Larazotide acetate: a pharmacological peptide approach to tight junction regulation

Abstract

Larazotide acetate (LA) is a single-chain peptide of eight amino acids that acts as a tight junction regulator to restore intestinal barrier function. LA is currently being studied in phase III clinical trials and is orally administered to adult patients with celiac disease as an adjunct therapeutic to enhance intestinal barrier function that has been disrupted by gliadin-induced immune reactivity. Mechanistically, LA is thought to act as a zonulin antagonist to reduce zonulin-induced increases in barrier permeability and has been associated with the redistribution and rearrangement of tight junction proteins and actin filaments to restore intestinal barrier function. More recently, LA has been linked to inhibition of myosin light chain kinase, which likely reduces tension on actin filaments, thereby facilitating tight junction closure. Small (rodent) and large (porcine) animal studies have been conducted that demonstrate the importance of LA as a tight junction regulatory peptide in conditions other than celiac disease, including collagen-induced arthritis in mice and intestinal ischemic injury in pigs.

INTRODUCTION

Celiac disease is an autoimmune gastrointestinal (GI) disease in which patients experience an intestinal inflammatory response after ingesting food material containing gluten. This can result in various clinical symptoms including vomiting, diarrhea, constipation, and fatigue (1). Throughout the Western world, celiac disease has a prevalence of ∼1%, affecting ∼3 million people within the United States alone, where it reportedly has a diagnosis rate of only 15% (2–4). As of now, the only therapeutic option for people with celiac disease is the avoidance of dietary gluten, although patients adhering to a gluten-free diet may not be free of persistent symptoms. The importance of protecting the intestinal barrier before injury and restoration of barrier function after injury is crucial in various disease pathologies including celiac disease. Therefore, studying and understanding methods that can target intestinal tight junction proteins to protect or restore barrier function is crucial for the development of therapeutic options.

Larazotide acetate (LA; INN-202), is an intestinal tight-junction regulator drug from 9 Meters Biopharma, Inc., previously known as Innovate Biopharmaceuticals before a merger with RDD Pharma, that is currently being tested in adult patients with celiac disease in a phase III clinical trial (5). This single-chain peptide is derived from human zonulin, a known modulator of tight junction permeability that has elevated intestinal tissue expression in patients with celiac disease (6, 7). LA also shares structural similarities to the Vibrio cholerae zonula occludens toxin (Zot), which is an enterotoxin that disrupts intercellular tight junction structure through actin polymerization and cytoskeletal modifications that result in increased intestinal permeability through intracellular signaling (8). Although LA has shown promise in clinical trials for patients with celiac disease, recent studies using various animal injury models indicate a potential use for LA for other diseases because of its ability to regulate tight junctions. This mini-review aims to provide a detailed source of current information on this octapeptide.

TIGHT JUNCTION STRUCTURE OVERVIEW

The intestinal epithelial monolayer serves as both an innate immune barrier to restrict translocation of potentially harmful luminal contents, including microorganisms, toxins, and foreign antigens, as well as acting as a selective filter to allow for proper absorption of nutrients, water, and electrolytes into the circulation (9). Paracellular and transcellular pathways maintain these functions with intestinal tight junction proteins acting as the primary regulators of paracellular permeability. Tight junction structures are made up of various elements, including transmembrane proteins (claudins, occludin, and tricellulin), tight junction-associated proteins (junctional adhesion molecules 1–3), and intracellular scaffold proteins (zonula occludens 1–3), and regulate paracellular passage of molecules based on charge and size selectivity under homeostatic conditions (10). In addition, intracellular scaffold proteins interface with the actin cytoskeleton, linking transmembrane tight junction proteins to the cytoskeleton and subsequently acting as a molecular target for tight junction dysregulation in various pathways, including the Zot and zonulin pathways, which both involve the dysregulation of tight junction proteins to induce alterations in intestinal permeability (7, 11–15). Maintenance of these structures are crucial, because the dysregulation of tight junction proteins during injurious events, such as in celiac disease or intestinal ischemia, and the inability to repair tight junction structures after injury can result in intestinal inflammation, sepsis, and multiple organ dysfunction in clinical patients (16, 17). Therefore, understanding what factors can negatively modulate tight junction permeability and how pharmacological approaches such as LA act to protect or restore the tight junction barrier before or after injury is critical for developing preventative or therapeutic approaches.

