Endometrial epithelium is the presumed tissue of origin for both eutopic and endometriosis-derived clear cell and endometrioid carcinomas. We had previously hypothesized that the morphological, biological and clinical differences between these carcinomas are due to histotype-specific mutations. Although some mutations and genomic landscape features are more likely to be found in one of these histotypes, we were not able to identify a single class of mutations that was exclusively present in one histotype and not the other. This lack of genomic differences led us to an alternative hypothesis that these cancers could arise from distinct cells of origin within endometrial tissue, and that it is the cellular context that accounts for their differences. In a proteomic screen, we identified cystathionine γ-lyase (CTH) as a marker for clear cell carcinoma, as it is expressed at high levels in clear cell carcinomas of the ovary and endometrium. In the current study, we analysed normal Müllerian tissues, and found that CTH is expressed in ciliated cells of endometrium (both eutopic endometrium and endometriosis) and fallopian tubes. We then demonstrated that other ciliated cell markers are expressed in clear cell carcinomas, whereas endometrial secretory cell markers are expressed in endometrioid carcinomas. The same differential staining of secretory and ciliated cells was demonstrable in a three-dimensional organoid culture system, in which stem cells were stimulated to differentiate into an admixture of secretory and ciliated cells. These data suggest that endometrioid carcinomas are derived from cells of the secretory cell lineage, whereas clear cell carcinomas are derived from, or have similarities to, cells of the ciliated cell lineage. Copyright © 2017 Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd.

Conflicting data has emerged regarding a role for eosinophils in IgA production, with some reports that eosinophils support both secretory and circulating IgA levels during homeostasis. Previous studies have compared antibody levels between wildtype and eosinophil-deficient mice, but these mice were obtained from different commercial vendors and/or were not littermates. Thus, the possibility remains that extrinsic environmental factors, rather than an intrinsic lack of eosinophils, are responsible for the reports of reduced IgA in eosinophil-deficient mice. Here we used wild-type and eosinophil-deficient (dblGATA) mice that were purchased from a single vendor, subsequently bred in-house and either co-housed as adults, co-reared from birth or raised as littermates. We found no differences in the levels of secretory IgA or in the numbers of small intestinal IgA-producing plasma cells between wild-type and dblGATA mice, demonstrating that under controlled steady-state conditions eosinophils are not essential for the maintenance of secretory IgA in the intestinal tract. While we found that levels of IgM and IgE were significantly elevated in the serum of dblGATA mice compared to co-reared or co-housed wild-type mice, no significant differences in these or other circulating antibody isotypes were identified between genotypes in littermate-controlled experiments. Our results demonstrate that eosinophils are not required to maintain secretory or circulating IgA production and the absence of eosinophils does not impact circulating IgG1, IgG2b, IgM, or IgE levels during homeostasis. These findings emphasize the importance of optimally controlling rearing and housing conditions throughout life between mice of different genotypes.

Intestinal helminth infection can impair host resistance to co-infection with enteric bacterial pathogens. However, it is not known whether helminth drug-clearance can restore host resistance to bacterial infection. Using a mouse helminth-Salmonella co-infection system, we show that anthelmintic treatment prior to Salmonella challenge is sufficient to restore host resistance to Salmonella. The presence of the small intestine-dwelling helminth Heligmosomoides polygyrus at the point of Salmonella infection supports the initial establishment of Salmonella in the small intestinal lumen. Interestingly, if helminth drug-clearance is delayed until Salmonella has already established in the small intestinal lumen, anthelmintic treatment does not result in complete clearance of Salmonella. This suggests that while the presence of helminths supports initial Salmonella colonization, helminths are dispensable for Salmonella persistence in the host small intestine. These data contribute to the mechanistic understanding of how an ongoing or prior helminth infection can affect pathogenic bacterial colonization and persistence in the mammalian intestine.

