P12.03 Bi-specific T cell engagers targeting IL13Rá2 activate tumor-infiltrating lymphocytes and improve survival in pre-clinical models of glioblastoma

BACKGROUND

The outstanding efficacy of bi-specific T-cell engagers (BiTE against hematological malignancies offers hope that they can similarly target solid tumors like GBM. In this study, we have designed a BiTE protein with specificity to the tumor-associated antigen, IL13Rα2, and investigated how BiTE protein engages a host’s T cell immune response to promote anti-glioma activity in pre-clinical models of GBM.

MATERIAL AND METHODS

BiTE molecule consisting of two single chain variable regions (scFv) of antibodies against either murine or human CD3ε and scFv47 against human IL13Rα2 connected through a flexible linker (BiTEIL13Rα2) were sub-cloned in a lentiviral expression cassette. The BiTE molecule (BiTEIL13Rα2off) modified to abrogate the interaction of scFv47 with IL13Rα2 served as a negative control. The BiTE proteins were isolated from the supernatants of HEK293T cells using His-Tag affinity chromatography and validated in SDS-PAGE, Western Blot, and ELISA assays. The BiTEIL13Rα2-induced T cells activation was measured in (i) cytotoxicity assay against IL13Rα2+ glioma cells, (ii) flow cytometry measuring for CD69 and CD25 T cells’ activation markers, and (iii) the production of cytokines, IFNγ and TNFα. For in vivo analysis, VmDk and C57Bl/6 mice bearing established intracranial glioma were treated systemically with BiTE proteins. The survival of the mice was recorded and analyzed using the log-rank test.

RESULTS

Here we show that BiTEIL13Rα2 specifically binds to IL13Rα2 but not to IL13Rα1, whereas BiTEIL13Rα2off has no binding activity to both IL13 receptors. The co-culture of naïve murine or donor’s human CD3+ T cells with IL13Rα2+ glioma cells in the presence of BiTEIL13Rα2 but not with BiTEIL13Rα2off (i) activates CD3+CD8+ T cells as judged by upregulation of CD69, CD25, and production of IFNγ and TNFα and (ii) results in concentration- and antigen-dependent cytotoxicity in glioma cells. Furthermore, a direct comparison of CD3+ T cells obtained from the peripheral blood and tumor tissue of GBM patients revealed that BiTEIL13Rα2 induces a potent cytotoxic activity of CD3+ T cells against IL13Rα2+ glioma cells. Finally, treatment of immunocompetent mice bearing IL13Rα2+ murine glioma with BiTEIL13Rα2 resulted in a higher frequency of intratumoral CD8+ T cells, and significant (p<0.05) improvement of survival over a negative control, BiTEIL13Rα2off group of mice.

CONCLUSION

Our data demonstrate that BiTEIL13Rα2 protein activates CD3+ T cells in an antigen-specific fashion. Furthermore, systemic treatment with BiTEIL13Rα2 protein confers a significant survival benefit in pre-clinical syngeneic glioma models, warranting investigations in other IL13Rα2-expressing cancers and translation to clinical settings.

Fine epitope mapping of monoclonal antibodies 9B9 and 3G8 to the N domain of angiotensin-converting enzyme (CD143) defines a region involved in regulating angiotensin-converting enzyme dimerization and shedding

A panel of monoclonal antibodies (mAbs) raised against both the N and C domains of angiotensin-I-converting enzyme (ACE, peptidyl dipeptidase, EC 3.4.15.2) have been extensively mapped and have facilitated the study of various aspects of ACE structure and biology. In this study, we characterize two mAbs, 9B9 and 3G8, that recognize the N domain of ACE and that influence shedding and dimerization. Fine epitope mapping was performed, which mapped the epitopes for these mAbs to the N terminal region of the N domain where they overlap to a large extent, despite having different effects on ACE processing. The mAb 3G8 epitope appears to be shielded by the C domain and to be carbohydrate dependent as binding increased significantly as a result of underglycosylation, whereas these factors did not influence mAb 9B9 recognition. Three mutations within the overlapping region of these two epitopes, Q18H, L19E, and Q22A, which decreased mAb 3G8 binding to the soluble N domain, were introduced into full-length somatic ACE (sACE) to determine their influence on ACE expression and processing. Increased ACE expression, cell surface expression, and basal shedding were observed with all three mutations. Furthermore, cross-linking and western blotting of Chinese hamster ovary (CHO) cell lysates detected two distinct ACE dimers, a native and cross-linked dimer. Increasing amounts of the cross-linked dimer were observed for the mutant sACEQ22A, further implicating the overlapping region of the mAb 9B9 and 3G8 epitopes in ACE processing.

