Society: AGA

Paneth cells reside at the bottom of the crypts in the small intestine and are crucial for maintaining homeostasis of the epithelium by engendering host protection from enteric pathogens. Defects in Paneth cells compromise the intestinal epithelial host defense and lead to a range of mucosal pathologies. MicroRNA-195 (miR-195) is evolutionally conserved among different species and it regulates the stability and translation of target mRNAs and is involved in many aspects of cell processes. MiR-195 can interact with RNA-binding proteins including HuR to jointly regulate target gene expression synergistically or antagonistically. We recently reported that intestinal epithelial tissue-specific transgenic expression of miR-195 resulted in Paneth cell dysfunction in mice, but the exact mechanism underlying miR-195 in the regulation of Paneth cell function remains unknown. Sox9 is a specific transcription factor essential for Paneth cell differentiation and its deletion stops Paneth cell development. Here we tested the hypothesis that miR-195 regulates Paneth cells by altering Sox9 expression via interaction with HuR. Methods: Studies were conducted in our newly generated miR-195 transgenic (miR195-Tg) mice, HuR knockout (IE-HuR-/-) mice, and cultured HEK cells. Paneth cells were examined by lysozyme-immunostaining assays. The functions of miR-195 and HuR were determined by their gene silencing or overexpression in vitro. Levels of Sox9 mRNA and protein were measured by qPCR and Western blotting analyses. The interactions of Sox9 mRNA with HuR or miR-195 were examined by RNP or biotin immunoprecipitation (IP)/qPCR analysis. Results: miR195-Tg mice exhibited reduced levels of Sox9 protein (by ~70%) in the small intestinal mucosa, which was associated with a significant decrease in the number of Paneth cells. Ectopically overexpressed miR-195 in HEK cells also decreased cellular abundance of Sox9 primarily by inhibiting Sox9 translation since miR-195 overexpression decreased the levels of newly synthesized Sox9 protein without effect on its mRNA content. In contrast, HuR overexpression increased Sox9 expression in HEK cells. IP/qPCR analysis revealed that miR-195 did not directly bind to the Sox9 mRNA but prevented HuR association with Sox9 transcript. Ectopically expressed HuR rescued Sox9 expression in cells overexpressing miR-195. Double mutant mice bearing a miR-195 transgene and HuR ablation were generated by crossbreeding miR195-Tg mice and IE-HuR-/- mice and displayed further decreases in the levels of Sox9 and Paneth cells in the small intestine. Conclusions: These results indicate that 1) miR-195 inhibits Sox9 translation through interaction with HuR; and 2) control of Sox9 expression by miR-195 and HuR plays an important role in Paneth cell development and function.

Diarrhea is an important global health problem and secretory diarrhea outbreaks, such as cholera, remain a major cause of mortality in certain regions. Current treatment for secretory diarrhea is primarily supportive, and fluid replacement with oral rehydration solution (ORS) is the mainstay of cholera treatment. In addition to ORS, the WHO currently recommends oral zinc use for 10-14 days in children with diarrhea. Zinc has antidiarrheal efficacy in patients with and without zinc deficiency suggesting that it has direct antidiarrheal effects; however its exact mechanism of antidiarrheal action is not well understood. Extracellular calcium-sensing receptor (CaSR) is a regulator of intestinal ion transport and a therapeutic target for secretory diarrhea. We recently showed that CaSR activator cinacalcet has marked antidiarrheal effects in human intestinal epithelial cells and mouse intestine. Cinacalcet effect was via stimulating cAMP hydrolysis that leads to inhibition of cAMP-gated ion channels (CFTR and basolateral membrane K⁺ channels). CaSR is physiologically activated by divalent and trivalent cations, and zinc (Zn²⁺) is a divalent heavy metal. We hypothesized that CaSR activation is a key mechanism for antidiarrheal effects of zinc. In human intestinal T84 cells natively expressing CaSR, zinc concentration-dependently inhibited forskolin-induced secretory I_sc by up to 60%, which was comparable to CaSR activator cinacalcet effect. Similarly, 100 μM zinc inhibited I_sc induced by clinically relevant cAMP agonists cholera toxin and vasoactive intestinal peptide (VIP) by 65%. Interestingly, 100 μM zinc had no effect on cGMP-agonist heat-stable E. coli enterotoxin-induced secretory I_sc in T84 cells, suggesting its antisecretory effects are specific to cAMP agonists. Using selective membrane permeabilization and ion gradients in T84 cells, we found that zinc inhibits apical membrane CFTR Cl^- conductance and basolateral membrane K⁺ channel conductance by ~65%. Zinc had no effect on forskolin-induced secretory I_sc in CFTR-transfected Fischer Rat Thyroid cells that do not express CaSR. Zinc treatment resulted in intracellular Ca²⁺ elevation (secondary messenger for CaSR) in CaSR-transfected HEK293 cells, but not in wild type HEK293 cells, suggesting that zinc is a CaSR agonist. In mouse intestine, zinc (100 μM) inhibited forskolin-induced secretory I_sc by 40% in wildtype mice, with no effect in intestinal epithelia-specific CaSR knockout mice (Vil1-Cre; Casr-flox). Our findings suggest that CaSR activation is a major mechanism for the antidiarrheal effects of zinc. In addition to its current clinical use in cholera, zinc can also be effective in other cAMP-mediated diarrheas such as VIP-secreting tumors and bile acid diarrhea.

