Spinal cord injury-induced neurogenic bowel: A role for host-microbiome interactions in bowel pain and dysfunction

Background and aims

Spinal cord injury (SCI) affects roughly 300,000 Americans with 17,000 new cases added annually. In addition to paralysis, 60% of people with SCI develop neurogenic bowel (NB), a syndrome characterized by slow colonic transit, constipation, and chronic abdominal pain. The knowledge gap surrounding NB mechanisms after SCI means that interventions are primarily symptom-focused and largely ineffective. The goal of the present studies was to identify mechanism(s) that initiate and maintain NB after SCI as a critical first step in the development of evidence-based, novel therapeutic treatment options.

Methods

Following spinal contusion injury at T9, we observed alterations in bowel structure and function reflecting key clinical features of NB. We then leveraged tissue-specific whole transcriptome analyses (RNAseq) and fecal 16S rRNA amplicon sequencing in combination with histological, molecular, and functional (Ca2+ imaging) approaches to identify potential mechanism(s) underlying the generation of the NB phenotype.

Results

In agreement with prior reports focused on SCI-induced changes in the skin, we observed a rapid and persistent increase in expression of calcitonin gene-related peptide (CGRP) expression in the colon. This is suggestive of a neurogenic inflammation-like process engaged by antidromic activity of below-level primary afferents following SCI. CGRP has been shown to disrupt colon homeostasis and negatively affect peristalsis and colon function. As predicted, contusion SCI resulted in increased colonic transit time, expansion of lymphatic nodules, colonic structural and genomic damage, and disruption of the inner, sterile intestinal mucus layer corresponding to increased CGRP expression in the colon. Gut microbiome colonization significantly shifted over 28 days leading to the increase in Anaeroplasma, a pathogenic, gram-negative microbe. Moreover, colon specific vagal afferents and enteric neurons were hyperresponsive after SCI to different agonists including fecal supernatants.

Conclusions

Our data suggest that SCI results in overexpression of colonic CGRP which could alter colon structure and function. Neurogenic inflammatory-like processes and gut microbiome dysbiosis can also sensitize vagal afferents, providing a mechanism for visceral pain despite the loss of normal sensation post-SCI. These data may shed light on novel therapeutic interventions targeting this process to prevent NB development in patients.

Adipose-Tumor Crosstalk contributes to CXCL5 Mediated Immune Evasion in PDAC

Background

CXCR1/2 inhibitors are being implemented with immunotherapies in PDAC clinical trials. Cytokines responsible for stimulating these receptors include CXCL ligands, typically secreted by activated immune cells, fibroblasts, and even adipocytes. Obesity has been linked to poor patient outcome and altered anti-tumor immunity. Adipose-derived cytokines and chemokines have been implicated as potential drivers of tumor cell immune evasion, suggesting a possibility of susceptibility to targeting specifically in the context of obesity.

Methods

RNA-sequencing of human PDAC cell lines was used to assess differential influences on the cancer cell transcriptome after treatment with conditioned media from peri-pancreatic adipose tissue of lean and obese PDAC patients. The adipose-induced secretome of PDAC cells was then assessed by cytokine arrays and ELISAs. Lentiviral transduction and CRISPR-Cas9 was used to knock out CXCL5 from a murine PDAC cell line for orthotopic tumor studies in diet-induced obese, syngeneic mice. Flow cytometry was used to define the immune profiles of tumors. Anti-PD-1 immune checkpoint blockade therapy was administered to alleviate T cell exhaustion and invoke an immune response, while the mice were monitored at endpoint for differences in tumor size.

Results

The chemokine CXCL5 was secreted in response to stimulation of PDAC cells with human adipose conditioned media (hAT-CM). PDAC CXCL5 secretion was induced by either IL-1β or TNF, but neutralization of both was required to limit secretion. Ablation of CXCL5 from tumors promoted an immune phenotype susceptible to PD-1 inhibitor therapy. While application of anti-PD-1 treatment to control tumors failed to alter tumor growth, knockout CXCL5 tumors were diminished.

Conclusions

In summary, our findings show that known adipokines TNF and IL-1β can stimulate CXCL5 release from PDAC cells in vitro. In vivo , CXCL5 depletion alone is sufficient to promote T cell infiltration into tumors in an obese setting, but requires checkpoint blockade inhibition to alleviate tumor burden.

DATA AVAILABILITY STATEMENT

Raw and processed RNAseq data will be further described in the GEO accession database ( awaiting approval from GEO for PRJ number ). Additional raw data is included in the supplemental material and available upon reasonable request.

