Search

Rectal cancer ranks as the second leading cause of cancer-related deaths. Neoadjuvant therapy for rectal cancer patients often results in individuals that respond well to therapy and those that respond poorly, requiring life-altering excision surgery. It is inadequately understood what dictates this responder/nonresponder divide. Our major aim is to identify what factors in the tumor microenvironment drive a fraction of rectal cancer patients to respond to radiotherapy. We also sought to distinguish potential biomarkers that would indicate a positive response to therapy and design combinatorial therapeutics to enhance radiotherapy efficacy. To address this, we developed an orthotopic murine model of rectal cancer treated with short course radiotherapy that recapitulates the bimodal response observed in the clinic. We utilized a robust combination of transcriptomics and protein analysis to identify differences between responding and nonresponding tumors. Our mouse model recapitulates human disease in which a fraction of tumors respond to radiotherapy (responders) while the majority are nonresponsive. We determined that responding tumors had increased damage-induced cell death, and a unique immune-activation signature associated with tumor-associated macrophages, cancer-associated fibroblasts, and CD8+ T cells. This signature was dependent on radiation-induced increases of Type I interferons (IFNs). We investigated a therapeutic approach targeting the cGAS/STING pathway and demonstrated improved response rate following radiotherapy. These results suggest that modulating the Type I IFN pathway has the potential to improve radiation therapy efficacy in RC.

The ability to alter the cytokine microenvironment has the potential to shape immune responses in many physiological settings, including the immunotherapy of tumours. We set out to develop a general approach in which cytokines could be functionally attenuated until activated. We report the development and initial characterization of fusion proteins in which human or mouse interleukin-2 (IL-2), a potent growth factor for immune cells, is joined to a specific IL-2 inhibitory binding component separated by a protease site. The rationale is that upon cleavage by a protease the cytokine is free to dissociate from the inhibitory component and becomes biologically more available. We describe the successful development of two attenuation strategies using specific binding: the first uses the mouse IL-2 receptor alpha chain as the inhibitory binding component whereas the second employs a human antibody fragment (scFv) reactive with human IL-2. We demonstrated that the fusion proteins containing a prostate-specific antigen or a matrix metalloproteinase (MMP) protease cleavage site are markedly attenuated in the intact fusion protein but had enhanced bioactivity of IL-2 in vitro when cleaved. Further, we showed that a fusion protein composed of the IL-2/IL-2 receptor alpha chain with an MMP cleavage site reduced tumour growth in vivo in a peritoneal mouse tumour model. This general strategy should be applicable to other proteases and immune modulators allowing site-specific activation of immunomodulators while reducing unwanted side-effects.

The omentum is an adipose tissue that contains milky spots (MS). The MS are similar to secondary lymphoid organs and can generate B and T cell immune responses to peritoneal antigens. Additionally, omentum CD4+CD25+FoxP3+ T regulatory cells (Tregs) display an activated phenotype CD44hiCD62Llow. Moreover, the omentum also collects metastasizing tumor cells in the peritoneal cavity and tumors growing in the omentum are associated with poor clinical prognosis. Here we show that intestinal microbiota affects Tregs activation profile and impairs tumor growth in omentum. Our transplantable tumor model in wild type (WT) mice shows that peritoneal tumors grow progressively in the omentum and peritoneal cavity in WT mice, whereas in germ free (GF) mice tumors do not grow. Omental tumor growth in WT mice is associated with an increase in PD-1+ Tregs, while tumor-specific CD8+ T are reduced. However, in GF mice numbers of PD1+ Tregs remain unaltered but tumor specific CD8+ T cells are still present. After co-housed GF mice with WT mice, ex-GF mice showed tumor growth and restore the increase in PD-1+ Tregs with low number of specific CD8+ T cells. The induction of tolerance requires PD-1+ Tregs. These data suggest that intestinal microbiota has a role in omental Tr activation and in the tolerance to peritoneal tumors.

Pro-inflammatory cytokines required for immunity against non-self, such as interferons (IFNs), are expressed at the maternal-fetal interface during normal pregnancy. However, neither the placental cell targets nor the functions of these cytokines have been defined. We hypothesized that semi-allogeneic trophoblast cells (TBCs), which are the only cells derived from the fertilized egg that are in direct contact with maternal tissue, are hypo-responsive to IFNs in order to evade maternal immune recognition. Therefore, we developed a novel whole mount immunofluorescence technique to identify cellular targets of IFN-γ and IFN-α/β by examining expression of phosphorylated STAT-1 (pSTAT-1) in intact human placental explants. Using this technique we were able to identify the major cell populations present in the placental explants, including fetal macrophages and distinct subpopulations of TBCs. Neither the syncytiotrophoblast layer in direct contact with maternal blood, nor the underlying cytotrophoblast cells expressed pSTAT-1 in response to either IFN. In contrast, extravillous trophoblast cells, which are in direct contact with maternal decidual tissue, and fetal macrophages induced strong expression of pSTAT-1 in response to IFNs. The identification of differential IFN responsiveness in distinct human trophoblast cell subpopulations provides important new insights into the roles of IFNs in normal pregnancy and placental pathology and infection.

