Identification of a novel membrane protein, HP59, with therapeutic potential as a target of tumor angiogenesis.

CM101, a polysaccharide isolated from the culture medium of Group B streptococcus, a neonatal pathogen, targets pathological angiogenesis and inhibits tumor growth in mice and humans. CM101 also targets neonatal lung and adult sheep lung endothelial cells. A gene encoding a transmembrane protein that interacts with CM101 was isolated from a sheep lung endothelial cell cDNA library. The gene, termed sp55, encodes a 495-amino acid polypeptide. COS-7 cells transfected with a vector containing sp55 express the SP55 protein-bound CM101 in a concentration-dependent manner. Stably transfected CHO cells also bound CM101. The corresponding human gene, hp59, was isolated from a human fetal lung cDNA library and had a predicted identity to SP55 of 86% over 495 amino acids. HP59 protein was shown by immunohistochemistry to be present in the pathological tumor vasculature of the lung, breast, colon, and ovary, but not in the normal vasculature, suggesting that the protein may be critical to pathological angiogenesis. The hp59 gene and/or the HP59 protein was not expressed in a variety of normal tissues, but was significantly expressed in human fetal lung, consistent with the pathophysiology of Group B streptococcus infections in neonates. Mice immunized with HP59 and SP55 peptides showed significant attenuation of tumor growth. Immunization effectively inhibited both the tumor angiogenesis and vasculogenesis processes, as evidenced by lack of both HP59- and CD34-positive vessels. These results and the immunohistochemistry data suggest a therapeutic potential for the CM101 target protein HP59 both as a drug target and as a vaccine against pathoangiogenesis.

CM101 stimulates cutaneous wound healing through an anti-angiogenic mechanism.

CM101, an anti-pathoangiogenic polysaccharide derived from group B streptococcus, has been shown to inhibit inflammatory angiogenesis and accelerate wound healing in a mouse model and minimize scarring/gliosis following spinal cord injury. To evaluate the in vivo effects of CM101 on cutaneous wound healing in the pig, intravenously delivered CM101 or placebo vehicle was given 1 h after cutaneous wounding and again at 72 h after injury. Tissues from partial-thickness and full-thickness excisions were collected at days 4 and 7 after wounding and evaluated for a variety of standard healing parameters. Both types of CM101-treated wounds showed significantly less evidence of inflammatory angiogenesis when assessed by macroscopic photography of the wound surface, qualitative histological observations, laser doppler perfusion imaging, and quantitative morphometric analysis of microvessel area from endothelium selectively immunostained for factor VIII. Resurfacing was accelerated in partial-thickness and full-thickness excisions that received two doses of CM101 as compared to the placebo-treated excisional wounds. Neodermal thickness was increased in CM101-treated wounds at day 4 and was slightly reduced in comparison with placebo by day 7. New collagen accumulation appeared to be unaffected by the CM101 treatment. Immunohistochemical staining using a polyclonal antisera directed against the anti-pathoangiogenic CM101 target protein HP59 on day 7 indicated a strong immunoreactivity on the microvessels present in the control wounds but not in wounds of the CM101-treated animals. In summary, the immunolocalization HP59 in the microvessels of the cutaneous wound bed in control but not in CM101 treated wounds suggests that CM101 inhibits the pathologic inflammatory angiogenesis accompanying the normal granulation processes. The net biological effect of inhibited inflammatory pathoangiogenesis is a diminished, suggested and purely physiologic, microvascular bed which translates into an enhanced rate of epithelial resurfacing and therefore an overall accelerated rate of wound repair.

CM101 treatment overrides tumor-induced immunoprivilege leading to apoptosis.