CHEMICAL PROPERTIES OF LARAZOTIDE ACETATE

LA is a single-chain peptide consisting of eight natural amino acids (Gly-Gly-Val-Leu-Val-Gln-Pro-Gly) derived from human fetal zonulin and is structurally similar to Zot, which is an enterotoxic product of the V. cholerae bacterium. The Zot enterotoxin was first discovered through Ussing chamber experiments in which rabbit ileal tissue was exposed to V. cholerae culture supernatants for a period of time and subsequently collected for electron microscopic observation of tight junction proteins. During the exposure period, Fasano et al. (15) observed modulated ileal permeability via elevations in transepithelial electrical conductance as well as a significant increase in the number of altered zonulae occludentes strands. Thus, the factor causing the modulations in intestinal permeability and tight junction alterations was designated as Zot. Following this, additional studies demonstrated the ability of the purified Zot protein to increase intestinal permeability exhibited by significant decreases in both jejunal and ileal tissue transepithelial electrical resistance (TEER) (18). Mechanistically, Zot acts to open tight junctions through actin microfilament polymerization in a protein kinase C-dependent manner (19). This mechanism of Zot’s action on actin was discovered through exposing rabbit ileal tissue and a rat small intestinal cell line, IEC-6 cells, to Zot and observing subsequent changes to the F-actin cytoskeleton. After exposure to Zot, F-actin redistribution was noted in both rabbit ileal tissue and IEC6 cells, with redistribution paralleling increased permeability in the ileal tissue (19).

Because of its ability to increase intestinal permeability, functional screening of Zot protein was conducted, leading to the discovery of a human intestinal Zot analogue, zonulin, which, like Zot, has been shown to reduce TEER of intestinal tissue (7). During this study by Wang et al. (7), zonulin was purified from both human fetal and human adult intestinal tissue and found to have some structural differences at the 8–15 NH₂-terminal amino acid residues between these two human zonulin sequences. However, the presence of conserved amino acids in the shared motif between Zot and zonulin lead these researchers to synthesize an octapeptide, now known as LA, with the same amino acid sequence as the NH₂-terminus of human fetal zonulin, positions 8–15 (7). In the same study, this synthetic octapeptide/LA was applied to Rhesus monkey intestinal tissue mounted on Ussing chambers in the presence of either Zot or zonulin. Although both Zot and zonulin-treated tissues exhibited reduced tissue resistance, LA prevented these Zot and zonulin-induced changes in TEER (7). This milestone study was crucial for understanding the relationship between Zot and zonulin, demonstrated an NH₂-terminus receptor-binding motif shared between Zot and zonulin, and provided the foundational peptide drug now being studied in phase III clinical trials for adult patients with celiac disease.

LA is the first pharmacological treatment ever tested in phase III clinical trials for celiac disease (5). A unique characteristic of this orally administered peptide is its restriction to the gut, resulting in a localized effect. Nonsystemic, intestine-targeted (NSIT) drugs are novel in that they are inherently designed to not be absorbed into systemic circulation but instead act locally within the GI tract (20). Interestingly, therapeutic targeting not only includes GI disorders and enteric infections but certain systemic diseases as well, including mineral metabolic disorders (hyperphosphatemia, hyperkalemia, or sodium overload) and diabetes. NSIT drug discoveries are extremely valuable since therapeutic activity within the GI tract is high whereas the risk for systemic exposure that can potentially lead to whole body toxicities is diminished (21).

Although LA is targeted specifically to the GI tract due to its nonabsorbed nature, understanding how long various formulations of the drug take to reach specific regions within the small intestine and in therapeutically effective concentrations are crucial for its use in other clinical conditions. A recent in vivo study using a porcine in vivo model demonstrated some pharmacokinetic properties of LA in its movement throughout the GI tract (22). This study sought to determine the concentration-versus-time profile of LA after oral administration of a specific delayed-release formulation of LA by measuring concentrations in intestinal fluid collected via ultrafiltration sampling probes placed at select locations in the small intestine. Although LA was detected throughout the small intestine during the entirety of the 4-h collection period following oral administration, peak concentrations were located in the duodenum and proximal jejunum 1 h after dosing (22). Indication of the presence and speed at which LA reaches select sites within the porcine intestinal tract provides useful knowledge for developing future LA formulations to target other intestinal regions and to hasten/delay the release of LA at these sites and for other indications. Taken together, LA is a novel therapeutic in that it exhibits NSIT characteristics due to its nonsystemic, gut-restricted nature, and reaches the small intestine intact where it is hoped that it will have a therapeutic effect in human patients with celiac disease.

