LPAR5 as a prospective therapeutic target for treating microvillus inclusion disease

1. Introduction

Microvillus inclusion disease (MVID) is a rare congenital diarrheal disorder that typically presents within the first week of life manifesting in severe dehydration and malnutrition. The small intestine of MVID patients have blunted villi and lack numerous apical nutrient transporters, giving rise to persistent watery diarrhea. MVID is an autosomal recessive disorder caused by inactivating mutations in the motor protein myosin VB (MYO5B) (as reviewed in (1)). Despite significant research investigating the brush border deficits in MVID, there are only two therapeutic options for MVID patients: continuous total parenteral nutrition (TPN) therapy or small bowel transplantation, both of which are associated with various chronic side effects.

Studies utilizing MYO5B-deficient mouse models have suggested new treatment opportunities for MVID. Therapeutic pathways of interest including bypassing the apical trafficking blockage induced by a loss of MYO5B function and/or enhancing proper enterocyte maturation from MYO5B-deficient progenitor cells. Based on previous work from others studying lysophosphatidic acid (LPA) and its associated receptors (LPARs) in the intestine, we recently evaluated the therapeutic potential of LPAR5 agonist, Compound 1, on MVID model mice and enteroids (2).

2. Epithelial Dysfunction in Microvillus Inclusion Disease

2.1. MYO5B as a regulator of apical trafficking in enterocytes

Myosins are actin-binding molecular motors, and myosin V family members transport cellular cargo for vesicle trafficking and cell polarity. MYO5B interacts with Rab small GTPases to allow for membrane recycling and the delivery of endosomes to the apical surface of epithelial cells (3). Enterocytes of both MVID patients and MYO5B-deficient animal models illustrate the importance of proper MYO5B function in apical protein transport, marked by shortened microvilli and internalization of nutrient transporters such as sodium-dependent glucose transporter 1 (SGLT1) and sodium hydrogen exchanger 3 (NHE3) (4, 5). However, not all apical transporters are mislocalized in MYO5B-deficient epithelial cells. Specifically, cystic fibrosis transmembrane conductance regulator (CFTR) is retained on the apical membrane (4, 6). Although the maintenance of CFTR likely exacerbates diarrheal symptoms through chloride secretion, the preserved localization of CFTR presents a potential therapeutic strategy for MVID by suggesting there are alternative, MYO5B-independent, apical trafficking pathways in enterocytes.

2.2. MYO5B as a regulator of small intestinal crypt function

In addition to the well-studied trafficking deficits in MYO5B-deficient cells, we have demonstrated differentiation defects within the small intestinal crypts in MVID model mice and crypt-derived organoids (enteroids). Similar to MVID patient biopsies, crypts are elongated in intestinal epithelial-specific MYO5B knockout (MYO5BΔIEC) mice and MYO5B(G519R) mice, which possess an MVID patient-derived point mutation (7, 8). This crypt phenotype highlights the increase in immature, proliferative cells in the MYO5B-deficient intestine. Progenitor cells within these crypts have disordered differentiation, marked by significant decreases in both tuft cell and mature enterocyte numbers (9). The MVID characteristic enterocytes with shortened microvilli and a lack of villus-specific apical transporters could be explained by this immature cell phenotype.

Cellular mechanisms underlying crypt function alterations caused by MYO5B loss have begun to be investigated. Bulk RNA-sequencing of MYO5BΔIEC mouse jejunum has revealed an imbalance in Wnt:Notch signaling, where transcription levels of Wnt ligands and targets were decreased while Notch-related gene expression was maintained. Proper Wnt:Notch signaling is necessary for both maintaining the proliferative progenitor cell niche and proper cell differentiation (10). The decrease in Wnt signaling in the elongated MYO5B-deficient crypt was unexpected as Wnts are essential for proliferation, suggenting that alternative pro-proliferative factors might be present in the MYO5B-deficient intestine. Although treatment with a Notch / gamma-secretase inhibitor, dibenzazepine (DBZ), in MYO5BΔIEC mice restored normal secretory cell numbers, DBZ did not correct the diarrhea-induced weight loss seen in these mice possibly due to Notch inhibition toxicity. Additionally, our recent study suggests that alterations in epithelial cell metabolism, such as mitochondrial structure and lipid metabolism, could contribute to the crypt dysfunction seen in MVID (2). Correcting these adverse metabolic changes in MVID intestines could improve cell differentiation and promote proper cell differentiation.

3. LPAR5 as a Therapeutic Target for MVID

3.1. LPA and the Small Intestine

LPA is a bioactive phospholipid molecule found in a variety of foods and synthesized by various organs, including the intestine, by phospholipases and autotaxin (recently reviewed in (11)). LPA has growth factor-like effects and acts through G protein-coupled receptors LPARs 1–6 and other orphan receptors.

LPAR transcription has been detected in the human and mouse small intestine. Previous work has demonstrated LPAR subtype-specific activation in intestinal epithelial cells: LPAR1 promotes intestinal organoid growth and differentiation, LPAR2 inhibits CFTR-dependent cholera toxin-induced diarrhea, and LPAR5 promotes water absorption through the apical trafficking of NHE3. Furthermore, LPAR5 knockout mouse models reveal the importance of LPAR5 in the protection of Lgr5+ stem cells and mucosal regeneration (12, 13).