Epitope mapping of mAbs to denatured human testicular ACE (CD143)

Angiotensin I-converting enzyme (ACE; CD143) has two homologous enzymatically active domains (N and C) and plays a crucial role in blood pressure regulation and vascular remodeling. A wide spectrum of monoclonal antibodies (mAbs) to different epitopes on the N and C domains of human ACE have been used to study different aspects of ACE biology. In this study, we characterized a set of nine mAbs, developed against the C domain of human ACE, which recognize the denatured forms of ACE and thus are suitable for the detection and quantification of somatic ACE (sACE) and testicular ACE (tACE) using Western blotting and immunohistochemistry on paraffin-embedded human tissues. The epitopes for these mAbs were defined using species cross-reactivity, phage display library screening, Western blotting and ACE mutagenesis. Most of the mAbs recognized common/overlapping region(s) on both somatic and testicular forms of human ACE, whereas mAb 4E10 was relatively specific for the testicular isoform and mAb 5B9 mainly recognized the glycan attached to Asn 731. This set of mAbs is useful for identifying even subtle changes in human ACE conformation because of denaturation. These mAbs are also sensitive tools for the detection of human sACE and tACE in biological fluids and tissues using proteomic approaches. Their high reactivity in paraffin-embedded tissues provides opportunities to study changes in the pattern of ACE expression and glycosylation (particularly with mAb 5B9) in different tissues and cells.

Monoclonal antibodies to native mouse angiotensin-converting enzyme (CD143): ACE expression quantification, lung endothelial cell targeting and gene delivery

We demonstrated previously that the monoclonal antibody 9B9 to angiotensin-converting enzyme (ACE), which accumulates very selectively into the rat lung after systemic injection, is a powerful tool for immunotargeting of therapeutic agents or genes to the rat lung vascular bed. Bearing in mind a high research and therapeutic potential of lung targeting via ACE, we obtained a new set of rat monoclonal antibodies to different epitopes of mouse ACE in order to expand this approach to mice. Nine new monoclonal antibodies, recognizing epitopes on the N- and C-domains of catalytically active mouse ACE, were obtained and examined for their efficacy to bind ACE both in vitro and in vivo. This set of monoclonal antibodies was proved to be useful for ACE quantification (by flow cytometry and cell enzyme-linked immunosorbent assay) on the surface of different mouse ACE-expressing cells: endothelial cells, monocytes, macrophages, dendritic cells and spermatozoa. Moreover, gene delivery into mouse ACE-expressing cells using adenoviruses increased 40-fold after redirecting of these viruses to ACE (by coating these viruses with anti-ACE monoclonal antibodies). Radiolabelled (I(125)) monoclonal antibodies specifically accumulated in the mouse lung after systemic injection. Monoclonal antibodies 3G8.17, 4B10.5 and 4B10.17 demonstrated the highest level of lung uptake, 40-50% of injected dose, and high selectivity of lung uptake. Influence of monoclonal antibodies on ACE shedding was negligible, except monoclonal antibody 1D10.11. None of the tested monoclonal antibodies inhibited ACE activity in vitro. In conclusion, a new set of rat monoclonal antibodies to mouse ACE was obtained suitable to study ACE biology in mice and for ACE expression quantification on mouse cells in particular. These monoclonal antibodies also demonstrated highly efficient and selective lung accumulation and thus has the potential for targeting drugs/genes to the pulmonary vasculature in different mouse models of human lung diseases, including numerous knockout models.

Development and characterization of rat monoclonal antibodies to denatured mouse angiotensin-converting enzyme