Heligmosomoides polygyrus is a helminth which naturally infects mice and is widely used as a laboratory model of chronic small intestinal helminth infection. While it is known that infection with H. polygyrus alters the composition of the host’s bacterial microbiota, the functional implications of this alteration are unclear. We investigated the impact of H. polygyrus infection on short-chain fatty acid (SCFA) levels in the mouse intestine and sera. We found that helminth infection resulted in significantly upregulated levels of the branched SCFA isovaleric acid, exclusively in the proximal small intestine, which is the site of H. polygyrus colonization. We next set out to test the hypothesis that elevating local levels of isovaleric acid was a strategy used by H. polygyrus to promote its own fitness within the mammalian host. To test this, we supplemented the drinking water of mice with isovalerate during H. polygyrus infection and examined whether this affected helminth fecundity or chronicity. We did not find that isovaleric acid supplementation affected helminth chronicity, however, we found that it did promote helminth fecundity, as measured by helminth egg output in the feces of mice. Through antibiotic-treatment of helminth-infected mice, we found that the bacterial microbiota was required in order to support elevated levels of isovaleric acid in the proximal small intestine during helminth infection. Overall, our data reveal that during H. polygyrus infection there is a microbiota-dependent localized increase in the production of isovaleric acid in the proximal small intestine and this supports helminth fecundity in the murine host.

Background Chronic infections with intestinal helminths occur primarily in world regions where infections with bacterial pathogens are also common. Research so far has shown that helminths can promote bacterial infection, however, the precise mechanisms of this are unknown. Current helminth control strategies involve mass deworming programs, and it is as yet unclear how deworming impacts susceptibility to bacterial infections in helminth-affected areas. Aims Our aims were to decipher the mechanistic details by which helminths can promote intestinal bacterial infection, and to determine how deworming affects susceptibility to intestinal colonization by bacterial pathogens. Methods Using a mouse model of enteric helminth-bacterial coinfection, we examined how the small intestinal helminth Heligmosomoides polygyrus promotes intestinal colonization by Salmonella enterica serovar Typhimurim. To study the effect of deworming of concurrent Salmonella infection, we treated our helminth-infected mice with the anthelmintic drug Strongid P both before and after Salmonella (co-) infection. To examine whether heightened Salmonella colonization during helminth infection depended on the ability of Salmonella to invade host tissue, we compared the ability of wild-type Salmonella or an invasion-deficient mutant (ΔinvA) to colonize during helminth co-infection. Results An ongoing helminth infection resulted in high levels of Salmonella in the small intestine after co-infection, however, when mice were dewormed prior to Salmonella co-infection, they were no longer susceptible to high small intestinal Salmonella burdens. In contrast, when helminth-infected mice are dewormed after Salmonella has already co-colonized, high Salmonella burdens persisted in the small intestine. Further, we found that during helminth co-infection, Salmonella primarily expands in the gut lumen rather than in the small intestinal tissue, and in line with this, a Salmonella invasion mutant was able to colonize the intestine to a similar extent to wild-type Salmonella during helminth infection. Conclusions Deworming experiments have revealed that the effects of H. polygyrus on promoting Salmonella colonization in the small intestines depend on the ongoing presence of the helminth. Deworming did not revert bacterial burdens once Salmonella had colonized, suggesting that an ongoing helminth infection reduces colonization resistance to Salmonella infection but was not required for Salmonella to persist after initial colonization by Salmonella. Further, we discovered that Salmonella expansion during helminth infection is independent of bacterial invasion of host tissue. Funding Agencies CIHR