CM101, a bacterial polysaccharide exotoxin produced by group B Streptococcus (GBS), also referred to as GBS toxin, has been shown to target pathological neovasculature and activate complement (C3), thereby inducing neovascularitis, infiltration of inflammatory cells, inhibition of tumor growth, and apoptosis in murine tumor models. Data from refractory cancer patients in a Phase I clinical trial with CM101 indicated a similar mechanism of tumor-targeted inflammation. To further our understanding of the mechanism of action of CM101 as an antitumor agent, we examined the role of the inflammatory response in inducing tumor apoptosis in a normal mouse and tumor-bearing mouse model. The i.v. infusion of CM101 into B16BL-6 melanoma tumor-bearing mice elevated p53 mRNA in circulating leukocytes as measured by reverse transcription-PCR, and immunohistochemistry demonstrated infiltration and sequestration of leukocytes. Whole tumor lysates from excised tumors exhibited an increase in binding to the murine p21(Waf1/Cip1) derived p53 DNA binding sequence compared with control whole tumor lysates, in which minimal or no DNA binding was observed. CM101 infusion led to elevated levels of Fas protein within the tumors as well as a decrease in the expression of fas ligand (fasL). Furthermore, tumors were apoptotic as determined by terminal deoxynucleotidyl transferase-mediated nick end labeling and DNA fragmentation assays. Collectively, these data suggest that CM101 up-regulates p53 in tumor-infiltrating leukocytes, initiating a loss of tumor immunoprivilege and consequently rendering the tumor sensitive to Fas/fasL-mediated apoptosis. CM101 induced loss of tumor immunoprivilege through changes in the expression of leukocyte p53, tumor Fas and fasL coupled with neovascularitis and leukocyte infiltration, constitutes a plausible molecular pathway for tumor reduction observed in cancer patients.

Isolation and identification of the group B streptococcal toxin CM101 from infants with sepsis.

OBJECTIVE: To determine whether the group B streptococcal (GBS) polysaccharide exotoxin CM101, which induces a complement-activated cytokine-driven inflammatory response, is present in body fluids of infants with GBS disease. STUDY DESIGN: With a sandwich enzyme-linked immunosorbent assay, CM101 was measured in plasma, urine, and cerebrospinal fluid from newborn infants who were evaluated for possible infection and from older infants with culture-confirmed GBS disease. RESULTS: Urine from 11 newborn infants with culture-confirmed early-onset disease contained large amounts of CM101 (1.0 to 5.5 mg/48 h). Plasma concentrations were 62.6 +/- 10.5 microg/mL in these infants and were 69.0 +/- 21.2 microg/mL in 4 older infants with late-onset disease. Plasma CM101 concentrations did not correlate with indexes of illness severity, leukocyte counts, or interleukin-6 or interleukin-8 plasma concentrations. CM101 was present in cerebrospinal fluid of 5 infants with meningitis (8.4 +/- 1.6 microg/mL). CM101 was not found in control samples. CM101 isolated from urine had molecular weight and sugar composition similar to those obtained from GBS culture media, and they both elicited a comparable pathophysiologic response when infused intravenously in lambs. CONCLUSIONS: CM101 is present in infants with GBS disease, and it appears to be the same as CM101 obtained from GBS culture media.

Quantified color Doppler sonography of tumor vascularity in an animal model.

This study was designed to evaluate the accuracy of a system to quantitate tumor vascularity with amplitude (power) color Doppler sonography two- and three-dimensionally. The vascularity of 20 transplanted murine tumors was determined with quantitated amplitude color Doppler sonography both two- and three-dimensionally and compared to tumor vascularity estimated by histologic examination. Serial examinations were performed 15, 30, 45, and 60 min after the injection of the exotoxin CM-101 and saline solution to assess changes in tumor vascularity. Three-dimensional amplitude color Doppler sonography best depicted the overall vascularity of tumor when compared to histologic estimation of vessel density. However, neither two- nor three-dimensional amplitude color power angiography correlated well to the microvessel count, probably a reflection of the difference in the method for vessel quantification using sonographic versus histologic techniques. Three-dimensional amplitude Doppler sonography correlated better with counts of large vessels (> 100 microm) as opposed to small vessels (> 15 microm). Time-activity curves showed no difference in tumor flow at the times measured in the experimental group injected with CM-101 or when compared to saline solutions in either the peripheral or central portions of the tumor. This three-dimensional amplitude color Doppler sonographic system affords global quantification of tumor vascularity and flow that may, in turn, be useful in determining the probability of malignancy (by determination of branching patterns and vessel regularity) or tumor response or both to treatment.

