at every concentration was found. Cells incubated with 6.25, 12.5, 25, and 50 microM antisense directed against TGF-alpha had a mean inhibition of 29.3%, 33.3%, 21.7%, and 46.6%, respectively. Cells similarly treated with oligos against EGFR had a mean inhibition of 77.9%, 80.3%, 82.0%, and 83.7%, respectively, and cells incubated with 6.25, 12.5, 25 and 50 microM of each oligo had a mean inhibition of 74.7%, 70.6%, 70.8%, and 76.3%, respectively. Lastly, in a paired experiment, cells treated with 0, 0.39, 0.78, 1.56, 3.125, and 6.25 microM of oligos, either specifically directed against EGFR or a random control, for 3 d were evaluated for both thymidine incorporation and EGFR expression. Statistically significant inhibition of 3H-thymidine incorporation was seen in cells with the oligo specifically directed against EGFR at 3.125 microM and 6.25 microM when compared to non-oligo containing controls. This was accompanied by a comparable significantly decreased expression of a low-MW reactive derivative of EGFR at 3.125 microM and 6.25 microM in Western blots, and of a high-MW reactive EGFR at 6.25 microM. The significant effect against high-MW EGFR was observed vs both the non-oligo containing control and the random sequence. Oligo concentrations between 0.78 and 1.5 microM also resulted in decreased expression of the low-MW form, but not significant differences in thymidine radiolabeling. In recovery experiments, cells treated initially with greater oligo concentrations required significantly increased time to recover, particularly in cells treated with EGFR directed oligos. Intracellular uptake and nuclear localization was demonstrated with FITC tagged oligos. In summary, even at relatively low oligo concentrations and short exposure, oligos against TGF-alpha, and particularly EGFR, significantly inhibit in vitro growth of the T98G glioblastoma, possibly mediated by decreased EGFR expression.

Biotinylation of antisense oligonucleotides does not alter lipofectin enhanced cellular uptake in prostate cancer cell lines.

Biotinylation is a common modification made to pharmaceuticals, including antisense oligonucleotides (oligos), to enhance their specific delivery. Such agents bind to targets that have been previously labeled with conjugated avidin, or alternatively, heteroconjugate monoclonal antibodies that have dual biotin and tumor-specific antigen specificities may be employed. However, for a drug to be efficacious it must also be taken up by the targeted cells. This is frequently difficult for large molecular weight compounds and cationic lipids, like lipofectin, are often employed. However, the effect of biotinylation on oligo uptake has not been examined in the presence of lipofectin, particularly in prostate cancer cells. Oligos conjugated with biotin and FITC were incubated in vitro with LNCaP and PC-3 cells in the presence of a previously determined effective concentration of lipofectin. Fluorescent uptake and distribution was compared to similar oligos that were not biotinylated. The results demonstrate that biotinylation does not alter the uptake of oligos in LNCaP or PC-3 prostate cancer cells, nor does it alter their retention or cytoplasmic distribution in PC-3 cells when used with lipofectin.

In vivo establishment of T98G human glioblastoma.

Human derived T98G glioblastoma has long been utilized as an in vitro model for epidermal growth factor receptor (EGFR)-mediated growth regulation. Recently, T98G has been employed to develop new types of therapy directed at limiting EGFR expression such as by administration of antisense oligonucleotides directed against EGFR encoding mRNA. A major limitation to extending this model for in vivo application is that T98G implanted s.c. or intracerebrally has been reported not to grow in nude mice. In an effort to extend this model to permit in vivo studies, we evaluated the use of Matrigel and orthotopic (intracranial) implantation techniques. When equal volumes of Matrigel were mixed with T98G cell suspensions, tumors developed at both flank and orthotopic locations. Four groups of nude mice were inoculated into the flanks with either 10(5), 10(6), 4 x 10(6) or 10(7) T98G cells in a 150 microliters total volume with Matrigel. In 1/5, 3/5, 1/5 and 1/3 mice receiving 10(5), 10(6), 4 x 10(6) and 10(7) cells, respectively, tumors developed 11, 15, 15 and 15 weeks, respectively, following inoculation. Out of 4 mice inoculated orthotopically (intracranially into the frontal lobe) with only 4 x 10(4) cells and Matrigel, 2 developed tumors. However, all mice (4/4) inoculated orthotopically with 4 x 10(5) cells in a 10 microliters total volume with Matrigel developed tumors. Two were identified histologically following a scheduled sacrifice at 36 and 60 days and two more at 103 and 118 days after sacrifice following abnormal behavior. The best tumor establishment efficacy combined orthotopic implantation of 4 x 10(5) T98G cells with Matrigel. These techniques permit the use of T98G glioblastoma as an in vivo model for new forms of therapy.

