An anti-αVß3 antibody inhibits coronary artery atherosclerosis in diabetic pigs.

BACKGROUND AND AIMS: Diabetes is a major risk factor for the development of atherosclerosis. Hyperglycemia stimulates vascular smooth muscle cells (VSMC) to secrete ligands that bind to the αVß3 integrin, a receptor that regulates VSMC proliferation and migration. This study determined whether an antibody that had previously been shown to block αVß3 activation and to inhibit VSMC proliferation and migration in vitro, inhibited the development of atherosclerosis in diabetic pigs. METHODS: Twenty diabetic pigs were maintained on a high fat diet for 22 weeks. Ten received injections of anti-ß3 F(ab)2 and ten received control F(ab)2 for 18 weeks. RESULTS: The active antibody group showed reduction of atherosclerosis of 91 ± 9% in the left main, 71 ± 11%, in left anterior descending, 80 ± 10.2% in circumflex, and 76 ± 25% in right coronary artery, (p < 0.01 compared to lesions areas from corresponding control treated arteries). There were significant reductions in both cell number and extracellular matrix. Histologic analysis showed neointimal hyperplasia with macrophage infiltration, calcifications and cholesterol clefts. Antibody treatment significantly reduced number of macrophages contained within lesions, suggesting that this change contributed to the decrease in lesion cellularity. Analysis of the biochemical changes within the femoral arteries that received the active antibody showed a 46 ± 12% (p < 0.05) reduction in the tyrosine phosphorylation of the ß3 subunit of αVß3 and a 40 ± 14% (p < 0.05) reduction in MAP kinase activation. CONCLUSIONS: Blocking ligand binding to the αVß3 integrin inhibits its activation and attenuates increased VSMC proliferation that is induced by chronic hyperglycemia. These changes result in significant decreases in atherosclerotic lesion size in the coronary arteries. The results suggest that this approach may have efficacy in treating the proliferative phase of atherosclerosis in patients with diabetes.

Disruption of the association of integrin-associated protein (IAP) with tyrosine phosphatase non-receptor type substrate-1 (SHPS)-1 inhibits pathophysiological changes in retinal endothelial function in a rat model of diabetes.

AIMS/HYPOTHESIS: We have previously shown that the association of integrin-associated protein (IAP) with tyrosine phosphatase non-receptor type substrate-1 (SHPS-1) regulates the response of cells, including osteoclasts, osteoblasts, smooth muscle and retinal endothelial cells, to IGF-I. Here we sought to: (1) determine whether the regulation of IGF-I responsiveness by the association of IAP with SHPS-1 is a generalised response of endothelial cells; (2) identify the mechanism by which this association contributes to changes in endothelial cell responses to IGF-I; and (3) determine whether inhibition of this association alters pathophysiological changes occurring in vivo. METHODS: Endothelial cells were maintained in 5 mmol/l glucose and at hyperglycaemic levels, and exposed to an anti-IAP antibody that disrupts the association between IAP and SHPS-1. A rodent model of diabetes with endothelial cell dysfunction was used to investigate the role of the association of IAP with SHPS-1 in endothelial cell function in vivo. RESULTS: Endothelial cells maintained in 5 mmol/l glucose showed constitutive cleavage of the extracellular domain of IAP (which contains the SHPS-1 binding site), with no association between IAP and SHPS-1 being detected. In contrast, hyperglycaemia inhibited IAP cleavage, allowing IAP to associate with SHPS-1 and IGF-I to stimulate SHPS-1 tyrosine phosphorylation. Exposure to the anti-IAP antibody inhibited IGF-I-stimulated tube formation and increased permeability. In the rodent model, basal IAP-SHPS-1 association was not detected in retinal extracts from normal rats, but was fully restored in rats with diabetes. The anti-IAP antibody inhibited the association of IAP with SHPS-1, and reduced retinal vascular permeability and leucocyte adherence to levels similar to those in non-diabetic rats. The antibody also significantly inhibited the aberrant neovascularisation induced by hypoxia. CONCLUSIONS/INTERPRETATION: Our results demonstrate that the increased association of IAP with SHPS-1 contributes to the pathophysiological changes in the endothelium that are induced by hyperglycaemia and hypoxia.

Interaction between insulin-like growth factor-I receptor and alphaVbeta3 integrin linked signaling pathways: cellular responses to changes in multiple signaling inputs.