WHAT IS ALREADY KNOWN ON THIS TOPIC

Obesity is linked to a worsened patient outcome and immunogenic tumor profile in PDAC. CXCR1/2 inhibitors have begun to be implemented in combination with immune checkpoint blockade therapies to promote T cell infiltration under the premise of targeting the myeloid rich TME.

WHAT THIS STUDY ADDS

Using in vitro/ex vivo cell and tissue culture-based assays with in vivo mouse models we have identified that adipose derived IL-1β and TNF can promote tumor secretion of CXCL5 which acts as a critical deterrent to CD8 T cell tumor infiltration, but loss of CXCL5 also leads to a more immune suppressive myeloid profile.

HOW THIS STUDY MIGHT AFFECT RESEARCH PRACTICE OR POLICY

This study highlights a mechanism and emphasizes the efficacy of single CXCR1/2 ligand targeting that could be beneficial to overcoming tumor immune-evasion even in the obese PDAC patient population.

Abstract C081: Opposing effects of white and brown adipose tissue on PDAC growth

The 5-year overall survival for pancreatic ductal adenocarcinoma (PDAC) patients is an abysmal 12% with local disease and 3% for metastatic disease. Obesity is a major risk factor for PDAC showing a three-month decrease in overall survival when compared to lean PDAC patients. The causal relationship between obesity and cancer has driven seminal findings to support a pro-tumorigenic role of white adipose tissue (WAT). Our previous results agreed with wider based studies demonstrating obese WAT as highly inflamed and eliciting pro-tumorigenic effects, through secretion of adipose derived factors that increase tumor proliferation, migration, and xenograft size. While multiple studies have implicated obese WAT as eliciting pro-tumorigenic effects, the effects of brown adipose tissue (BAT) on PDAC have not been explored. Unlike WAT, the BAT is generally protective against obesity, regulates core body temperature, and supports an anti-inflammatory environment. While regulation of these metabolic effects by BAT are well established, the influence of BAT has on PDAC function remains to be determined. Considering its contradictory role to WAT, we hypothesized that BAT would exhibit anti-tumorigenic effects on pancreatic cancer. To test this hypothesis, we isolated and cultured adipose tissue from lean and obese mice in order to collect perigonadal WAT (pgWAT) and BAT conditioned media (CM). PDAC cells were cultured in the absence or presence of pgWAT-CM or BAT-CM and then assayed for alterations in proliferation. While pgWAT induced proliferation of PDAC cells, we detected a significant inhibition of proliferation by BAT-CM. In order to adapt to environmental demands, WAT can convert into BAT and vice versa. Taking account of this phenomenon, we hypothesized that induction of WAT browning would reduce tumor growth by limiting the release of energy-rich molecules from obese WAT and mimicking BAT. To test our hypothesis, 3T3-L1 pre-adipocytes were differentiated into mature white adipocytes and subsequently “browned” by utilizing Cl-316, 243, trametinib, or roscovitine. From these treatments, we validated that brown-like gene signatures were upregulated while white-like gene signatures were downregulated after the treatments. The adipose conditioned media (ACM) from these “browned” adipocytes was applied to a PDAC cell line and proliferation was measured. Our results demonstrated that the browned ACM inhibited PDAC proliferation when compared to non-treated white ACM. These results support the idea that efficient browning of adipose tissue could be therapeutically beneficial for obese PDAC patients. Currently, we are analyzing the ACM from brown adipose tissue and browned WAT to determine alterations in secreted cytokines. Identifying the adipose derived cytokines involved in promoting PDAC proliferation will potentially provide alternate targets for therapeutics. In conclusion, we have identified an anti-tumorigenic role for BAT that further suggests the predominant subtype of adipose is important in relation to PDAC progression.

Citation Format: Austin Eades, Michael VanSaun, McKinnon R. Walsh, Bailey Bye, Appolinaire Olou, Joe Ambrose, Jarrid Jack. Opposing effects of white and brown adipose tissue on PDAC growth [abstract]. In: Proceedings of the AACR Special Conference on Pancreatic Cancer; 2022 Sep 13-16; Boston, MA. Philadelphia (PA): AACR; Cancer Res 2022;82(22 Suppl):Abstract nr C081.