The United States continues to be a prime target for attack by terrorist organizations in which nuclear detonation or dispersal of radiological material are legitimate threats. These types of attacks could result in devastating consequences to a multitude of individuals in the form of radiation injury to particular organ systems. One of these organs, the cutaneous system, which provides the first line of defense against invading pathogens, is particularly sensitive to ionizing radiation (IR) leaving the host vulnerable to cutaneous pathogen invasion. Our laboratory demonstrated that exposure to IR depleted murine cutaneous dendritic cells (cDCs); a network of cells responsible for forming a cutaneous immunological barrier that is critical in protecting the host against microorganism infection. This IR-induced depletion was caused by a migration of cDCs out of the skin responding to abnormal upregulation of cDC migratory factors, CCR7 and CCL21. As a result, this breakdown in the immunological barrier left the host susceptible to a cutaneous Candida albicans challenge where the pathogen was not contained locally but rather disseminated systemically to the kidneys. Interestingly, a subcutaneous treatment with IL-12 prevented the IR-induced depletion of cDCs offering a potential radiation countermeasure to preserve the cutaneous immunological barrier and perhaps prevent the lethal dissemination of opportunistic pathogens following radiation exposure.

Cytokines are being used for tumor immunotherapy, but when delivered systemically, negative side effects limit their utility. To mitigate these side effects but maintain their anti-tumor effects, we have constructed a panel of fusion proteins (FPs) with IL-2 joined to a specific inhibitor, separated by specific protease cleavage sequences. The rationale is that at the tumor site, where the concentration of a protease is high, the FP will be cleaved, allowing the IL-2 to dissociate from the inhibitor, increasing the biological activity of the cytokine and enhancing anti-tumor effectors. We have expressed these FPs and showed that cleavage in vitro results in increased biological activity of IL-2. In the Colon 38 peritoneal tumor model, delivery of a FP results in increased frequency of T cells (p< 0.0001) and NK cells (p< 0.01) compared to vehicle control treatment, and a corresponding decrease in tumor growth. To test for systemic side effects and evaluate the effects of FP delivery, we used an adeno-associated virus (AAV) delivery system to achieve sustained, high systemic levels of the FPs in mice. AAV-expressed FPs persist at high levels (~500 pg/ml) for at least 9 weeks and do not appear immunogenic. Importantly, unlike injection of free IL-2, expression of the FPs did not generate an inflammatory cytokine response as determined by a 22-analyte Luminex assay. Thus, protease activated cytokine fusion proteins may be a valuable means to enhance tumor immunotherapy.

Tumor cell metastasis to the peritoneal cavity (PC) is a serious concern for many cancers, including ovarian and colon tumors. Upon introduction to the PC, tumor cells bind to the omentum, a tissue comprised of adipose embedded with organized immune aggregates (IA) thought to have regulatory function. Interestingly, it is to these IA that tumor cells initially bind and thrive. This presents a unique situation in which tumor cells are in close proximity to potential effector cells. However, without a stimulus to override their regulatory function, IA fail to control tumor growth. Thus, we attempted to stimulate a response by immunizing mice with a lethally irradiated colon adenocarcinoma cell line: Colon38 (C38). Our data demonstrates that i.p. immunization with C38 is capable of preventing growth of C38 upon live tumor challenge. Although T cells are necessary to generate this response, when CD8 or CD4 T cells were depleted during the effector phase, tumor growth was still controlled; suggesting T cells are not the effector cells. Interestingly, in mice immunized with C38 and challenged with another tumor cell line, E0771, tumor growth was also prevented. This non-specific response is unique to the PC, as mice that were immunized with C38 and challenged with E0771 by an alternate (i.m.) route developed tumors. Together, these data suggest that i.p. immunization results in T cell induction of an innate/non-specific anti-tumor response.