Integrins are heterodimeric transmembrane proteins that mediate cell attachment to extracellular matrix, migration, division, and inhibition of apoptosis. Because growth factors are also important for these processes, there has been interest in cooperative signaling between growth factor receptors and integrins. IGF-I is an important growth factor for vascular cells. One integrin, alphaVbeta3, that is expressed in smooth muscle cells modulates IGF-I actions. Ligand occupancy of alphaVbeta3 is required for IGF-I to stimulate cell migration and division. Src homology 2 containing tyrosine phosphatase (SHP-2) is a tyrosine phosphatase whose recruitment to signaling molecules is stimulated by growth factors including IGF-I. If alphaVbeta3 ligand occupancy is inhibited, there is no recruitment of SHP-2 to alphaVbeta3 and its transfer to downstream signaling molecules is blocked. Ligand occupancy of alphaVbeta3 stimulates tyrosine phosphorylation of the beta3-subunit, resulting in recruitment of SHP-2. This transfer is mediated by an insulin receptor substrate-1-related protein termed DOK-1. Subsequently, SHP-2 is transferred to another transmembrane protein, SHPS-1. This transfer requires IGF-I receptor-mediated tyrosine phosphorylation of SHPS-1, which contains two YXXL motifs that mediate SHP-2 binding. The transfer of SHP-2 to SHPS-1 is also required for recruitment of Shc to SHPS-1. Ligand occupancy of alphaVbeta3 results in sustained Shc phosphorylation and enhanced Shc recruitment. Shc activation results in induction of MAPK. Inhibition of the Shc/SHPS-1 complex formation results in failure to achieve sustained MAPK activation and an attenuated mitogenic response. Thus, within the vessel wall, a mechanism exists whereby ligand occupancy of the alphaVbeta3 integrin is required for assembly of a multicomponent membrane signaling complex that is necessary for cells to respond optimally to IGF-I.

Structural analysis of the role of the beta 3 subunit of the alpha V beta 3 integrin in IGF-I signaling.

The disintegrin echistatin inhibits ligand occupancy of the alpha V beta 3 integrin and reduces Insulin-like growth factor I (IGF-I) stimulated migration, DNA synthesis, and receptor autophosphorylation in smooth muscle cells. This suggests that ligand occupancy of the alpha V beta 3 receptor is required for full activation of the IGF-I receptor. Transfection of the full-length beta 3 subunit into CHO cells that have no endogenous beta 3 and do not migrate in response to IGF-I was sufficient for IGF-I to stimulate migration of these anchorage dependent cells. In contrast, transfection of either of two truncated mutant forms of beta 3 (terminating at W(715) or E(731)) or a mutant with substitutions for Tyr(747) Tyr(759) (YY) into either CHO or into porcine smooth muscle cells did not restore the capacity of these cells to migrate across a surface in response to IGF-I. This effect was not due to loss of IGF-I receptor autophosphorylation since the response of the receptor to IGF-I was similar in cells expressing either the full-length or any of the mutant forms of the beta 3 subunit. Echistatin reduced IGF-I receptor phosphorylation in cells expressing the full-length or the YY mutant forms of beta 3 subunit, but it had no effect in cells expressing either of two truncated forms of beta 3. A cell-permeable peptide homologous to the C-terminal region of the beta 3 subunit (amino acids 747-762) reduced IGF-I stimulated migration and receptor autophosphorylation of non-transfected porcine smooth muscle cells. These results demonstrate that the full-length beta 3 with intact tyrosines at positions 747 and 759 is required for CHO cells to migrate in response to IGF-I. Furthermore, a region of critical amino acids between residues 742-762 is required for echistatin to induce its regulatory effect on receptor phosphorylation. Since the IGF-I receptor does not bind to alpha V beta 3 the results suggest that specific but distinct regions of the beta 3 subunit interact with intermediary proteins to facilitate IGF-I stimulated cell migration and echistatin induced inhibition of IGF-I signal transduction.

Endogenous IGFBP-3 is protected from inducible IGFBP-3 protease activity in normal adult serum.