CM101-mediated recovery of walking ability in adult mice paralyzed by spinal cord injury.

CM101, an antiangiogenic polysaccharide derived from group B streptococcus, was administered by i.v. injection 1 hr post-spinal-cord crush injury in an effort to prevent inflammatory angiogenesis and gliosis (scarring) in a mouse model. We postulated that gliosis would sterically prevent the reestablishment of neuronal connectivity; thus, treatment with CM101 was repeated every other day for five more infusions for the purpose of facilitating regeneration of neuronal function. Twenty-five of 26 mice treated with CM101 survived 28 days after surgery, and 24 of 26 recovered walking ability within 2-12 days. Only 6 of 14 mice in the control groups survived 24 hr after spinal cord injury, and none recovered function in paralyzed limbs. MRI analysis of injured untreated and treated animals showed that CM101 reduced the area of damage at the site of spinal cord compression, which was corroborated by histological analysis of spinal cord sections from treated and control animals. Electrophysiologic measurements on isolated central nervous system and neurons in culture showed that CM101 protected axons from Wallerian degeneration; reversed gamma-aminobutyrate-mediated depolarization occurring in traumatized neurons; and improved recovery of neuronal conductivity of isolated central nervous system in culture.

Functional studies on the anti-pathoangiogenic properties of CM101.

Group B streptococcus (GBS) isolated from human neonates diagnosed with sepsis and respiratory distress produces a polysaccharide exotoxin (CM101) which has been previously described as GBS toxin. CM101 infused i.v. into tumor-bearing mice causes rapid tumor neovascularitis, infiltration of inflammatory cells, inhibition of tumor growth and tumor apoptosis. CM101 has successfully completed phase I studies in refractory cancer patients with very encouraging results. We have now demonstrated a mechanism of action for CM101. Using a normal mouse tumor model, we have examined tumor and normal tissues which were harvested at 0, 5, 15, 30 and 60min post-infusion of either CM101 or dextran. We present evidence that CM101 is rapidly (within the first 5min) bound to the tumor neovasculature. Complement is activated by the alternative pathway (C3) and leukocytes start to infiltrate the tumor within the first 5min. Through RT-PCR and immunohistochemical techniques, we demonstrate that proinflammatory cytokines, interleukin-6 and tumor necrosis factor (TNF)-alpha, are up-regulated in infiltrating leukocytes and TNF receptor 2 is up- regulated in the targeted tumor neovasculature. Combined, these events constitute possible explanations for the observed pathophysiology of tumor ablation.

Phase I study of the antineovascularization drug CM101.

CM101 is a bacterial polysaccharide that induces neovascular inflammation in malignant tumors. Fifteen patients with refractory malignancies received CM101 i.v. by a 15-min infusion every other day, three times in 1 week, at doses ranging from 1 unit (7.5 microgram)/kg to 5 units/kg. Serum was analyzed for anti-CM101 IgG and IgM weekly. Plasma levels of inflammatory cytokines, including tumor necrosis factor alpha, interleukin 8, interleukin 10, MIP-1alpha, and soluble E-selectin, were analyzed from -15 min to 12 h during each treatment. Dose-limiting toxicities, including grade IV dyspnea and arrhythmia, were encountered at the 5-unit/kg level. Toxicities occurred primarily within the first 12 h after therapy and included mild-to-moderate fever and chills, nausea, cough, headache, facial flushing, dyspnea, myalgias, and acute tumor-related pain. No patient developed detectable antibodies to CM101. All patients experienced marked time- and dose-dependent elevations in all cytokines studied. Three patients experienced tumor shrinkage. The results show that CM101 can be safely administered at doses that produce evidence for severe, and possibly tumor-specific, inflammation. Further study is necessary to better characterize the mechanism of action and determine the optimal dose and schedule of this new agent.