Evolution of experimentally induced papillary necrosis to focal segmental glomerulosclerosis and nephrotic proteinuria.

Bromoethylamine (BEA)-induced papillary necrosis is a reproducible model for analgesic nephropathy. We induced this lesion in groups of male Sprague-Dawley rats and followed the functional and histological changes for 1 year. We found that by 1 month, necrosis of the papilla was complete, glomerular filtration rate was depressed, and urine albumin excretion was increased. There was an extensive interstitial fibrosis characterized by a mononuclear cell infiltrate and patchy tubular atrophy. By 6 months, there was re-epithelialization of the papillary stump accompanied by a marked increase in albuminuria and an improvement in concentrating ability. Changes seen at 9 months were more advanced. There was extensive cortical fibrosis manifested by pitting of the surface of the kidney. At 1 year, renal function remained impaired (creatinine clearance reduced by 65% to 0.26 mL/min/100 g), and the animals were now markedly nephrotic, with albuminuria of 254 mg of albumin/24 h. In the BEA rats, there was selective destruction of the deep nephrons leading to an increase in the volume-ratio of superficial to deep nephrons. Glomerular changes, affecting approximately 60% of the glomeruli, were characteristic of focal segmental glomerular sclerosis. This model of papillary necrosis/interstitial fibrosis is associated with chronic renal insufficiency and leads to the development of focal glomerular sclerosis and nephrotic proteinuria by 6 to 12 months after its induction.

Angiotensin-converting enzyme inhibition reduces the effect of bromoethylamine-induced papillary necrosis and renal fibrosis.

Rats injected with a single, 50-mg dose of bromoethylamine (BEA) developed papillary necrosis accompanied by sever interstitial fibrosis. At 1 mo, the creatinine clearance decreased (control 0.66 versus BEA 0.33 ml/min per 100 g body wt, P = 0.02), and the urine albumin-to-creatinine ratio increased markedly (control 0.19 versus BEA 0.51, P = 0.02). In a group of animals given the angiotensin-converting enzyme inhibitor enalapril (Enal; 100 mg/L) in their drinking water for 4 wk, beginning 1 wk before BEA injection, creatinine clearance improved significantly (BEA 0.33 versus Enal + BEA 0.52 ml/min per 100 g body wt, P = 0.01) and albumin excretion fell to zero. Histologic examination revealed an 88% decrease in the area of papillary necrosis and a decrease in the degree of interstitial fibrosis in the corticomedullary junction. To determine whether this was due to changes in urine flow rate induced by enalapril, a group of animals was injected with BEA, and enalapril at the above dose was begun 1 wk later. After 1 mo, the enalapril-treated animals showed the same improvement in creatinine clearance (BEA 0.33 versus BEA + Enal 0.50 ml/min per 100 g body wt, P = 0.03) and suppression of albumin excretion. The area of papillary necrosis was reduced by 67%. In the BEA animals treated with enalapril, ED-1-positive cells, alpha-smooth muscle actin, and transforming growth factor-beta1 were decreased compared with BEA alone. It is concluded that in this model of papillary necrosis, enalapril protects renal function and decreases interstitial fibrosis mediated at least in part through an angiotensin II/bradykinin-dependent mechanism.

Paclitaxel, bropirimine and linomide: effect on growth inhibition in a murine prostate cancer model by different growth regulatory mechanisms.

Paclitaxel, bropirimine and linomide therapy was evaluated in a murine prostate cancer model. All drugs were effective in impeding tumor growth but the mechanisms of action varied. Paclitaxel inhibited bcl-2 expression suggesting an apoptotic mechanism. Bropirimine, while inhibiting bcl-2 expression also significantly depressed tumor necrosis factor-alpha (TNF-alpha) expression. In the bropirimine treated group there was also a correlation between angiogenesis and cyclin D expression. Finally, linomide significantly decreased angiogenesis. Since the mechanism of action of these drugs differ, combining them at lower doses might maintain therapeutic efficacy while reducing toxicity.

An evaluation of the markers p53 and Ki-67 for their predictive value in prostate cancer.

BACKGROUND AND OBJECTIVES: p53 and Ki-67 are but two markers being evaluated for their predictive value in prostate cancer. The purpose of this study was to compare p53 and Ki-67 with age, stage, Gleason score, and ploidy for their prognostic abilities in prostate cancer. METHODS: Prostate cancer specimens from 134 patients were immunohistochemically stained for p53 and Ki-67 expression and differences evaluated by SPSS analysis of variance (ANOVA) methods. The dependent variable was patient survival and the independent variables were age, stage, Gleason score, and ploidy. RESULTS: In decreasing order of prediction of survival were stage (P < 0.001), Gleason score (P < 0.001), age (P = 0.1869), Ki-67 (P = 0.2284), p53 (P = 0.4282) and ploidy (P = 0.8141). CONCLUSION: It is concluded that stage and Gleason score are significant predictors of survival while p53, Ki-67, age and ploidy are not.