Abstract 4027: Dual MEK and AKT inhibition suppresses pancreatic cancer growth and migration

Pancreatic cancer is one of the deadliest cancers, due to late diagnosis and very few available therapeutic treatments. The Kras gene is frequently mutated in pancreatic cancer, but previous clinical trials using RAS inhibitors have proven ineffective. Additionally, many drug treatments targeting the MAPK pathway have shown little success due to the inherent drug resistance of pancreatic cancer cells and to acquired resistance through activation of alternative proliferative pathways such as JAK-STAT and PI3K-AKT. Our pathway analysis after MEK inhibition in pancreatic cancer cells displayed sustained and/or upregulation of AKT activity in multiple cell lines. Therefore, we hypothesize that a combination treatment of Trametinib and Omipalisib would block two prominent mitogenic pathways (MEK and AKT, respectively) to suppress cancer cell proliferation and migration. Multiple analyses including Western blot, clonogenic, 5-Ethynyl-2’-Deoxyuridine (EDU), and scratch migration assays were employed to determine the augmentation of cellular function in murine and anthropic pancreatic cancer cell lines after single or combination treatment with Trametinib and/or Omipalisib. Results demonstrated that application of Omipalisib alone was successful at blocking pAKT but failed to suppress pERK. Conversely, Trametinib alone selectively inhibited pERK but did not affect pAKT levels. The combination treatment successfully suppressed both pathways in relatively low quantities, indicating their efficacy as a dual therapeutic. Dual therapy was further effective at inhibiting cell growth as evidenced by clonogenic and EdU assays. Recovery and closure from an in vitro scratch wound was also significantly inhibited with combination of Omipalisib and Trametinib. In vivo studies demonstrated that both the pancreatic xenograft mean tumor growth and final tumor size were significantly reduced in response to the combination treatment compared to vehicle. In conclusion, dual therapy with Omipalisib and Trametinib showed a greater anti-tumor efficacy than treatment with either drug alone. Currently, we are determining the prolonged effectiveness of this combination treatment program in genetically-engineered mouse models. If successful, a pharmacological application of Omipalisib and Trametinib has the potential to provide a beneficial therapeutic option to pancreatic cancer patients.

Citation Format: Jarrid Jack, Alexandra Pierce, Bailey Bye, McKinnon Walsh, Prabhakar Chalise, Michael N. VanSaun. Dual MEK and AKT inhibition suppresses pancreatic cancer growth and migration [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 4027.

Characterization of the Normal Mouse IGH Antibody Repertoire

While the process of antibody generation in response to pathogens and immunization is well understood, the normal antibody repertoire needs characterization to better understand the diversity of the antibody response. The antibody molecule is generated from multiple gene segments; variable (V), diversity (D), and joining (J). These segments rearrange (V, D, and J in the heavy chain, V and J in the light chain) and combine with a constant region, which determines the effector function within in the host. The V, D, and J gene segments are combined to form the complementary determining region 3 (CDR3), providing much of the variability within the B cell receptor repertoire. To address heavy chain rearrangement in the unstimulated antibody repertoire, we pooled and sequenced the total splenic RNA from three groups of normal C57BL/6J. V, D, J, and constant region usage and CDR3 formation were analyzed. We developed a bioinformatic workflow to analyze the most common usage and combinations of the V, D, and J gene segments and their pairings with constant regions. We examined length and composition of CDR3 along with the ontology of the observed antibodies. The preliminary screen identified over 33,000 sequences as possible antibodies; 5,743 were hypothesized to be productive sequences. The largest V-gene family represented was V1, followed by V2 and V6. The average CDR3 length was 11 amino acids and we identified a number of high-occurrence CDR3 amino acid sequences. These data provide a glimpse of the normal C57BL/6J heavy chain response and validate the bioinformatics needed for our future immunization studies. Supported by NASA grants NNX13AN34G and NNX15AB45G and NIH grant GM103418 and the Molecular Biology Core supported by CVM-KSU.

Medicare utilization by disabled-worker beneficiaries: a longitudinal analysis

Medicare eligibility for Social Security disabled-worker beneficiaries begins after 2 years of cash benefit receipt. Extension of the current coverage is often proposed as a way to encourage beneficiaries to return to work. Little is known, however, about the long-run Medicare costs for the disabled and how costs vary by demographic and health characteristics. This article describes Medicare utilization and reimbursement amounts for 1974-81 for a cohort of disabled-worker beneficiaries under age 62 and first entitled to cash benefits in 1972. The data come from a first-time linkage of Disability Insurance program data with data on Medicare utilization. The tables provide a detailed look at several factors that are associated with variation in Medicare costs among beneficiaries and over time.