PROPOSED MECHANISM OF ACTION OF ZONULIN AND LARAZOTIDE ACETATE

In patients with celiac disease, LA is proposed to restore disrupted tight junction structure, thereby preventing gliadin, the immunostimulatory protein generated from the breakdown of gluten, from permeating the epithelial barrier (23–26). Without exposure of the lamina propria to gliadin and the prevention of gliadin and gliadin fragments from entering into systemic circulation, the inflammatory immune reaction typically seen in patients with celiac disease can be averted. A proposed mechanism by which LA normalizes tight junction structure to prevent gliadin from crossing the intestinal epithelium is through regulation of the endogenous zonulin pathway. From a molecular standpoint, zonulin is known as prehaptoglobin-2 (pre-HP2), which can be enzymatically cleaved into haptoglobin-2 (HP2) and was previously believed to only serve as the inactive precursor to HP2 (27). Haptoglobins are known to form stable complexes with free hemoglobin (Hb), and these HP-Hb complexes act as antioxidants to prevent oxidative tissue damage through removal of the oxidative potential of Hb iron (28, 29). However, in addition to acting as a proprotein for HP2, zonulin is now known to function as a modulator of intestinal permeability through regulation of intercellular tight junction structure, and its expression is increased in disease states such as celiac disease (6, 30).

Currently, there are two described triggers by which zonulin is released into the lumen: 1) intestinal epithelial apical exposure to enteric pathogens including Escherichia coli and Salmonella typhi; and 2) luminal exposure to gliadin (31). Upon the release of zonulin into the lumen as a result of intestinal exposure to gliadin, zonulin can interact with two receptors, epidermal growth factor receptor (EGFR) and protease-activated receptor 2 (PAR2), which results in increased intestinal permeability (27, 32). It was previously proposed that the zonulin pathway is activated through direct binding of zonulin to EGFR, transactivation of EGFR through PAR2 activation, or a combination of both mechanisms (30). However, current literature suggests that alterations in intestinal permeability by zonulin is primarily dependent on PAR2 transactivation of EGFR since PAR2 has been described to transactivate EGFR, and in a study by Tripathi et al. (27), mice lacking PAR2 did not exhibit EGFR-dependent reduction in TEER when exposed to zonulin. Binding of zonulin to these receptors and downstream signaling results in actin polymerization and tight junction protein phosphorylation, which causes tight junction disassembly through protein displacement from the tight junction complex (33). The zonulin-induced disruption of the tight junction complex allows paracellular transport of gliadin and the eventual uptake of gliadin into systemic circulation, thereby causing the inflammatory response exhibited by patients with celiac disease (26, 34, 35). Although this mechanism by which zonulin dysregulates tight junctions has been well described thus far, it is important to note the controversy around EGFR localization within the apical membrane of intestinal epithelial cells and the implications this controversy may have on this specific mechanism. A study by Avissar et al. (36) demonstrated redistribution of EGFR to the apical membrane of rabbit intestinal epithelium following small bowel resection, whereas Kelly et al. (37) notes EGFR localization within both the apical and basolateral membranes of enterocytes of newborn and weaned porcine intestine. However, other studies have demonstrated the presence of EGFR concentrated solely within the basolateral membrane of intestinal epithelium in rat intestine as well as in normal human fetal and normal adult intestinal epithelium (38–40). These contrasting studies suggest that there are preexisting conditions of the gut resulting in redistribution of EGFR to the apical membrane that might allow interaction with EGFR and zonulin. Furthermore, increased permeability of the intestine in patients with celiac disease may allow zonulin to reach basolateral EGFR receptors. Nonetheless, the precise mechanisms of zonulin receptor binding remain incompletely understood and is an area in need of additional study.

To prevent the downstream effect of zonulin binding to its receptors after intestinal gliadin exposure, it is proposed that LA antagonizes the zonulin pathway by competitively inhibiting zonulin binding to target receptors, but further mechanistic studies are needed in order to support this proposal (Fig. 1) (26, 32). Further supporting the role of LA as a zonulin inhibitor, one study found that LA treatment reduced both rearrangement and redistribution of actin and the tight junction-associated protein zonula occludens-1 (ZO-1) in Caco-2 brush-border-expressing cells that were exposed to AT-1002, a synthetic hexamer peptide derived from the active Zot fragment that has been shown to increase tight junction permeability (24). AT-1002 alters intestinal permeability in a PAR2-dependent manner with increases in permeability comparable with Zot and zonulin-induced alterations (41–43). More specifically, the PAR2 activating peptide tethering motif (SLIGRL) and AT-1002 (FCIGRL) are structurally similar, and this motif similarity may be responsible for AT-1002-associated occludin and ZO-1 displacement, leading to increased intestinal permeability in the Caco-2 cells mentioned previously (41).

Figure 1.