Growth factor deprivation therapy of hormone insensitive prostate and breast cancers utilizing antisense oligonucleotides.

Antisense oligonucleotides (oligos) are artificial sequences of nucleotide bases which may be synthesized complementary to known regions within specific mRNAs. When these constructed oligos interact with protein encoding mRNA they may regulate expression of various growth factors and/or their receptors. Oligos directed against transforming growth factor-alpha (TGF-alpha) and its binding site, the epidermal growth factor receptor (EGFR), were employed: A) in vitro to affect the growth of hormone insensitive human derived PC-3 prostate cancer cells as well as the human derived UACC-893 breast cancer cell line; and B) in vivo to treat tumors established by these cell lines in athymic nude mice. The in vitro results for each oligo, and their combination, produced significant inhibition of both prostate and breast cell lines. In addition, the combination of oligos most efficiently diminished the immunohistochemical expression of both TGF-alpha and EGFR in PC-3 cells. Direct in vivo inoculation of oligos into established PC-3 or UACC-893 tumors in nude mice produced hemorrhagic necrosis within 2-3 days. Such therapy could represent a new tier of therapy for recurrent, hormone insensitive, tumors based upon the concept of growth factor deprivation.

Paclitaxel is more effective than thalidomide in inhibiting LNCaP tumor growth in a prostate cancer model.

LNCaP tumors were treated by either administration of paclitaxel, thalidomide or by orchiectomy in order to determine their relationship with markers pertaining to the process of tumor growth, apoptosis or angiogenesis. Forty rats bearing LNCaP tumors were divided into 4 groups of 10 and treated by either paclitaxel (20 mg/kg x 5 days); thalidomide (200 mg/kg x 5 days/week x 5 weeks); or orchiectomy. After 6 weeks serum samples were removed for PSA determination and the animals sacrificed for evaluation of: A) tumor volume; B) tissue bcl-2, cyclin D, PSA and factor VIII immunohistochemically graded (0-5 scale) for marker expression; and C) serum PSA. Comparisons were made to untreated LNCaP tumors. Statistically significant differences were determined using the nonparametric Mann-Whitney test. Paclitaxel produced significant differences in volume (p < 0.001), expression of bcl-2 (p < 0.043), cyclin D (p < 0.023), tissue PSA (p < 0.001) and serum PSA (p < 0.019) levels. Thalidomide altered expression of bcl-2 (p < 0.011) and tissue PSA (p < 0.002). Orchiectomy altered volume (p < 0.002) and bcl-2 expression (p < 0.001). All three therapies have been suggested for prostate cancer and each produced alterations in accepted markers for treatment response (either reduced volume or serum PSA). Paclitaxel significantly influenced the most markers. Of interest was that all treatments, especially thalidomide, a known antiangiogenesis agent, reduced factor VIII, although not significantly. Evidently each treatment evokes different pathways of activity.

Enhanced expression of bcl-2 following antisense oligonucleotide mediated growth factor deprivation.

Although the role of bcl-2 in apoptosis has been described, its involvement in prostate cancer (CAP) progression is less well understood, but thought to be involved with the transition of CAP from androgen-sensitivity to androgen-independence, where its expression is augmented following androgen ablation. For treating these recurrent androgen-independent tumors, following hormone treatment failure, a new tier of therapy based upon growth factor deprivation has been suggested, implemented by antisense oligonucleotides (oligos) directed against mRNA encoding a critical growth regulatory autocrine loop (comprised of transforming growth factor-alpha (TGF-alpha) and its binding site, the epidermal growth factor receptor (EGFR). To determine whether oligo-induced growth factor deprivation therapy similarly enhanced expression of bcl-2 (as follows androgen deprivation) human prostate cancer derived PC-3 cells were treated in vitro with oligos directed against TGF-alpha (MR-1) and/or EGFR (MR-2). After 5 days of treatment cells were immunochemically stained for human bcl-2. In similar experiments, cells were treated for 3 days prior to extraction of proteins, Western blot analysis, photography and computer evaluation of protein density by SigmaScan software. Immunostained cells treated with oligos directed against mRNA encoding TGF-alpha (MR-1) either alone or in combination with that directed against EGFR (MR-2) had increased bcl-2 expression (+3 to +5). In addition, the intensity of Western blots scanned for bcl-2 expression were 19%, 32% and 30% greater in cells treated with oligos directed against TGF-alpha, EGFR or their combination, respectively. We conclude that enhanced bcl-2 expression followed antisense oligo induced growth factor deprivation. This result is similar to that found upon androgen deprivation therapy, and also demonstrates additional biologic activity of this new class of molecular therapeutic agents.