As the global obesity epidemic continues to worsen, identification of therapeutic targets has become increasingly important. One possible target is inducible heat shock protein 70 (HSP70), a conserved protein essential for responding to cellular stress. Unpublished work from the labs of Dr. Ciancio and Dr. Martinez-Guryn has shown that overexpression of intestine-specific HSP70 protected mice from diet induced obesity (DIO), but the mechanism is unknown. We hypothesized that intestinal HSP70 overexpression protects against DIO by modulating intestinal fat metabolism. To examine this, HSP70 transgenic (TG) mice containing a villin-driven HSP70 transgene and non-transgenic littermates (NTG) were fed either low-fat diet (LFD; 10% kcal from fat) or high-fat diet (HFD; 60% kcal from fat) for 12 weeks and fasting plasma lipid levels, stool caloric content, RNA sequencing of jejunum mucosa and targeted qPCR of genes involved in fat metabolism and absorption were performed and data analyzed using a two-way ANOVA followed by Tukey’s post-hoc test (n=2-12/group). Small intestinal length and weight were also measured in standard chow fed TG vs NTG mice and analyzed using a student’s t-test (n=7-13). Plasma from HSP70 TG mice fed HFD displayed a 47% or 36% decrease in plasma low-density lipoprotein (p=0.0002) and total cholesterol levels (p<0.0001), respectively, compared to NTG mice fed HFD. Plasma triglyceride levels were also decreased by 15% in HFD-TG mice compared to HFD-NTG mice (F=10.86, p=0.0061, genotype effect). Caloric content of stool measured via bomb calorimetry was higher in HFD-fed HSP70 TG mice (1.10±0.24 kcal) compared to HFD-NTG mice (0.63±0.28 kcal, p=0.0039). Interestingly, TG mice had longer (37.7±2.9cm vs 35.0±1.8cm, p=0.0107) and heavier (0.887±0.057g vs 0.513±0.035g, p<0.0001) small intestines compared to NTG mice. To understand the mechanisms involved, RNA sequencing was performed on jejunum mucosal scrapings from TG and NTG mice fed LFD or HFD. Principal component analysis of the jejunal transcriptome revealed that 69% of the variation in the transcriptome was explained by the HSP70 transgene, compared to 10% of the variation explained by diet. Based on gene pathway analysis, TG mice on HFD displayed downregulation of lipid transport, biosynthesis, and metabolic processes compared to HFD-NTG mice. Targeted qPCR confirmed these findings; expression of several jejunal lipid metabolism genes was decreased in TG vs NTG mice fed HFD, including diacylglycerol acyltransferase 2 (Dgat2; 50% decrease), fatty acid translocase (Cd36; 93% decrease), and stearoyl-CoA desaturase 1 (Scd1; 95% decrease), p<0.0060. Taken together, these data suggest that DIO resistance in HSP70 TG mice may be due in part to altered lipid metabolism in the small intestine. These findings provide support for intestinal HSP70 as a novel therapeutic in fighting obesity.