We have recently demonstrated that the presence or absence of IGFBP-3 protease activity in physiological fluids may not be determined simply by the presence or absence of specific enzymes but rather the presence of inhibitors. In addition, it appears that these inhibitors may not only be associated with the protease(s) but with the IGFBP-3 itself, protecting it from proteolytic cleavage. To provide further evidence for this mechanism of regulation we investigated whether IGFBP-3 protease activity could be unveiled in normal adult serum (NS) and whether the endogenous IGFBP-3 was protected from this activity. The addition of a range of concentrations of heparin, induced IGFBP-3 protease activity in NS. This was comparable to that seen in pregnancy serum (PS) by virtue of the fragmentation pattern and inhibitor profile. While the addition of zinc also revealed IGFBP-3 protease activity in NS the pattern of fragments differed to that seen in PS. Under both conditions, however, the endogenous IGFBP-3 was not proteolytically modified. These results demonstrate that IGFBP-3 protease activity is present in NS and can be activated, although the endogenous IGFBP-3 is relatively protected from such activity.

Alterations in insulin-like growth factor binding protein-3 proteolysis and complex formation in the arthritic joint.

Increased concentrations of insulin-like growth factor (IGF) system components have previously been observed in rheumatoid arthritis (RA) and osteoarthritis (OA); however, disruption of the IGF axis and the implications for the disease process remain largely unaddressed. This study was undertaken to characterise the IGF binding protein (IGFBP)-3 proteolysis and complex formation systems in synovial fluid and to investigate changes in these systems in arthritic disease, and their impact on the availability of IGF. Western blotting or autoradiography of SDS gels was used to visualise IGFBP-3 or its proteolysis. IGF-I and IGFBP-3 concentrations were determined by radioimmunoassays and acid-labile subunit (ALS) was measured by ELISA. A shift in distribution of IGFBP-3 and IGF-I in RA and OA synovial fluids (RASynF, OASynF) and an associated increase in ALS suggested the presence of 150 kDa ternary complexes. IGFBP-3 proteolysis was decreased in RASynF and OASynF, but was apparent in size-fractionated fluid and resembled serum activity. The presence of serum-like inhibitors of IGFBP-3 proteolysis in RASynF was also demonstrated by the ability of this fluid, and 150 kDa fractions from its size fractionation, to inhibit IGFBP-3 proteolysis in the other synovial fluid. A marked disruption in the IGF system was observed, as considerably more IGF-I was retained in ternary complexes. We also classified the IGFBP-3 proteolysis system in synovial fluid and found it to be disturbed in RASynF and OASynF. These changes may be caused by an increased flux of circulatory proteins into synovial fluid, resulting from an inflammation-induced increase in vascular permeability. The net result in RA and OA would be a decrease in IGF availability in arthritic joints, and therefore loss of a potential anabolic stimulus. This disruption to the IGF axis would influence disease progression in RA and OA.

The role of cell surface attachment and proteolysis in the insulin-like growth factor (IGF)-independent effects of IGF-binding protein-3 on apoptosis in breast epithelial cells.

We have recently reported that insulin-like growth factor (IGF)-binding protein-3 (IGFBP-3) can significantly increase ceramide-induced apoptosis in an Hs578T breast carcinoma cell line in an IGF-independent manner. It was observed in that study that IGFBP-3 added to the cultures was proteolytically modified, generating a specific pattern of fragmentation. We have also previously reported that almost all of the IGFBP-3 outside the circulation in extravascular fluids is in a fragmented form, apparently due to the activity of a cation-dependent serine protease. The aim of this study was to investigate the role of proteolysis in the IGFBP-3 enhancement of C2-induced apoptosis. In this study we confirmed that preincubation of Hs578T cells with IGFBP-3 enhances the apoptotic effect of the ceramide analog C2. The presence of IGF-I completely inhibited the enhancement effect, apparently by inhibiting cell surface association and proteolytic modification. The presence of a serine protease inhibitor [4-(2-aminoethyl)benesulfonyl fluoride] completely inhibited the enhancement effect of IGFBP-3, and Western immunoblotting of conditioned medium and cell surface-associated IGFBP-3 revealed that proteolytic fragmentation of the IGFBP-3 was reduced. In addition, fragments from the incubation of IGFBP-3 with plasmin were able to enhance the susceptibility of Hs578T cells to C2. The effect of these fragments could, however, also be reduced by 4-(2-aminoethyl)benesulfonyl fluoride despite the fact that IGFBP-3 was already fragmented. This suggests additional roles for serine proteases in the IGFBP-3 effect on C2-induced apoptosis in addition to the cleavage of the binding protein.

Active and inhibitory components of the insulin-like growth factor binding protein-3 protease system in adult serum, interstitial, and synovial fluid.