Lack of toxicity associated with the systemic administration of antisense oligonucleotides for treatment of rats bearing LNCaP prostate tumors.

Antisense oligonucleotides (oligos) are now in clinical trials for the treatment of a variety of diseases. However, concern is sometimes expressed as to the toxicity of such compounds, particularly those with phosphorothioated backbones. We have previously reported (J. Surg. Oncol. 62, 194, 1996) our experience in treating nude mice bearing human PC-3 prostate tumors with phosphorothioated antisense oligos directed against mRNA encoding transforming growth factor-alpha (TGF-alpha) and the epidermal growth factor receptor (EGFR). This therapy resulted in a 75% (9/12) response rate for the intralesional treatment and a 100% (3/3) response rate for the systemic administration utilizing Alzet diffusion pumps. In the current study, athymic nude rats bearing orthotopically implanted LNCaP tumors, whose establishment was confirmed by the expression of human PSA, were implanted subcutaneously with Alzet diffusion pumps and treated systemically for 14 days with a total of 1 mg of each oligo (2 mg total). Controls consisted of five untreated rats similarly inoculated with LNCaP cells, but which did not receive antisense oligos. After 2 weeks the rats were sacrificed and serum samples were evaluated for BUN, creatinine, LDH and SGOT. Lungs, kidneys, livers, spleens and prostates were also removed for pathologic evaluation. There were no serum marker differences between groups nor was there histologic evidence of oligo toxicity seen in any evaluated tissue. Of interest was the observation that the livers and spleens, as well as prostates, of treated animals revealed mild lymphocytic infiltration compared to controls. We conclude that at this level of administration, there is no toxicity associated with 14-day oligo treatment.

Antitumor effect of CPT-11, a new derivative of camptothecin, against human prostate cancer (PC-3) in vitro and prostate rat tumor (AT-3) in vivo.

In an effort to evaluate the efficacy of CPT-11 on prostate cancer we utilized the PC-3 human prostate cancer cell line (in vitro), and the Dunning R3327 AT-3 rat prostate cancer tumor line (in vivo). PC-3 cells were initially seeded and cultured prior to the addition of CPT-11 at different concentrations 48 and 120 h later. After an additional 48 h the number of cells was determined using MTT dye uptake. CPT-11 at concentrations between 10 ng/ml and 50 micrograms/ml inhibited the growth of the PC-3 prostate human cancer cell line up < 0.001). AT-3 prostate rat cancer cells were injected into the right flank of 30 Copenhagen X Fischer rats, divided into treated and control groups. A total dose of CPT-11 (200 mg/kg) was given intraperitoneally, administered 1, 3 and 5 days postimplantation. We evaluated tumor growth by size and weight (5 and 10 days postimplantation). A total dose of 200 mg/kg inhibited the rapid growth of the prostate AT-3 tumor in vivo (p < 0.001).

Antisense oligonucleotide intralesional therapy for human PC-3 prostate tumors carried in athymic nude mice.

Previously we reported hemorrhagic necrosis in human-derived PC-3 prostate tumors, in athymic nude mice, produced by the intralesional injection of antisense oligonucleotides (oligos) directed against mRNAs encoding transforming growth factor-alpha (TGF-alpha) and its target, the epidermal growth factor receptor (EGFR). We now describe our experience with these oligos in treating additional mice with various doses and modes of administration. During prolonged treatment, a dose-response effect was observed, with the optimal dosage consisting of the combination of 400 micrograms of each oligo. Although responses varied, based upon amount and how oligos were administered, we found that tumors were best treated when initially less than 156 mm3. Intralesional inoculations produced necrosis and yielded responses, ranging from complete response (CR) or cure to partial responses (PR) in 9 of 12 tumors treated with full dose (400 micrograms of each oligo) and 1 of 1 treated with 800 micrograms of each oligo, against a large tumor. Included among the 9 positive responses with full-dose administration were 2 tumors that regressed (one completely). A single tumor treated with twice (2X) the normal dosage (800 micrograms of each oligo) also regressed. A single tumor treated with half (1/2) dose (200 micrograms of each) progressed similar to controls, as did 3 of 12 treated with the full dose. Limited experience with ALZET diffusion pumps gave CR (1 of 3) or PR (2 of 3) in 100% of tumors treated (including one mouse cured of multiple tumors in a five day period). It appears that multiple inoculations consisting of 400 micrograms of each oligo is most effective against these tumors, particularly when administered against tumors of <156 mm3 in initial size.