Four new rat monoclonal antibodies, generated to denatured mouse somatic angiotensin-converting enzyme (ACE, CD143), detect mouse ACE with high sensitivity in Western blotting. Epitope mapping for the monoclonal antibodies--B12, 4G6 and 5C4--was also performed. Two monoclonal antibodies--B12 and 5C4--are directed to various epitopes on the N-domain--i.e., they recognized only the somatic isoform of mouse ACE. The monoclonal antibody H7 recognized an epitope on the C-domain of mouse ACE. The monoclonal antibody 4G6 was directed to a sequence on the N-domain of mouse ACE, which is homologous to a region of the C-domain and, as a result, also recognizes mouse testicular ACE (tACE) by means of Western blotting. In paraffin-embedded mouse tissues, all monoclonal antibodies detected all known expression sites of somatic ACE (sACE), e.g., the epithelial cells of the kidney proximal tubules, intestine and epididymis, and heterogeneously in endothelial cells. The monoclonal antibodies 4G6 and H7 additionally stained mouse tACE in spermatozoa and in mature spermatids. The monoclonal antibody 4G6 also demonstrated cross-reactivity with sACE from a broad spectrum of animal species, including human, rat, rabbit and bovine. However, this monoclonal antibody did not recognize the testicular isoform of ACE of these species. This set of monoclonal antibodies is useful for identifying even subtle changes in mouse ACE conformation because of denaturation. These monoclonal antibodies are also sensitive tools for the detection of mouse ACE in biological fluids and tissues by using proteomics approaches. Their high reactivity in paraffin-embedded tissues opens up opportunities to study possible changes in the pattern of ACE expression in knockout mouse models and may prove useful for correlating ACE expression in these models with human diseases.

Monoclonal antibodies to denatured human ACE (CD 143), broad species specificity, reactivity on paraffin sections, and detection of subtle conformational changes in the C-terminal domain of ACE

Two new mouse monoclonal antibodies (mAbs) were generated to denatured human angiotensin-converting enzyme (ACE, CD143). The clones 2E2 and 3C5, each of the IgG1 kappa chain isotype, detect ACE with high sensitivity, respectively, at 20 ng and 2 ng of protein per lane in Western blotting. They both recognize different epitopes on the C-domain of ACE located between amino acid residues 740 and 992. In formalin-fixed and paraffin-embedded human tissues, immunohistochemistry revealed all known expression sites of ACE, e.g. the epithelial brush borders of proximal kidney tubules, epithelial cells of epididymis, endothelial cells, activated macrophages as well as germ cells during spermatogenesis. In contrast to other mAbs to denatured human ACE, mAbs 2E2 and 3C5 demonstrate cross-reactivity with a broad spectrum of animal species such as monkey, rat, rabbit, cattle, dog, cat, and guinea pig. In addition, mAb 2E2 recognized mouse ACE in Western blotting and on paraffin sections. Our findings suggest that mAbs 2E2 and 3C5 are useful for identifying even subtle changes in ACE conformation resulting from denaturation. These mAbs are also sensitive tools for the detection of minimal amounts of ACE in biological fluids and tissues using proteomics approaches. Their reactivity in routinely processed tissues of various species may prove useful for correlation of ACE expression in animal models to human diseases.

Antibody-mediated lung endothelium targeting: in vivo model on primates

We have recently provided evidence that angiotensin-converting enzyme (ACE) is a rational target and anti-ACE monoclonal antibodies (mAbs) are suitable molecules for directing gene/drug delivery into the pulmonary endothelium of rodents. As a step towards gene therapy clinical trials using this approach, the present study evaluated the potential of anti-ACE mAbs for in vivo lung endothelium targeting in 10 species of primates. Cross-reactivity of 10 distinct mAbs directed to human ACE with ACE from baboon, macaques, cercopithecus and chimpanzee revealed that the highest binding with ACE from baboon and macaques was with mAb i2H5, from chimpanzee - mAb 9B9, and from human - 9B9 and i2H5. Thereafter, in vivo biodistribution of mAbs i2H5 and 9B9 was estimated in Macaca arctoides. MAb i2H5, which binds to macaque ACE with substantially higher affinity than mAb 9B9, also more effectively accumulates in their lungs than mAb 9B9. Immunospecificity of lung accumulation (mAb/control IgG ratio) was 37 for i2H5 and 0.5 for 9B9. Lung selectivity of i2H5 uptake (lung/blood ratio) was around 10. Therefore mAb i2H5 may be useful for in vivo lung targeting in non-human primates, whereas 9B9 may be most useful in primates that are closer to humans (chimpanzee). A combination of these two mAbs may be particularly useful for human clinical trials of gene/drug therapy for lung disorders such as pulmonary hypertension and lung metastases.

Point mutation in the stalk of angiotensin-converting enzyme causes a dramatic increase in serum angiotensin-converting enzyme but no cardiovascular disease

BACKGROUND

Angiotensin-converting enzyme (ACE) metabolizes many small peptides and plays a key role in blood pressure regulation. Elevated serum ACE is claimed to be associated with an increased risk for cardiovascular disease. Previously, two families with dramatically increased serum ACE were described, but no systematic survey of affected individuals was performed, and the molecular background of this trait is unknown.