Circulating insulin-like growth factor binding protein-3 (IGFBP-3) proteolytic activity is normally low but increases in serum from pregnant women and from patients with various pathologies. In contrast, we have recently reported that outside the circulation, such activity is normally high but decreases in various pathologies. We have now compared components of the IGFBP-3 proteolytic system revealed after size fractionation of serum and extravascular fluids with different intrinsic levels of such activity. Normal serum, serum from pregnant women, and synovial fluid from patients with rheumatoid arthritis revealed high and low molecular weight (MW) areas of activity. However, only the low MW activity was apparent in interstitial fluid from normal skin (N Inst F) or psoriatic lesions (P Inst F) and in synovial fluid from normal volunteers (N Syn F) or patients with osteoarthritis (OA Syn F). Addition of inhibitors revealed both areas to comprise more than one enzyme, including serine proteases and metalloproteinases; both could also be inhibited by P Inst F, NS, RA Syn F, and inhibitory fractions from the separation of the latter two. These findings demonstrate low and high MW regions of proteolytic activity, which may contribute to the IGFBP-3 protease system, the former always present, whereas the latter seems to be retained within the circulation apart from inflammatory conditions. The variations apparent in IGFBP-3 protease activity in the intact samples related to the presence of an inhibitor, which may protect IGFBP-3 from proteolysis, rather than to changes in the component proteases.

Ovarian dysfunction in endometriosis-associated and unexplained infertility.

PURPOSE: The impact of endometriosis and unexplained infertility on follicular function and fertilization of oocytes in cycles totally unperturbed by exogenous gonadotrophins, when compared with controls with tubal damage, were examined. METHODS: In natural cycles, without any exogenous gonadotropins, endocrine and ultrasonographic studies of follicular maturation in 18 women with minor endometriosis (41 cycles), 15 women with unexplained infertility (31 cycles), and 34 women with tubal damage (88 cycles) were performed. RESULTS: The endometriosis group had a significantly longer follicular phase (median: 15, 13, and 13 days). Both endometriosis and unexplained infertility had significantly reduced LH concentrations in follicular fluid compared with tubal damage (median: 12.1, 11.5, and 15.9 IU/L, respectively). Endometriosis was associated with a significantly reduced fertilization rate compared with unexplained infertility or tubal damage (46, 65, and 69%, respectively). CONCLUSIONS: These data show continuing evidence of ovulatory dysfunction leading to reduced fertilization rates in women with minor endometriosis.

Antibiotic treatment based on seminal cultures from asymptomatic male partners in in-vitro fertilization is unnecessary and may be detrimental.

We questioned the policy of routine microbiological culture of semen prior to in-vitro fertilization (IVF) with a view to prescribing antibiotics to reduce the risk of introducing seminal infection into the embryo culture system. An initial retrospective study examined serum microbiology reports of 449 couples undergoing IVF or gamete intra-Fallopian transfer (GIFT). In semen samples taking >/=1 days to reach the microbiology laboratory compared with same-day delivery there was increased frequency of significant culture of enterococci (27 versus 15%, P < 0.01). In samples taking >/=2 days there was increased frequency of significant culture of Gram-negative bacilli (31 versus 12%, P < 0.01) and of overall culture of other potentially pathogenic organisms (26 versus 14%, P < 0.01). We questioned diagnostic accuracy and relevance. Therefore, in a prospective study, semen and high vaginal swabs obtained on the day of oocyte collection were cultured from 100 couples having IVF or GIFT, of whom 52 male partners had been treated with antibiotics following positive pre-IVF semen culture. The presence of bacteria in semen samples used only for IVF (n = 90) did not reduce fertilization rates nor lead to infection of the embryo culture system. However, there was an increased incidence of significant culture of vaginal Gram-negative bacilli in patients with treated partners compared with untreated partners [15/52 (29%) versus 5/48 (10%), P < 0.05]. Thus antibiotic therapy in the male partner may increase the likelihood of inoculation of antibiotic-resistant pathogenic bacteria from the vagina into the embryo culture system during vaginal oocyte collection. In asymptomatic patients, microbiological screening of semen samples prior to IVF treatment and subsequent treatment with antibiotic therapy in those with positive cultures appears to be unnecessary and may be detrimental to IVF outcome.

Reduced preovulatory granulosa cell steroidogenesis in women with endometriosis.

