Background: The mucin MUC16 and the glycosylphosphatidylinositol anchored glycoprotein mesothelin likely facilitate the peritoneal metastasis of ovarian tumors. The biochemical basis and the kinetics of the binding between these two glycoproteins are not clearly understood. Here we have addressed this deficit and provide further evidence supporting the role of the MUC16-mesothelin interaction in facilitating cell-cell binding under conditions that mimic the peritoneal environment. Results: In this study we utilize recombinant-Fc tagged human mesothelin to measure the binding kinetics of this glycoprotein to MUC16 expressed on the ovarian tumor cell line OVCAR-3. OVCAR-3 derived sublines that did not express MUC16 showed no affinity for mesothelin. In a flow cytometry-based assay mesothelin binds with very high affinity to the MUC16 on the OVCAR-3 cells with an apparent K_d of 5–10 nM. Maximum interaction occurs within 5 mins of incubation of the recombinant mesothelin with the OVCAR-3 cells and significant binding is observed even after 10 sec. A five-fold molar excess of soluble MUC16 was unable to completely inhibit the binding of mesothelin to the OVCAR-3 cells. Oxidation of the MUC16 glycans, removal of its N-linked oligosaccharides, and treatment of the mucin with wheat germ agglutinin and erythroagglutinating phytohemagglutinin abrogates its binding to mesothelin. These observations suggest that at least a subset of the MUC16-asscociated N-glycans is required for binding to mesothelin. We also demonstrate that MUC16 positive ovarian tumor cells exhibit increased adherence to A431 cells transfected with mesothelin (A431-Meso⁺). Only minimal adhesion is observed between MUC16 knockdown cells and A431-Meso⁺ cells. The binding between the MUC16 expressing ovarian tumor cells and the A431-Meso⁺ cells occurs even in the presence of ascites from patients with ovarian cancer. Conclusion: The strong binding kinetics of the mesothelin-MUC16 interaction and the cell adhesion between ovarian tumor cells and A431-Meso+ even in the presence of peritoneal fluid strongly support the importance of these two glycoproteins in the peritoneal metastasis of ovarian tumors. The demonstration that N-linked glycans are essential for mediating mesothlein-MUC16 binding may lead to novel therapeutic targets to control the spread of ovarian carcinoma.

Recombinant immunotoxins (RITs) are hybrid proteins used to treat cancer. These proteins are composed of an Fv that reacts with cancer cells joined to a portion of Pseudomonas exotoxin A, which kills the cell. Because the toxin is a foreign protein, it can induce neutralizing antibodies and thereby limit the number of doses a patient can receive. We previously identified seven major mouse B-cell epitopes in the toxin, and subsequently silenced them using point mutations that converted large hydrophilic amino acids to alanine, yet retained full antitumor activity. Here we present results in which we identify and silence human B-cell epitopes in the RIT HA22. We obtained B cells from patients with antibodies to RITs, isolated the corresponding variable fragments (Fvs), and constructed a phage-display library containing Fvs that bind to the RITs. We then used alanine scanning mutagenesis to locate the epitopes. We found that human and mouse epitopes frequently overlap but are not identical. Most mutations that remove mouse epitopes did not remove human epitopes. Using the epitope information, we constructed a variant immunotoxin, HA22-LR-LO10, which has low reactivity with human antisera, yet has high cytotoxic and antitumor activity and can be given to mice at high doses without excess toxicity. The toxin portion of this RIT (LR-LO10) can be used with Fvs targeting other cancer antigens and is suitable for clinical development.