Soluble E-selectin in cancer patients as a marker of the therapeutic efficacy of CM101, a tumor-inhibiting anti-neovascularization agent, evaluated in phase I clinical trail.

A polysaccharide toxin, GBS toxin, is produced by group B Streptococcus (GBS) isolates from neonates who died of "early-onset disease". GBS toxin, named CM101 in the clinic, was hypothesized, on the basis of our previous in vivo studies, to induce inflammation in pulmonary neovasculature in neonates by cross-linking of embryonic receptors still expressed after birth and in tumor neovasculature in adults. Immunohisto chemical in vitro analysis of human biopsies showed that tumor neovasculature is indeed a binding site for CM101. In vivo studies in mice have demonstrated that CM101 induced inflammatory responses in neoplastic tumor neovasculature causing inhibition of tumor growth and tumor cell necrosis. These experimental observations warranted a phase I clinical trial for CM101 as an anti-neovascularization agent in human cancer therapy. Cancer patients received one cycle of therapy consisting of three treatments during 1 week. CM101 was administered over 15 min by i.v. infusion. Dosages of 7.5 micrograms/kg (1 U/kg), n = 3; 15 micrograms/kg (2 U/kg), n = 6; 24.75 micrograms/kg (3.3 U/kg), n = 3; and 37.5 micrograms/kg (5 U/kg), n = 3 were used. Enzyme-linked immunosorbent sandwich assays (ELISA) of the patients sera showed a marked elevation of soluble E-selectin with a peak concentration observed at 8-12 h after each CM101 infusion. The average baseline value for soluble E-selectin prior to the first treatment was 97.3 +/- 23.4 ng/ml (mean +/- SEM, n = 15) and the average peak level at 8 h was 441.6 +/- 62.4 (mean +/- SEM, n = 15; P < 0.001). Subsequent treatments gave average maximum soluble E-selectin levels again at 8 h of 466.9 +/- 87.6 and 412.0 +/- 67.8 ng/ml, for treatments 2 and 3 respectively. Baseline values for treatments 2 and 3 were 192.3 +/- 26.4 and 226.4 +/- 26.1 ng/ml respectively (p < 0.01 versus treatment 1). Out of 15 patients, 5 showed tumor reduction or stabilization and were given additional cycles of therapy. CM101 induced an increase in soluble E-selectin levels, which remained elevated over baseline at the start of the following treatment cycles. The baseline remained elevated for several weeks after the final treatment, i.e., P < 0.01 for levels before treatment 1 compared to those at week 4 after treatment. Elevated soluble E-selectin is considered proof of endothelial engagement in an inflammatory process. Our data support the contention that the inflammatory response observed in these cancer patients is targeting the tumor neovasculature and that measurement of soluble E-selectin levels in patients treated with CM101 can provide important information on the magnitude of CM101-mediated neovascular endothelial activation and tumor cell damage in cancer of endothelial origin, or cancer with a major neo-angiogenic component.

Quantification of tumor vascularity and flow with amplitude color Doppler sonography in an experimental model: preliminary results.

This study was designed to evaluate a system to quantitate vascularity and tumor blood flow with amplitude (power) color Doppler sonography. The vascularity of nine transplanted murine tumors was determined with quantitated amplitude color Doppler sonography and compared to tumor vascularity estimated by histologic examination. The system used seemed to provide an accurate depiction of the vascularity of tumor vis-àa-vis histologic estimation of vessel density (r = 0.80). Time-activity curves showed greater flow in the experimental group injected with an exotoxin than in the group injected with saline solution. Vascular density quantification with amplitude color Doppler sonography also was more accurate when an intravascular agent (such as an exotoxin) was used than when saline infusions were given. This quantification scheme may allow the development of a system to assess the probability of malignancy and to monitor tumor response to treatment on the basis of the vascularity of the mass.

Acute inflammatory changes in subcutaneous microtumors in the ears of mice induced by intravenous CM101 (GBS toxin).