Both clear cell ovarian carcinoma (CCOC) and endometrioid ovarian carcinoma (ENOC) are associated with ovarian endometriotic cysts, which is believed to be their precursor lesion. However, genomic evidence is lacking which could explain how these two clinically distinct histotypes of ovarian cancer arise from the same precursor lesion. We therefore hypothesized that these cancers arise from distinct cells of origin within endometrial tissue. Global proteomic analysis of ovarian cancer histotypes identified CTH as a marker for CCOC. We further found that CTH is highly expressed in the ciliated cells of endometrium (both ectopic endometrium and endometriosis), and of the fallopian tube, with little expression in the secretory cells. We also find that other ciliated cell markers are expressed in CCOC, whereas endometrial secretory cell markers are expressed in ENOC. We propose a new model of CCOC and ENOC histogenesis wherein ENOC is derived from cells of secretory cell lineage whereas CCOC is derived from cells of ciliated cell lineage. However, it remains unclear how external factors in the endometriotic cyst cooperate with cell of origin and mutation to promote cancer formation. To study normal tissue biology, we are using organoid cultures of normal endometrium. As ciliated cells of the endometrium are rare, and we have a particular interest in determining whether they have other features that may link them to CCOC, we used a Notch inhibitor, DBZ, to force ciliated cell differentiation in the organoids. We observed a dramatic shift in the cellular content with DBZ treamtment towards ciliated cells. We performed single cell RNA sequencing (scRNAseq) on these endometrial organoids. In the normal endometrial organoids, cells were predominantly a secretory phenotype, characterized by high ESR1 expression, with a minor ciliated cell population. The ciliated cell population expressed several known ciliated markers (FOXJ1 and DNAH12). Upon treatment with DBZ, the number of secretory cells decreases dramatically and two populations of cells emerge which have ciliated cell markers. The larger ciliated cell population is similar to the ciliated cells in the untreated organoids. The smaller ciliated cell population in the DBZ treated organoids express some ciliated cell markers, but clusters separately from normal ciliated cells. We believe this population may reprepsent an intermediary population, which has not fully differentiated. Interestingly, this population expresses the cytokine IL6, while the normal ciliated cell population does not. This is of note because CCOCs express more IL6 compared to the other histotypes. Therefore, we can speculate that this intermediary ciliated cell population may represent cells from which CCOC arise, however more testing is needed. In the future, the scRNAseq data from organoids will be compared to CCOC and ENOC tumors to determine whether the tumors resemble more closely one population of normal cells. We will use viral transduction to introduce mutations into the organoid cultures to determine whether specific mutation leads to transformation towards a CCOC or ENOC-like phenotype. These studies will enable us to tease apart the relative contribution of mutation, microenvironment and the cell of origin to promote tumor formation. Citation Format: Dawn R Cochrane, Basile Tessier-Cloutier, Germain Ho, Kieran Campbell, Evan Gibbard, Katherine M Lawrence, Tayyebeh Nazeran, Anthony N. Karnezis, Clara Salamanca, Angela S Cheng, Jessica N McAlpine, Sohrab Shah, Lien N Hoang, C Blake Gilks and David G Huntsman. CELL OF ORIGIN, MUTATION AND MICROENVIRONMENT: MODELING EARLY EVENTS OF ENDOMETRIOSIS ASSOCIATED CANCERS [abstract]. In: Proceedings of the 12th Biennial Ovarian Cancer Research Symposium; Sep 13-15, 2018; Seattle, WA. Philadelphia (PA): AACR; Clin Cancer Res 2019;25(22 Suppl):Abstract nr GMM-020.