Potential mechanism of zonulin pathway inhibition by larazotide acetate (LA) before or after intestinal exposure to gluten. The left panel of this figure represents a previously described mechanistic overview of the zonulin-dependent disruption of tight junction structure after intestinal gliadin exposure, whereas the right panel depicts possible mechanisms for the action of LA to protect against or restore intestinal barrier function before or after gliadin exposure, respectively. The stepwise presentation is as follows: 1) gliadin is ingested through oral consumption of glutenous food. 2) gliadin binds a chemokine receptor, CXCR3, to elicit a MyD88-dependent release of zonulin from intestinal epithelial cells into the intestinal lumen (32). 3 and 4) zonulin binds to EGFR and PAR2, which transactivates EGFR in a PAR2-dependent manner, resulting in tight junction displacement via actin polymerization and tight junction protein internalization (33). This ultimately allows gliadin to pass paracellularly between intestinal epithelial cells, down to the lamina propria, and into systemic circulation. 5 and 6) oral administration of LA before the consumption of gluten (protective) or after consumption (restorative) is proposed to competitively inhibit zonulin binding to its receptors, thereby preventing tight junction disassembly or restoring the tight junction barrier.

Other studies suggest that LA stabilizes tight junction structure through rearrangement of actin filaments and by promoting tight junction assembly (44). Specifically, one study found that when leaky Madin-Darby canine kidney type II cells were treated with LA, tight junction proteins, including occludin and claudin-3, had less intracellular distribution (and concurrently more membrane expression), whereas actin was more associated with intercellular junctions (25). A possible mechanism by which LA regulates rearrangement of actin filaments to promote tight junction reassembly is through the myosin light chain (MLC) pathway. MLC phosphorylation (pMLC) by either myosin light chain kinase (MLCK) or ρ-associated coiled coil containing protein kinase (ROCK) stimulates tight junction regulation via contraction of the perijunctional actomyosin ring (PAMR) (45, 46). This contractile event leads to opening of tight junction structures, as transmembrane tight junction proteins are linked to the PAMR through peripheral membrane proteins such as ZO-1 (45). A recent study has implicated an effect of LA on MLC through inhibition of pMLC (47). In this study, Caco-2 brush-border-expressing cells that were pretreated with LA and exposed to anoxia/reoxygenation (A/R) contained significantly increased levels of membrane-bound occludin when compared with untreated anoxia-injured cells. In addition, A/R-induced increases in pMLC were attenuated by pretreatment with LA in these cells, demonstrating the role of LA in the MLC pathway of tight junction regulation. Further mechanistic studies are needed to determine how LA regulates the MLC pathway through potentially MLCK and/or ROCK inhibition (Fig. 2). Taken together, these data suggest that LA regulates tight junction assembly in leaky epithelia in part via a mechanism related to regulation of MLC phosphorylation.

Figure 2.

Potential mechanisms by which larazotide acetate (LA) regulates the TJ barrier after intestinal ischemic injury. The TJ barrier is highly regulated by the phosphorylation state of myosin light chain (MLC). This phosphorylation process is primarily controlled by myosin light chain kinase (MLCK) and myosin light chain phosphatase (MLCP). ρ-associated protein kinase (ROCK) inhibits MLCP through phosphorylation of an MLCP subunit. Therefore, we propose that LA may act to inhibit MLCK and/or ROCK to facilitate reduced tight junction disruption or tight junction repair.

LARAZOTIDE ACETATE AND ITS ROLE IN REGULATION OF TIGHT JUNCTION PROTEINS IN OTHER INJURY MODELS

Although LA is currently being studied in adult patients with celiac disease, its use to “tighten” intestinal tight junction structure to ameliorate or restore pathologically induced increases in intestinal permeability in other conditions is promising. For example, one research group studied the effect of LA in collagen-induced arthritis (CIA) mice and its activity as a zonulin antagonist (48). In this study, intestinal barrier function was found to be diminished before onset of arthritis in mice. In addition, zonulin has found to be elevated in the serum of patients with rheumatoid arthritis. In a small-animal model of arthritis, zonulin acted as a disruptor of the tight junction barrier by reducing expression of occludin and ZO-1, associated with increased intestinal permeability to macromolecules (48). Interestingly, LA treatment of CIA mice before onset of arthritis prevented the zonulin-induced increases in macromolecular flux as well as increasing mRNA expression of tight junction proteins. These data support the hypothesis that LA acts as a zonulin antagonist and demonstrates its potential for reducing onset of other disease conditions, aside from celiac disease.