CM101, a bacterial polysaccharide derived from group B streptococcus, induces pronounced inflammatory changes in and around tumor blood vessels 60 min after i.v. injection. A technique has been developed for implanting small numbers of tumor cells in the ear skin of mice. This allows macroscopic examination of the tumor and its supporting blood vessels as it reaches the 10000 cell size and greater. Treatments can be monitored in this model for effects on small "metastatic-like" tumor nodules by direct observation and by histological examination. Inflammatory changes were indicated by increased numbers of polymorphonuclear leukocytes (PMN) adjacent to and marginating within thin-walled blood vessels and within the tumor tissue. PMN were seen in the process of migrating through venules and enlarged capillaries, each with prominent endothelial cells. Tumor morphology was variable with evidence of occasional single necrotic cells. This contrasted with tumors in ears of dextran-treated or untreated mice, which had uniform tumor morphology, and acute inflammatory cells were rarely present.

CM101, a polysaccharide antitumor agent, does not inhibit wound healing in murine models.

CM101 (previously called GBS toxin), a new anticancer polysaccharide that induces inflammatory reactions in neovasculature of tumors, does not cause similar reactions in neovasculature of healing wounds. It appears that treatment with CM101 will not interfere with normal wound healing in cancer patients.

Identification of two cardiac glycosides as Na(+)-pump inhibitors in rat urine and diet.

Endogenous Na(+)-pump specific inhibitors are present in the plasma, urine, and tissues of humans and animals. To date, the source of these inhibitors has not been rigorously defined. In the present study, large amounts of several Na(+)-pump specific inhibitors have been demonstrated to exist in the urine of rats raised on a regular chow diet and tap water. All of the inhibitor levels have been found to increase 1.5-8-fold by the surgical preparation of reduced renal mass (RRM) and one-kidney, one-clip (IK, IC) hypertension. These urinary inhibitors, however, except for the ouabain-like inhibitor which eluted from a high performance liquid chromatography C18 column at the same retention time as [3H]ouabain, disappeared within a week after switching the diet from regular diet (number 5001, PMI Feeds, Inc.) to pure synthetic diet (number 5755). The urinary level of the ouabain-like inhibitor decreased to only one-half of the level in the control rat raised on a regular diet. Two of these inhibitors were purified from both urine and diet by a combination of Amberlite XAD-2 adsorption chromatography, reverse phase low pressure liquid chromatography, and several high performance liquid chromatographies. Reverse phase high performance liquid chromatography, liquid secondary ion and gas-liquid mass spectrometries, and proton nuclear magnetic resonance spectroscopy identified these inhibitors as a stereoisomer of convalloside, probably neoconvalloside, and a mono-rhamnoside of periplogenin or its stereoisomer. These cardiac glycosides exhibited inhibitory potencies comparable to ouabain against ouabain-displacement from Na+,K(+)-ATPase and against 86Rb uptake into human erythrocytes, and they also exhibited cross-reactivity to anti-ouabain antibodies and anti-digoxin antibodies. These results clearly demonstrate that the principal source of most of the inhibitors in rat urine is the diet. The results suggest that the ouabain-like inhibitor may be derived from an endogenous origin.

Effects of group B Streptococcus toxin on long-term survival of mice bearing transplanted Madison lung tumors.

GBS toxin is a polysaccharide exotoxin produced by group B Streptococcus. This organism causes sepsis and respiratory distress in human neonates (so-called early onset disease). This disease is marked by a strong inflammatory response only in the lung, with pulmonary sequestration of granulocytes and extensive capillary endothelial damage, and occurs only during the first few days after birth. We have found that a similar inflammatory response can be induced by i.v. infusion of picomole quantities of GBS toxin in the developing vasculature of transplanted tumors in mice and can significantly retard the tumor growth. When optimum treatment with GBS toxin was started shortly after tumor implantation, a majority of tumors in the mice regressed and the mice remained tumor-free for over 5 months. Some tumors regressed in mice receiving short-term treatment with GBS toxin, but recurred after the treatment was stopped. Median survival times were extended by all regimens and all doses of GBS toxin tested. No evidence of toxicity to the vasculature of other tissues was observed. GBS toxin is being tested for cancer therapy in humans.