METHODS AND RESULTS

Eight families were identified with autosomal dominant inheritance of a dramatic (5-fold) increase of serum ACE activity. Strikingly, no clinical abnormalities were apparent in the affected subjects. Isolated blood cells were used for genetic and biochemical analysis. The level of ACE expression on the blood leukocytes and dendritic cells and total cell-associated ACE of the affected individuals was similar to that in nonaffected relatives; however membrane-bound mutant ACE was much more efficiently clipped from the cell surface compared with its wild-type counterpart. A point mutation causing Pro1199Leu in the stalk region of the ACE molecule cosegregates with the increase in serum ACE (LOD score, 6.63).

CONCLUSIONS

A point mutation in the stalk region of the ACE protein causes increased shedding, leading to increased serum ACE, whereas cell-bound ACE is unaltered, and affected individuals exhibit no clinical abnormalities. These findings qualify the importance of serum ACE and establish a new determinant of ACE solubilization.

Combined transductional and transcriptional targeting improves the specificity of transgene expression in vivo

The promise of gene therapy for health care will not be realized until gene delivery systems are capable of achieving efficient, cell-specific gene delivery in vivo. Here we describe an adenoviral system for achieving cell-specific transgene expression in pulmonary endothelium. The combination of transductional targeting to a pulmonary endothelial marker (angiotensin-converting enzyme, ACE) and an endothelial-specific promoter (for vascular endothelial growth factor receptor type 1, flt-1) resulted in a synergistic, 300,000-fold improvement in the selectivity of transgene expression for lung versus the usual site of vector sequestration, the liver. This combined approach should be useful for the design of other gene delivery systems.

A targetable, injectable adenoviral vector for selective gene delivery to pulmonary endothelium in vivo

Adenoviral (Ad) vectors are promising gene therapy vehicles due to their in vivo stability and efficiency, but their potential utility is compromised by their restricted tropism. Targeting strategies have been devised to improve the efficacy of these agents, but specific targeting following in vivo systemic administration of vector has not previously been demonstrated. The distinct aim of the current study was to determine whether an Ad-targeting strategy could maintain fidelity upon systemic vascular administration. We used a bispecific antibody to target Ad infection specifically to angiotensin-converting enzyme (ACE), which is preferentially expressed on pulmonary capillary endothelium and which may thus enable gene therapy for pulmonary vascular disease. Cell-specific gene delivery to ACE-expressing cells was first confirmed in vitro. Administration of retargeted vector complex via tail vein injection into rats resulted in at least a 20-fold increase in both Ad DNA localization and luciferase transgene expression in the lungs, compared to the untargeted vector. Furthermore, targeting led to reduced transgene expression in nontarget organs, especially the liver, where the reduction was over 80%. Immunohistochemical and immunoelectron microscopy analysis confirmed that the pulmonary transgene expression was specifically localized to endothelial cells. Enhancement of transgene expression in the lungs as a result of the ACE-targeting strategy was also confirmed using a new noninvasive imaging technique. This study shows that a retargeting approach can indeed specifically modify the gene delivery properties of an Ad vector given systemically and thus has encouraging implications for the further development of targetable, injectable Ad vectors.

Cyclic adenosine monophosphate regulates the expression of the intercellular adhesion molecule and the inducible nitric oxide synthase in brain endothelial cells

The authors studied whether cyclic AMP (cAMP), a widespread regulator of inflammation, modulates the cytokine-mediated expression of the intercellular adhesion molecule, intercellular adhesion molecule-1 (ICAM-1), and the inflammatory nitric oxide synthase 2 (NOS-2), in primary and immortalized brain endothelial cell cultures (GP8.3 cell line). When measured by enzyme-linked immunosorbent assay (ELISA), ICAM-1 was constitutively expressed and was up-regulated twofold by interleukin-1beta, with no effect of interferon-gamma. The NOS-2 activity, assessed by nitrite accumulation, was absent from untreated cultures but was induced by interleukin-1beta and interferon-gamma acting synergistically. Stimulation of cAMP-dependent pathways with forskolin or dibutyryl cAMP decreased ICAM-1 protein expression, whereas it increased NOS-2 protein expression. For both ICAM-1 and NOS-2, mRNA expression correlated with protein expression. Blockade of NOS activity with L-N-monomethylargiuine (L-NMMA) did not alter ICAM-1 expression, indicating that the nitric oxide released by NOS-2 did not cause the down-regulation of ICAM-1. Analysis of NFKB activation indicated that cAMP acted through a mechanism other than inhibition of nuclear translocation of NFKB. The authors conclude that cAMP modulates the expression of proinflammatory molecules in brain endothelium. This suggests that inflammatory processes at the blood-brain barrier in vivo may be regulated by perivascular neurotransmitters via cAMP.

