Madeleine Birgersson

Human papillomavirus (HPV)‐associated tonsillar squamous cell carcinoma (TSCC) is suggested to arise in tonsillar crypts. It also constitutes a rare exception in the literature of carcinoma development, as carcinoma in situ (CIS) is not a recognized entity and dysplastic stages are not evident. Here we investigated the evidence of this. Hence, a systematic review and meta‐analysis was performed to study tumor origin. MEDLINE was searched and all original studies reporting tumor origin in HPV‐associated oropharyngeal cancer were included. We also used spatial transcriptomics (10× Visium) on HPV‐associated TSCC, cervical, and HPV‐independent oral cancer cases. We compared tonsillar CIS to cervical high‐grade intraepithelial lesion (HSIL). We studied epithelial cell gene expression across the sample types and epithelial cellular states. A trajectory analysis was performed in HPV‐associated tumors. In total, 14% of all TSCC were of surface origin in the systematic review. Spatial transcriptomics revealed that tonsillar CIS and HSIL clustered with a high correlation when using a pseudo‐bulk approach. Moreover, when all epithelial cells were analyzed unsupervised, ten epithelial cell clusters that correlated to histology were identified. Two of these clusters were dysplastic and occurred in all samples. A sequential progression from dysplasia to invasion was observed in both TSCC and cervical cancer in the trajectory analysis. We conclude that a subset of HPV‐associated TSCC may arise from the surface and that there is a sequential progression toward invasive disease in HPV‐associated TSCC, similar to that of cervical cancer. The findings suggest that pure tonsillar dysplastic lesions should hypothetically exist.

Inflammatory intestinal conditions are a major disease burden. Numerous factors shape the distribution of immune cells in the colon, but a spatial characterization of the homeostatic and inflamed colonic immune microenvironment is lacking. Here, we use the COMET platform for multiplex immunofluorescence to profile the infiltration of nine immune cell populations in mice of both sexes (N = 16) with full spatial context, including in regions of squamous metaplasia. Unsupervised clustering, neighborhood analysis, and manual quantification along the proximal-distal axis characterized the colonic immune landscape, quantified cell-cell interactions, and revealed sex differences. The distal colon was the most affected region during colitis, which was pronounced in males, who exhibited a sex-dependent increase of B cells and reduction of M2-like macrophages. Regions of squamous metaplasia exhibited strong infiltration of numerous immune cell populations, especially in males. Females exhibited more helper T cells and neutrophils at homeostasis and increased M2-like macrophage infiltration in the mid-colon upon colitis. Sex differences were corroborated by plasma cytokine profiles. Our results provide a foundation for future studies of inflammatory intestinal conditions.

A high-fat diet can lead to gut microbiota dysbiosis, chronic intestinal inflammation, and metabolic syndrome. Notably, resulting phenotypes, such as glucose and insulin levels, colonic crypt cell proliferation, and macrophage infiltration, exhibit sex differences, and females are less affected. This is, in part, attributed to sex hormones. To investigate if there are sex differences in the microbiota and if estrogenic ligands can attenuate high-fat diet-induced dysbiosis, we used whole-genome shotgun sequencing to characterize the impact of diet, sex, and estrogenic ligands on the microbial composition of the cecal content of mice. We here report clear host sex differences along with remarkably sex-dependent responses to high-fat diet. Females, specifically, exhibited increased abundance of Blautia hansenii, and its levels correlated negatively with insulin levels in both sexes. Estrogen treatment had a modest impact on the microbiota diversity but altered a few important species in males. This included Collinsella aerofaciens F, which we show correlated with colonic macrophage infiltration. In conclusion, male and female mice exhibit clear differences in their cecal microbial composition and in how diet and estrogens impact the composition. Further, specific microbial strains are significantly correlated with metabolic parameters.

There are significant sex differences in colorectal cancer (CRC), including in incidence, onset, and molecular characteristics. Further, while inflammatory bowel disease (IBD) is a risk factor for CRC in both sexes, men with IBD have a 60% higher risk of developing CRC compared to women. In this study, we investigated sex differences during colitis-associated CRC (CAC) using a chemically induced CAC mouse model. The mice were treated with azoxymethane (AOM) and dextran sodium sulfate (DSS) and followed for 9 and 15 weeks. We performed RNA-sequencing of colon samples from males (n = 15) and females (n = 15) to study different stages of inflammation and identify corresponding transcriptomic sex differences in non-tumor colon tissue. We found a significant transcriptome response to AOM/DSS treatment in both sexes, including in pathways related to inflammation and cell proliferation. Notably, we found a stronger response in males and that male-specific differentially expressed genes were involved in NFκB signaling and circadian rhythm. Further, an overrepresented proportion of male-specific gene regulations were predicted to be targets of Stat3, whereas for females, targets of the glucocorticoid receptor (Gr/Nr3c1) were overrepresented. At 15 weeks, the most apparent sex difference involved genes with functions in T cell proliferation, followed by the regulation of demethylases. The majority of sex differences were thus related to inflammation and the immune system. Our novel data, profiling the transcriptomic response to chemically induced colitis and CAC, indicate clear sex differences in CRC initiation and progression.

Epidemiological studies highlight a strong association between obesity and colorectal cancer (CRC), especially in men. Estrogen, on the other hand, is associated with protection against both the metabolic syndrome and CRC. The colon is the first organ to respond to a high-fat diet (HFD), and estrogen receptor beta (ERβ) in the intestine appears to prevent CRC. How estrogen impacts the colon under HFD condition has, however, not been investigated. Estrogen can act through three different receptors (ERα, ERβ, GPER1) which all may impact metabolism. In an effort to dissect this, we fed mice a control diet or a high-fat diet (HFD) for 13 weeks and administered receptor-selective estrogenic ligands for the last three weeks. We recorded corresponding physiological impact on fat distribution, fasting glucose, colon crypt proliferation and immune cell infiltration, and the colon transcriptome response. We identify clear sex-differences at the transcriptome level, both at base line and after HFD and ligand treatments. An unexpected observation was the significant sex-differences and impact by HFD and estrogens on circadian clock gene expression, such as Npas2 and Arntl (Bmal1) , in the colon. Both sexes also exhibited an increased infiltration of F4/80+ macrophages as a result of HFD. In males, but not females, this was accompanied by changes in colonic epithelial cell proliferation. ERα-selective PPT treatment had significant systemic effects, reducing body weight in both sexes, whereas ERβ-selective DPN treatment did not impact body weight, but reduced infiltration of F4/80+ macrophages in colon of both sexes and attenuated HFD-induced proliferation of male colon crypts. Both ERα and ERβ activation contributed to circadian clock gene regulations. We detail for the first time how HFD and estrogens modulate the colon transcriptome and physiology in a sex and ER-specific manner.