Both clear cell ovarian carcinoma (CCOC) and endometrioid ovarian carcinoma (ENOC) are associated with ovarian endometriotic cysts, which is believed to be the precursor lesion of these cancers. Women with endometriotic cysts have up to a 3-fold increased risk of developing CCOC and ENOC. It is perplexing that these two clinically distinct histotypes of ovarian cancer arise from the same precursor lesion. We have performed whole-genome sequencing of ovarian cancer histotypes and found that while some genomic features are more common to one histotype than the other, there is not a single feature that is unique to either histotype. Lacking genomic evidence that could explain the differences between these histotypes, we hypothesized that these cancers arise from distinct cells of origin within endometrial tissue, and the cellular context accounts for their differences. We performed global proteomic analysis of ovarian cancer histotypes and identified CTH as a marker for CCOC. Upon examination of normal Müllerian tissues, we found that CTH is highly expressed in the ciliated cells of endometrium (both ectopic endometrium and endometriosis) and of the fallopian tube, with very little expression in the secretory cells of these tissues. We also find that other ciliated cell markers are expressed in CCOC, whereas endometrial secretory cell markers are expressed in ENOC. We propose a new model of CCOC and ENOC histogenesis wherein ENOC is derived from cells of secretory cell lineage whereas CCOC is derived from, or shares similarities to, cells of ciliated cell lineage. There remain, however, many unanswered questions. For example, while CCOC and ENOC occur at roughly equal prevalence, ciliated cells of the endometrium are rare compared to secretory cells. Cells in the endometriotic cyst are exposed to factors such as inflammation and reactive oxygen species, which could influence differentiation of endometrial progenitor cells into the secretory or ciliated cell lineage. To test factors that promote ciliated cell differentiation in normal endometrium, we treated organoid cultures of normal endometrium with IL-6 and Notch pathway modulators. We propose that ovarian cancer histotypes arise from different cells of origin and that the biology of the normal cells will be partly responsible for determining the phenotype of the cancers. Citation Format: Dawn Cochrane, Basile Tessier-Cloutier, Katherine Lawrence, Tayyebeh Nazeran, Anthony Karnezis, Clara Salamanca, Timothy Lee, Angela Cheng, Jessica McAlpine, Lien Hoang, Blake Gilks, David Huntsman. The origins of endometriosis-associated cancers. [abstract]. In: Proceedings of the AACR Conference: Addressing Critical Questions in Ovarian Cancer Research and Treatment; Oct 1-4, 2017; Pittsburgh, PA. Philadelphia (PA): AACR; Clin Cancer Res 2018;24(15_Suppl):Abstract nr B39.

Background During homeostasis eosinophils are highly abundant in the lamina propria of the small intestine. Eosinophils have been reported to play a role in promoting barrier function in the steady state intestinal tract, and in the control of intestinal colonization by harmless or symbiotic members of the bacterial microbiota. However, the role eosinophils play during enteric infection with a bacterial pathogen is unknown. Aims Our study aims to confirm the previously reported role of eosinophils in supporting intestinal barrier function and to investigate how the absence of eosinophils affects intestinal colonization by an enteric bacterial pathogen. Methods We used wildtype BALB/c and eosinophil-deficient (dblGATA knockout) BALB/c mice that had been cohoused, or bred as littermate controls, to normalize bacterial microbiota populations between mice used for experiments. To investigate steady state barrier function in these mice we used naïve mice to quantify levels of small intestinal IL-1α and IL-1β by cytometric bead arrays, and we used ELISAs to measure levels of immunoglobulin A (sIgA) and serum IgA. Levels of lamina propria-resident B220-IgA+ plasma cells were quantified using flow cytometry. To investigate the contributions of eosinophils to enteric bacterial infection, mice were orally infected with Salmonella enterica serovar Typhimurium. Salmonella burdens along the intestinal tract as well as in the liver and spleen were quantified. Results We found that levels of IL-1α and IL-1β were significantly decreased in the small intestine of naive eosinophil-deficient mice, compared to wildtype mice. In naïve wildtype and eosinophil-deficient littermate control mice, we did not detect any differences in sIgA or serum IgA levels. Additionally, levels of IgA producing plasma cells were similar in the small intestinal lamina propria between wildtype and eosinophil-deficient mice. Following oral Salmonella infection, Salmonella burdens were similar between wildtype and eosinophil-deficient mice both 24 hours and three days post-infection. Conclusions Our data supports a role for eosinophils in modifying steady-state cytokine levels in the intestinal tract. For the first time we report data on IgA levels from littermate controls of wildtype and eosinophil-deficient mice, and contrary to previously published reports, we found that eosinophils are not critical for the maintenance of intestinal sIgA, serum IgA or lamina propria resident IgA producing plasma cells. Our data further concluded that an absence of eosinophils did not impact control of Salmonella infection. Funding Agencies CIHR