A new paradigm has emerged demonstrating that the immune system mediates many of the anti-tumor effects of radiation therapy (RT). RT continues to be a cornerstone in the treatment for many cancers, unfortunately not all individuals respond effectively to RT resulting clinically into two groups consisting of non-responders (progressive disease) and partial/complete responders (tumor control/cure). We have designed a unique model that mirrors this clinically relevant phenomenon in which mice bearing colon38, a colon adenocarcinoma, were treated locally with 15Gy RT resulting in both non-responders and responders. Interestingly, our data uncovered a window of interest four days following RT in which critical events occurred that ultimately determined the effectiveness of therapy. Intratumoral IFNγ was increased in responsive tumors and licensed CD8 T cells to exhibit potent lytic activity against tumor cells; a response that was greatly diminished in tumors refractory to RT. We explored whether enhancing IFNγ and CD8 T cells via immunotherapy would increase the responder to non-responder ratio. Indeed, multi-modality therapy comprised of IL-12 and RT in mice bearing colon38 tumors resulted in complete remission of cancer in 100% of cases compared to a cure rate of only 40% with RT alone. Our data suggests that the efficacy of RT may depend on the immune response generated shortly after treatment and targeting this component may greatly enhance the effectiveness of RT.

Interferon alpha (IFNα) has been FDA-approved for treatment of high-risk melanoma since 1995. Besides having direct antitumor effects, IFNα and IFNβ, the other major type I IFN (IFN-I), have been found to act through enhancing multiple pathways within the host immune system. However, the mechanisms by which IFN-Is mediate these responses remain unclear. Our previous data demonstrated that treatment of in vivo B16 melanoma with a single high dose local radiation induced an increase in IFNγ levels within the tumor microenvironment. Interestingly, this increase is abolished in the absence of IFNα/β signaling within host cells, as observed using IFNα/β receptor knockout mice (KO). Importantly, the induction of downstream IFNγ-dependent genes such as the CXCR3 family of chemokines is also affected. Since CXCR3 chemokines have been shown to be crucial chemoattractants for T cells in tumors, it is not surprising that the lack of IFNα/β receptor also resulted in decreased recruitment of CD8 T cells to tumors in response to radiation treatment. Corresponding to these effects, radiation-treated KO mice exhibit poor tumor growth control compared to wild type mice. Taken together, our data suggest that IFN-Is play a critical role in radiation-mediated antitumor responses. An important mechanism through which this occurs depends on the ability of IFN-Is to enhance IFNγ responses, which in turn upregulates CXCR3 chemokines in the tumor, resulting in enhanced recruitment of CD8 T cells.

Radiation (IR) therapy has long been an effective cancer treatment acting primarily through the induction of tumor cell damage at a molecular level. However, a new concept is emerging suggesting that the immune system plays a pivotal role in determining the effectiveness of IR therapy. We have previously demonstrated that IR therapy promotes antitumor immunity through the release of tumor antigens and induction of a pro-inflammatory tumor microenvironment. Our work here has identified IFNγ as an essential cytokine which mediates the efficacy of IR therapy. Local IR (15Gy) to mice bearing colon38, a colon adenocarcinoma, resulted in decreased tumor burden in wild type animals. Interestingly, IR therapy had no effect on tumor burden in IFNγKO mice. We further determined that intratumoral (i.t.) levels of IFNγ spiked two days following IR which directly correlated with a decrease in tumor burden that was not a result of direct cytotoxic effects of IFNγ on tumor cells. CD8+ T cells and NK cells were the predominant producers of IFNγ, however both radiosensitive populations were depleted by IR; a paradox not yet completely understood. IR treated tumors could be further divided into responders and non-responders. Responders were shown to possess higher IFNγ levels, and i.t. treatment with IFNγ further enhanced the effectiveness of IR therapy. These data suggest that IFNγ may play a pivotal role in mediating the antitumor effects of IR therapy.

Tumor metastasis to the peritoneal cavity is often observed in patients with primary ovarian cancer as well as resected colon cancers. Metastatic cells frequently attach to the omentum, a peritoneal adipose tissue that contains many immune aggregates (IA). Interestingly, it is to these IA that the tumor cells initially bind and proliferate. Our lab has been investigating the capability of harnessing these IA to generate productive anti-tumor immune responses using the tumor cell line Colon38, a colon adenocarcinoma syngeneic to C57BL/6 mice. Immunization of mice with lethally irradiated Colon38 cells grown in serum-free media results in a marked increase in the number of CD8 T cells as well as their expression of activation markers. Challenge of the mice with live Colon38 14 days after immunization results in no observable tumor growth on the omentum and rechallenge experiments also yield no tumor growth. Unexpectedly, when we challenge mice with an alternate tumor cell line, B16.F0, we also observe no tumor growth on the omentum. These results have been repeated in the BALB/c strain of mice using syngeneic tumor cell lines. Our data not only highlight the uniqueness of the peritoneal cavity, but also suggest that this immunization strategy generates an initial non-specific anti-tumor response that likely involves components of the innate immune system as well as a long-term tumor-specific response involving CD8 T cells.