The tumor microenvironment plays a critical role in controlling tumor progression and immune surveillance. We produced an immunotoxin (2E4-PE38) that kills mouse cells expressing CD25 by attaching the Fv portion of monoclonal antibody 2E4 (anti-mouse CD25) to a 38-kDa portion of Pseudomonas exotoxin A. We employed three mouse cancer tumor models (AB1 mesothelioma, 66c14 breast cancer, and CT26M colon cancer). Tumors were implanted at two sites on BALB/c mice. On days 5 and 9, one tumor was directly injected with 2E4-PE38, and the other was not treated; 2E4-PE38 produced complete regressions of 85% of injected AB1 tumors, 100% of 66c14 tumors, and 100% of CT26M tumors. It also produced complete regressions of 77% of uninjected AB1 tumors, 47% of 66c14 tumors, and 92% of CT26M tumors. Mice with complete regressions of 66c14 tumors were immune to rechallenge with 66c14 cells. Mice with complete regressions of AB1 or CT26M tumors developed cross-tumor immunity rejecting both tumor types. Injection of anti-CD25 antibody or a mutant inactive immunotoxin were generally ineffective. Tumors were analyzed 3 days after 2E4-PE38 injection. The number of regulatory T cells (Tregs) was significantly reduced in the injected tumor but not in the spleen. Injected tumors contained an increase in CD8 T cells expressing IFN-γ, the activation markers CD69 and CD25, and macrophages and conventional dendritic cells. Treatment with antibodies to CD8 abolished the antitumor effect. Selective depletion of Tregs in tumors facilitates the development of a CD8 T cell-dependent antitumor effect in three mouse models. Significance Regulatory T cells (Tregs) protect cancer cells from immune attack. Tregs express CD25 on their surface and can be killed by antibodies and immunoconjugates targeting CD25. These agents have not been effective in controlling tumor growth, probably because they also kill cytotoxic CD8 T cells expressing CD25 that are needed for antitumor activity. To overcome these deficiencies, we have directly injected tumors with a potent anti-CD25 immunotoxin, 2E4-PE38, that kills CD25-expressing Treg cells in the tumor, causes injected and distant tumors to regress, and induces antitumor immunity. Because 2E4-PE38 is present at very low concentrations in the blood, it does not kill CD25-expressing cells at distant sites. This strategy could be employed for treating tumors in people.

Immunogenicity remains the “Achilles9 heel” of protein-based therapeutics. Anti-drug Abs produced in response to protein therapeutics can severely limit both the safety and efficacy of this expanding class of agent. In this article, we report that monotherapy of mice with tofacitinib (the JAK inhibitor) quells Ab responses to an immunotoxin derived from the bacterial protein Pseudomonas exotoxin A, as well as to the model Ag keyhole limpet hemocyanin. Thousand-fold reductions in IgG1 titers to both Ags were observed 21 d post immunization. In fact, suppression was evident for all IgG isotypes and IgM. A reduction in IgG3 production was also noted with a thymus-independent type II Ag. Mechanistic investigations revealed that tofacitinib treatment led to reduced numbers of CD127⁺ pro–B cells. Furthermore, we observed fewer germinal center B cells and the impaired formation of germinal centers of mice treated with tofacitinib. Because normal Ig levels were still present during tofacitinib treatment, this agent specifically reduced anti-drug Abs, thus preserving the potential efficacy of biological therapeutics, including those used as cancer therapeutics.

Protein therapeutics, such as antibodies, enzymes and toxins, are very promising reagents for the treatment of human disease. However, many therapeutic proteins are known to elicit immune responses when administered to humans. Certain antibodies work by neutralization; others reduce drug efficacy. It is clear that helper T cells are an important factor in the development of class-switched and affinity-maturated anti-therapeutic protein antibodies. Elimination of the T cell epitope seems reasonable, but it is probably impossible to remove all T cell epitopes from protein drugs because T cell epitopes are closely related to the major histocompatibility complex (MHC) molecules, which are known to be highly polymorphic. Accordingly, a possible practical approach for reducing immunogenicity involves the removal of B cell epitopes. In this case, reducing the affinity between the antigen and the B cell receptor may reduce B cell activation, even though the T cell will still be activated. Also a B cell epitope is not restricted by MHC class II molecules. This review seeks to address the identification and the characterization of B cell epitopes, and reports on the development of strategies for reducing immune response with modified B cell epitopes.

Recombinant immunotoxins (RITs) are fusions of an Fv-based targeting moiety and a toxin. Pseudomonas exotoxin A (PE) has been used to make several immunotoxins that have been evaluated in clinical trials. Immunogenicity of the bacterial toxin and off-target toxicity have limited the efficacy of these immunotoxins. To address these issues, we have previously made RITs in which the Fv is connected to domain III (PE24) by a furin cleavage site (FCS), thereby removing unneeded sequences of domain II. However, the PE24 containing RITs do not contain the naturally occurring disulfide bond around the furin cleavage sequence, because it was removed when domain II was deleted. This could potentially allow PE24 containing immunotoxins to be cleaved and inactivated before internalization by cell surface furin or other proteases in the blood stream or tumor microenvironment. Here, we describe five new RITs in which a disulfide bond is engineered to protect the FCS. The most active of these, SS1-Fab-DS3-PE24, shows a longer serum half-life than an RIT without the disulfide bond and has the same anti-tumor activity, despite being less cytotoxic in vitro. These results have significance for the production of de-immunized, low toxicity, PE24-based immunotoxins with a longer serum half-life.