<b>HSP70 Significantly Alters the Jejunal Transcriptome</b>. RNA sequencing of jejunal scrapings harvested from HSP70 transgenic (TG) and non-transgenic (NTG) littermates that were fed a low (LFD) or high (HFD) fat diet for 12 weeks and then fasted for 4 hours. <b>A</b>) Principal coordinate analysis demonstrating genotype (PC1, 69%) as the major influencer compared to diet (PC2, 10%). <b>B</b>) Volcano plot of differentially expressed genes in HFD-fed NTG vs TG mice. <b>C</b>) Pathways downregulated in HSP70 TG vs NTG mice fed HFD. <b>D</b>) Pathways upregulated in HSP70 TG vs NTG mice fed HFD (n=5/group).

HSP70 Significantly Alters the Jejunal Transcriptome. RNA sequencing of jejunal scrapings harvested from HSP70 transgenic (TG) and non-transgenic (NTG) littermates that were fed a low (LFD) or high (HFD) fat diet for 12 weeks and then fasted for 4 hours. A) Principal coordinate analysis demonstrating genotype (PC1, 69%) as the major influencer compared to diet (PC2, 10%). B) Volcano plot of differentially expressed genes in HFD-fed NTG vs TG mice. C) Pathways downregulated in HSP70 TG vs NTG mice fed HFD. D) Pathways upregulated in HSP70 TG vs NTG mice fed HFD (n=5/group).

BACKGROUND The gastrointestinal tract is a mechanically vibrant organ. Mechanical inputs are critical as it coordinates transport, digestion and secretion. We recently described the sense of “gut touch”, whereby mechanical inputs are detected by specialized mechanoreceptors in the GI epithelium, similar to Merkel cells that mediate touch sensing in the skin. The gut touch sensors are enteroendocrine cells (EECs), which use Piezo2 mechanogated ion channels to detect mechanical stimuli and transduce them into calcium signals that lead to the release of hormones and neurotransmitters. In cell models, Piezo2 interacts with E-cadherin, which mediates cell-cell contacts in the epithelium. The Piezo2-E-cadherin interaction is a potential mechanism for tethering Piezo2 to the actin cytoskeleton.
METHODS Piezo2 subcellular localization was discerned in tissue EECs and in intestinal organoids via super-resolution imaging. Immunofluorescence was used to determine Piezo2 and E-cadherin co-localization. A high-sensitivity Simple Western was used to test co-immunoprecipitation of E-cadherin with Piezo2 in primary colonic epithelium. To assess functional relevance, EECs were mechanically stimulated after E-cadherin knockdown (KD) and intracellular calcium dynamics measured with a fluorescent genetically encoded calcium indicator (GCaMP5).
RESULTS We identified EECs in small bowel primary tissue and organoids that were double positive for Piezo2 and serotonin (5-HT). We observed that in these cells Piezo2 distributes to 3 subcellular locations: (1) lateral wall, where Piezo2 colocalizes with the cortical actin cytoskeleton; (2) basal cytoplasm, where apparently intracellular Piezo2 clusters accumulate at the base of the cell; and (3) perinuclear, where Piezo2 clusters localize close to the nucleus. Remarkably, no Piezo2 clusters localized to the apical cell compartment. Of these 3 pools, Piezo2 that localizes to the lateral wall and the basal cytoplasm co-localizes with E-cadherin. E-cadherin and actin co-immunoprecipitate with Piezo2 from primary epithelium. Compared to non-target siRNA controls, E-cadherin KD decreased peak calcium responses to mechanical stimulation by 65.2% and total area under the curve (AUC) by 69.3% (n = 13 for NT KD, n = 21 for E-cad KD; p = 0.009 for peak responses, p = 0.045 for AUC).
CONCLUSIONS Piezo2 localizes to the perinuclear and basolateral compartments of EECs and appears to be consistently excluded from the apical compartment. Piezo2 interacts with E-cadherin in primary GI EECs, which we show for the first time in primary tissue. The Piezo2-E-cadherin interaction is functionally relevant for EEC mechanosensitivity. These results shed light on how Piezo2-expressing EECs organize force-sensing molecules within the cell to efficiently transduce forces.

Kimberlee Kossick

Appearances