Cytokines have been used to treat tumors, however, side effects often limit their use when delivered systemically. We set out to develop an approach in which cytokines would be preferentially activated at the site of growing tumors. We have constructed a novel fusion protein (FP) consisting of the cytokine interleukin-2 (IL-2) joined to a specific inhibitor moiety, the interleukin-2 receptor alpha chain separated by a protease cleavage sequence recognized by matrix metalloproteinases (MMPs). The rationale is that the cytokine would be relatively inactive in the intact fusion protein but its biological activity would increase after protease cleavage. We analyzed a baculovirus expressed FP biochemically by immunoblot analyses and functionally using the IL-2 dependent CTLL-2 cell line. IL-2 activity is attenuated over 50 fold in the intact FP compared to the same molar amount of IL-2. MMP cleavage releases IL-2 resulting in an increase in the amount of biologically available IL-2. Additionally, the FP showed efficacy in vivo using a peritoneal tumor model employing Colon 38. This general approach may be applicable to other cytokines as well as other proteases.

Interferon-alpha (IFN-α) is clinically approved for treatment of melanoma. Besides having direct antitumor effects, type I IFNs (IFN-α/β) can also enhance IFN-γ responses in cells. Previously, we found that local treatment with a single high dose of radiation induced an increase of IFN-γ within the tumor microenvironment, and this increase is important for radiation-induced antitumor responses. Since the interplay between IFN-α/β and IFN-γ has been shown to be important in several immunological processes, we hypothesize that the lack of type I IFN signaling will result in poor radiation-induced IFN-γ responses and as a consequence, a lack of tumor control. To test our hypothesis, we first compared the growth of B16 tumors between wild type (WT) mice and IFN-α/β receptor knockout (KO) mice. Interestingly, tumor growth was faster in KO mice than WT mice, suggesting that type I IFNs play a role in antitumor immunity. In mice that were treated with 15Gy radiation, the tumor burden reduction observed in KO mice was comparatively smaller than that in WT mice. Corresponding to lower reduction in tumor growth after radiation treatment, KO mice have limited levels of radiation-induced intratumoral IFN-γ and number of tumor-infiltrating CD8+ T cells compared to that in radiation-treated WT mice. Taken together, our data suggest that type I IFN signaling within host cells is important for antitumor immunity, including IFN-γ-mediated responses induced by radiation treatment.

Tumor metastasis within the peritoneal cavity is a concern for those diagnosed with cancers of peritoneal organs. A tissue implicated in this process is the omentum, which is composed of adipose tissue embedded with immune cell aggregates. We, and others have demonstrated that upon introduction of tumor cells into the peritoneal cavity, omental immune aggregates are the initial site of attachment. Due to their high vascularity, they provide an ideal microenvironment for tumor growth. Thus, treatment of peritoneal metastases often includes removal of the omentum. However, we hypothesize that stimulation of omental immune cells might result in reduced tumor growth. Our initial experiments demonstrate that injection of particles into the peritoneal cavity resulted in their uptake by macrophages present in the peritoneal fluid and rapid migration to immune aggregates. To determine if tumor antigens would result in effective immunity when bound to these aggregates, mice were immunized intraperitoneally with irradiated EMT6 tumor cells and later challenged with live EMT6/GFP cells. Omenta and peritoneal fluid were analyzed for the presence of memory CD8+ T cells and tumor cells. Our results show that immunization results in increased memory CD8+ T cells and decreased tumor burden on the omentum, indicating that omental immune cells are capable of responding against tumor cells. Further studies are being done to examine tumor burden in mice immunized after tumor establishment.

A number of T cell functions were followed in mice after thymectomy in adult life. It was found that T cells involved in these functions could be divided roughly into two subpopulations: one whose activity was relatively long-lived (longer than 20 weeks) and one, relatively short-lived (on the order of a few weeks). In the long-lived class were those T cells involved in the helper function in the primary IgM and IgG responses to SRBC in vivo and in vitro. Priming for the secondary demonstration of helper activity or delayed hypersensitivity specific for SRBC seemed to involve both classes, although the contribution from the short-lived class was greater. These findings are consistent with models for T cell heterogeneity in which a short-lived population of T cells contains precursors distinct from a second long-lived population containing memory cells. Both classes can apparently contribute to a pool of activated effector cells in a process driven by antigenic stimulation.