Immunotargeting of catalase to ACE or ICAM-1 protects perfused rat lungs against oxidative stress

The pulmonary endothelium is susceptible to oxidative insults. Catalase conjugated with monoclonal antibodies (MAbs) against endothelial surface antigens, angiotensin-converting enzyme (MAb 9B9) or intercellular adhesion molecule-1 (MAb 1A29), accumulates in the lungs after systemic injection in rats (V. Muzykantov, E. Atochina, H. Ischiropoulos, S. Danilov, and A. Fisher. Proc. Natl. Acad. Sci. USA 93: 5213-5218, 1996). The present study characterizes the augmentation of antioxidant defense by these antibody-catalase conjugates in isolated rat lungs perfused for 1 h with catalase conjugated with either MAb 9B9, MAb 1A29, or control mouse IgG. Approximately 20% of the injected dose of Ab-125I-catalase accumulated in the perfused rat lungs (vs. <5% for IgG-125I-catalase). After elimination of nonbound material, the lungs were perfused further for 1 h with 5 mM hydrogen peroxide (H2O2). H2O2 induced an elevation in tracheal and pulmonary arterial pressures (126 +/- 7 and 132 +/- 5%, respectively, of the control level), lung wet-to-dry weight ratio (7.1 +/- 0.4 vs. 6.0 +/- 0.01 in the control lungs), and ACE release into the perfusate (436 +/- 20 vs. 75 +/- 7 mU in the control perfusates). Both MAb 9B9-catalase and MAb 1A29-catalase significantly attenuated the H2O2-induced elevation in 1) angiotensin-converting enzyme release to the perfusate (215 +/- 14 and 217 +/- 38 mU, respectively), 2) lung wet-to-dry ratio (6.25 +/- 0.1 and 6.3 +/- 0.3, respectively), 3) tracheal pressure (94 +/- 4 and 101 +/- 4%, respectively, of the control level), and 4) pulmonary arterial pressure (103 +/- 3 and 104 +/- 7%, respectively, of the control level). Nonconjugated catalase, nonconjugated antibodies, nonspecific IgG, and IgG-catalase conjugate had no protective effect, thus confirming the specificity of the effect of MAb-catalase. These results support a strategy of catalase immunotargeting for protection against pulmonary oxidative injury.

Modulation of angiotensin-converting enzyme in cultured human vascular endothelial cells

Previous work has suggested that not all immunoreactive angiotensin-converting enzyme (ACE) in tissues or cells is in a biologically active state. We have explored this possibility in cultured human umbilical vein endothelial cells (HUVEC), one of the most widely studied in vitro endothelial cell systems. Our approach included characterization of the effect of increasing passage number on ACE activity and expression of immunoreactive ACE at the single cell level, the subcellular compartmentalization of active ACE, and the effect of phorbol ester (PMA) treatment. We found that both ACE activity and expression of ACE antigen were downregulated by cultivation (30% of ACE-positive cells at seventh passage vs. 90% in primary culture). ACE downregulation is specific (number of CD31-positive cells did not change with cultivation) and correlated with downregulation of factor VIII-antigen. The percentage of ACE-positive cells in permeabilized HUVEC at third passage was almost twice that in nonpermeabilized HUVEC (90% vs. 50%), indicating that HUVEC contain intracellular immunoreactive ACE. ACE activity, however, was similar when measured in intact cells and in cell lysates. Moreover, diazonium salt of sulfanilic acid (DASA), a membrane-impermeable ACE inhibitor, inhibited ACE activity in intact cells and in cell lysates at the same extent, thus implying that intracellular ACE is inactive. PMA (100 nM) treatment increased the percentage of ACE-positive cells at third passage from 57 to 96%. ACE activity was increased 3-fold in cell and 1.5-fold in the culture medium of PMA-treated cells. Analysis of ACE activity in intact monolayers and cell lysates of control and PMA-treated cells revealed that all enzymatically active ACE in PMA-treated cells is localized on the plasma membrane and acts as an ectoenzyme. We conclude that expression of ACE by HUVEC is downregulated by repeated passage in culture but can be restored by PMA treatment. In addition, ACE expression is heterogeneous between neighboring cells, and total immunoreactive ACE protein associated with HUVEC includes an inactive pool of the enzyme.