3.2. LPA and LPAR5 Treatment in MVID Model Mice

Based on the data demonstrating LPA signaling promoting intestinal water absorption, we have assessed LPA effects on MVID models. LPA(18:1) intraperitoneal injection (IP) or oral gavage was compared to the LPAR2 agonist GRI977143 (IP) in MYO5BΔIEC mice (7). LPA administration by IP yielded the best results, illustrated by a reduction in villus blunting, a partial rescue of SGLT1 and NHE3 brush border localization, a decrease in CFTR-mediated anion secretion, and an increase in SGLT1 activity. LPAR2 agonist treatment decreased CFTR anion secretion similarly to LPA (IP), but did not significantly improve nutrient transporter apical trafficking or SGLT1 activity. These data suggest LPA improves brush border maturation in a non-LPAR2 manner. LPA treatment did not alleviate MYO5B loss-induced crypt elongation or weight loss, likely due to inconsistent exposure of immature enterocytes to LPA.

To elucidate which LPAR subtype is responsible for the therapeutic effects of LPA treatment and overcome the low solubility and stability of natural LPA, we recently evaluated a synthetic LPAR5 agonist (Compound-1) in MYO5BΔIEC and MVID patient-modeled MYO5B(G519R) mice (2). Although expression levels of all detectable Lpar subtypes are significantly decreased by MYO5B loss, Lpar5 levels are still prominent in MYO5B-deficient epithelial cells. Both MYO5B(G519R) and MYO5BΔIEC intestines display cell metabolic deficits, disrupted mitochondrial structures, and brush border defects in intestinal epithelial cells. Compound-1 reduced villus blunting and improved SGLT1 and NHE3 apical localization and partially improved mitochondrial distribution in MYO5B-deficient enterocytes, compared to vehicle-treated mice. These data suggest the therapeutic potential of LPAR5 activation for MVID (Figure 1).

Figure 1.

Trophic effects of LPAR5 activation in MVID model mice. LPA and synthetic LPAR5 agonist are predicted to target progenitor cell metabolism and functions in intestinal crypts. Facilitated differentiation and maturation of sensory tuft cells and absorptive enterocytes result in amelioration of villus blunting. The illustration was created with BioRender.com.

This recent study also illustrates the differential effects of Compound-1 on MYO5B knockout versus point mutation-harboring mice, and sex-specific responses. Compound-1 partially rescued diarrhea-induced weight loss of only male MYO5B(G519R) mice, but not female mice or MYO5BΔIEC mice of either sex. These observations indicate that models with real-world patient-based mutations, not only genetic knockout models, and models of both sexes should be included in drug assessment. In pararell, Compound-1 treatment reversed the decreased tuft cell numbers in MVID pig enteroids, carrying a patient-modeled point mutation at P663L (orthologue of human MYO5B(P660L)). These observations suggest that LPAR5 acts through epithelial cell-autonomous and systemic pathways, and LPAR5 pathway is likely influenced by MYO5B mutation status.

4. Expert Opinion

To date, there are no non-invasive therapeutics for MVID patients. To provide more treatment options beyond TPN or intestinal transplant, several pathways of interest have been under investigation. Antidiarrheal drugs have been proposed to target chloride channels in intestinal epithelial cells, such as CFTR and calcium-activated chloride channels (CaCCs) (14). Another treatment approach aims to stimulate enterocyte maturation to facilitate proper nutrient uptake in MVID patients. Recent findings from basic research shed light on alterations in MYO5B-deficient intestinal progenitor cell signaling and mitochondrial metabolism, which serve as novel therapeutic targets to enhance enterocyte maturation and overcome the severe malabsorption symptoms of MVID. Treatment with Compound-1, a synthetic LPAR5 agonist, in vivo and in enteroid models demonstrated positive effects on cell differentiation and maturation, although the efficacy of Compound-1 varied depending on sex and MYO5B mutation (2). A combination treatment with LPAR5 activation and chloride channel inhibition could be utilized in future studies to investigate additive therapeutic effects for MVID patients.

Recently developed in vitro platforms involving patient-derived enteroids or genetically modified healthy enteroid lines are useful tools to screen potential therapeutics in human cells. However, they typically lack physiological interactions with non-epithelial cells. Patient-modeled mouse models overcome this limitation and allow us to examine various treatment routes. Initially, we have examined oral gavage and subcutaneous pellet insertion of Compound-1, however, neither route demonstrated consistent effects to a similar extent to IP injection in MVID mouse models. Still, human tissue-specific drug kinetics and efficient drug delivery routes need to be optimized and future long-term studies are necessary to evaluate adverse side effects of LPAR5 agonists for clinical use.

Compound-1 has similar effects to LPA on alleveating MVID nutrient malabsorption. However, compared to natural LPA, Compound-1 shows high affinity and preference to human LPAR5 and has greater water solubility. These characteristics allow for Compound-1 to be developed as a progenitor cell-targeted, manageable drug for MVID. The therapeutic potential of Compound-1 in these studies was likely limited due to inconsistent exposure of the drug on developing enterocytes (single IP injection daily). Future work will aim to provide a continuous delivery method of Compound-1 to promote proper cell maturation in congenital diarrheal diseases including, but not limited to, MVID, which is arising from different gene mutations (1, 15).