To examine the cause of altered follicular fluid steroid levels and lower in vitro fertilization rate observed in infertile women with minor endometriosis, we have compared the production of estradiol (aromatase activity) and progesterone of freshly isolated granulosa cells (3h. incubation) from such women and a control group with tubal or unexplained infertility, having IVF during unstimulated or gonadotropin-stimulated cycles. As previously observed, mature oocytes from women with endometriosis had a reduced fertilization and cleavage rate in vitro in unstimulated cycles (19/37[51%] vs. 69/94[73%], p < 0.05) and stimulated cycles (20/37[57%] vs. 32/39[82%], p < 0.01). Median [95%CI] basal aromatase activity was lower in endometriosis compared with control in unstimulated cycles (2.84[2.03-3.49] pmol E2/10(3) cells/3h, n = 31 vs. 3.63[2.72-3.49], n = 55, p = 0.057) and stimulated cycles (0.31[0.16-0.50], n = 14 vs. 0.99[0.70-1.52], n = 20, p < 0.001). Progesterone production followed a similar pattern in unstimulated (0.56[0.50-0.89] pmol/10(3) cells/3h, n = 29 vs. 1.23[0.69-1.54], n = 52,) and stimulated (0.37[0.20-0.73], n = 16 vs. 0.95[0.72-1.17], n = 21) cycles (p < 0.05). Addition of FSH, LH or hCG (30ng/mL) to the incubation medium enhanced progesterone production 2 to 3-fold, but had no effect on aromatase activity. Our results indicate a defect in granulosa cell steroidogenesis associated with endometriosis, which could affect oocyte function and explain the reduction in fertilizing capacity and subsequent competence of the corpus luteum, and the associated subfertility.

Pituitary-ovarian dysfunction as a cause for endometriosis-associated and unexplained infertility.

This study examines circulating and follicular hormone concentrations and fertilization of oocytes in cycles totally unperturbed by exogenous gonadotrophins in 10 women (25 cycles) with untreated minimal-mild endometriosis and nine women (23 cycles) with prolonged unexplained infertility compared with 16 women (50 cycles) with tubal damage as functional controls. Endometriosis was associated with a significantly longer follicular phase (median 15, 12, 13 days respectively) and reduced oestrogen secretion (median index area under the curve 3063, 3842, 3805 units respectively) compared with controls. Both endometriosis and unexplained infertility had significantly reduced serum luteinizing hormone (LH) surges [median peak serum (LH) 43, 39, 55 IU/l respectively and median area under the curve 661, 687, 823 units respectively] and reduced LH concentrations in follicular fluid (median 19.6, 10.6, 9.2 IU/l respectively). These findings suggest that infertility associated with minor endometriosis and of apparently unexplained aetiology share a common pathophysiology in impaired LH surge secretion. Whether that represents a primary pituitary disorder or is secondary to a defective ovarian signal is discussed.

Time-dependent effects of transforming growth factor alpha on aromatase activity in human granulosa cells.

Transforming growth factor alpha (TGF alpha) is implicated as a paracrine growth factor in the regulation of human granulosa cell function. To investigate this further, we have examined the actions of TGF alpha on the basal and follicle-stimulating hormone (FSH)-stimulated aromatase activity of human granulosa cells to determine how this growth factor influences oestrogen biosynthesis in the follicle. Granulosa cells from women having in-vitro fertilization during untreated or gonadotrophin-stimulated cycles were cultured for 1-6 days in the presence or absence of FSH or TGF alpha at a range of doses. Aromatase activity, expressed as oestradiol production, was determined after culture during a 3 h test period. After 2 days, TGF alpha (1-300 ng/ml) decreased basal and FSH-stimulated aromatase activity in a dose-dependent manner (ED50 = 3 ng/ml). In contrast, after 4 days, TGF alpha enhanced both basal and FSH-stimulated aromatase activity. Repeated experiments revealed a consistent pattern of inhibition on day 2, which was more marked in the presence of FSH (reduction by 30.6 +/- 9.1%, mean +/- SEM; n = 14; P < 0.01), and stimulation on day 4 in both the absence (increased by 61.4 +/- 20.6%, mean +/- SEM; n = 6; P < 0.05) and presence of FSH (increased by 36.0 +/- 15.2%, mean +/- SEM; n = 8; P < 0.05). The results provide further evidence that TGF alpha is a paracrine factor in the control of oestrogen biosynthesis, but the actions can be either inhibitory or stimulatory depending on the duration of exposure.