Antitumor effects of GBS toxin: a polysaccharide exotoxin from group B beta-hemolytic streptococcus.

A group B streptococcus (GBS) isolated from human neonates diagnosed with sepsis and respiratory distress ("early-onset disease") produces a polysaccharide exotoxin (GBS toxin) that, when infused in sheep, causes lung pathophysiology similar to that seen in humans. Histological studies have demonstrated that GBS toxin induces a strong inflammatory response in the lung, with pulmonary sequestration of granulocytes and extensive capillary endothelial damage. The susceptibility of humans to GBS toxin is age-dependent and limited to about 4 days after birth. It is rarely evident thereafter. This suggests that the binding of GBS toxin to the target endothelium occurs via specific components in the developing lung endothelial cells of the newborn that are later lost. We report here that GBS toxin can also bind to developing endothelium associated with neoplasia and induce an inflammatory response. GBS toxin was shown by immunohistochemistry to bind to capillary endothelium of human large-cell carcinomas. In nude mice bearing human tumor xenografts, intravenously administered GBS toxin caused tumor necrosis and hemorrhagic lesions, and substantially inhibited the rate of growth of the tumors. In BALB/c mice bearing Madison lung tumors, GBS toxin induced an inflammatory response resulting in marked changes in tumor morphology, including vasodilation, endothelial and tumor cell necrosis, invasion of lymphocytes and macrophages, and capillary thrombosis. In these tumor models, no evidence of toxicity to the vasculature of other tissues was observed. The reported pathophysiology of GBS in human neonates, the lack of disease in non-neonates colonized with GBS, and these results suggest that GBS toxin may have potential as a well tolerated agent in cancer therapy of some human tumors.

Monomer sequence determination of carbohydrates using fast-atom bombardment mass spectrometry of periodate-oxidized acetate ester derivatives.

A derivatization method, adapted from that of Angel et al. (ref. 10), for sequencing sugar residues in partially degraded poly- and oligo-saccharides using positive-ion f.a.b.-m.s. is described. Derivative selection provides sequence information by directing fragmentation exclusively to both sides of glycosidic oxygen atoms and, in the case of opened rings, between glycosidic carbon and ring oxygen atoms. Polysaccharides or oligosaccharides are subjected to sequential periodate oxidation, borodeuteride reduction, and acetylation. The derivatized polysaccharides are then subjected to partial degradation, acetylation, and high-performance liquid chromatography (h.p.l.c.) purification. F.a.b.-m.s. data obtained on model compounds, using 3-nitrobenzyl alcohol as matrix for f.a.b.-m.s., demonstrated direction of fragmentation to both sides of the glycosidic oxygen atom in unoxidized residues, and to both sides of the acetal oxygen atoms in oxidized residues. Oligosaccharide linkage and sequence may thus be determined by observing fragmentation from both the reducing and non-reducing ends of the molecule. Two Salmonella lipopolysaccharides, derivatized by this procedure, were partially hydrolyzed and then acetylated. Analysis of the h.p.l.c.-purified oligosaccharide derivatives by f.a.b.-m.s. demonstrated the applicability of the technique for sequencing nmol quantities of branched structures.

Structures of asparagine-linked oligosaccharides of the glycoprotein fetuin having sialic acid linked to N-acetylglucosamine.

In the accompanying paper (Bendiak et al., 1989), the separation of a series of oligosaccharides released from asparagine residues of fetuin was described. A series of NMR experiments, which included one- and two-dimensional nuclear Overhauser enhancement, two-dimensional correlation spectroscopy, and two-dimensional relayed-coherence spectroscopy, as well as permethylation analyses, established a Gal beta 1----3(NeuAc alpha 2----6)GlcNAc beta 1----4Man unit common to a series of purified structures. These oligosaccharides contained either three, four, or five glycosidically linked sialic acid residues. The NeuAc residue in alpha 2----6 linkage to GlcNAc gives rise to diagnostic chemical shift perturbations of particular proton signals in the oligosaccharides.