As an example of the activity of LA in nonceliac digestive disease, LA was recently used to enhance recovery of ischemia-injured jejunum in a large animal model of intestinal ischemia/reperfusion injury (49). When ischemia-injured jejunum was recovered ex vivo on Ussing chambers in the presence of a single dose of apically administered LA, restoration of barrier function was significantly enhanced in comparison with untreated ischemic controls, as indicated by elevated TEER and decreased mucosal-to-serosal macromolecular flux. In addition, LA treatment altered membrane-bound versus cytoplasmic expression of tight junction proteins. Together, these data suggest that LA directly modulates tight junction structure during recovery after acute injury.

In the aforementioned study, an interesting finding was that tissue treated with a higher dose (10 μM) of LA was ineffective in stimulating repair after intestinal ischemic injury versus the effective dose of 1 μM (49). This is potentially due to the formation of inhibitory fragments by brush-border enzymatic degradation of LA during the ex vivo recovery process. These fragments of LA formed by intestinal brush-border enzymes were synthesized and applied to the mucosal tissue during recovery during separate experiments, and one fragment, LA fragment 2 (VLVQPG) which has two glycine residues removed from NH₂-terminus of LA, exhibited an inhibitory effect on recovery when given as 10 times overdose (10 μM) with 1 μM LA compared with untreated ischemia-injured tissue (49). LA fragment 2 is proposed to be formed by brush-border enzyme aminopeptidase M when incubated in vitro and indicates that fragments of LA have the potential to block the recovery effects of LA. These results may explain the ineffectiveness of the 10 μM dose of LA treatment, as the increased formation of peptide fragments (particularly LA fragment 2) may share enough similarities to the Zot and zonulin receptor-binding motif (GXXXVQXG) to competitively inhibit LA binding to EGFR and/or PAR2. In addition, there may be potential that these fragments are binding to these same receptors to elicit a zonulin-like effect in tight junction dysregulation, versus an inhibition of zonulin activity, although mechanistic studies need to be conducted to fully explore this possibility.

The formation of LA inhibitory fragments presents a pharmacological dilemma, because it has the potential to limit efficacy of the parent LA molecule. To potentially prevent or slow down LA fragmentation by the brush-border enzymes responsible for LA fragmentation, chiral modification of the entire LA sequence to all D amino acids was conducted, which resulted in the generation of LA Analog 6 (A6) by Innovate Biopharmaceuticals, Inc. (now 9 Meters Biopharma, Inc.) (49, 50). Interestingly, A6 enhanced recovery of ischemia-injured jejunum at a 10-fold lower dose (0.1 μM) than the effective dose of LA in the same injury model (50). The brush-border enzyme fragmentation of A6 was significantly slower than LA. However, fragments of A6 were still detected with mass spectrometry analysis, and a high dose (10 µM) of one fragment did exhibit inhibitory effects on recovery when applied in tandem with A6 (0.1 µM). An alternative approach that has been explored experimentally in the porcine ischemic injury model to limit formation of LA inhibitory fragments while maintaining the ability of LA to induce recovery of barrier function involves addition of small doses of LA over time and suggest that 0.1 μM LA applied every 45 min is effective (49). Further experiments will be required to assess this “micro-dosing” but suggest a delayed release LA formulation may be worth pursuing.

CONCLUSIONS

Development of novel therapeutics that restore the tight junction barrier after gliadin exposure in patients with celiac disease is crucial, given that the only current treatment option is adherence to a strict, gluten-free diet, which is not wholly successful in all patients with celiac. LA has shown to be effective as a tight junction regulator and shows promise in its efficacy as an adjunct therapeutic with a gluten-free diet (44, 51, 52). In addition, patients given LA are at lower risk for developing toxicity issues due to its nonsystemic, gut-restricted effects. As a proposed zonulin antagonist and tight-junction regulatory peptide, the use of LA may prove useful for other patient indications given current data in both small and large animal models of injury.

GRANTS

Supported was given by Innovate Biopharmaceuticals, Inc. (367652 to A.T. Blikslager).

DISCLOSURES

J. Madan was a major share owner in Innovate Biopharmaceuticals Inc., which has since merged to form 9 Meters Biopharma. He is not affiliated with 9 Meters Biopharma. B.R. Krishnan and A.T. Blikslager consulted for Innovate Biopharmaceuticals. A.T. Blikslager was funded by Innovate Biopharmaceuticals (declared under funding). Z.M. Slifer has no conflicts of interest, financial or otherwise, to disclose.

AUTHOR CONTRIBUTIONS

Z.M.S. and B.R.K. conceived and designed research; B.R.K. analyzed data; Z.M.S. interpreted results of experiments; Z.M.S. drafted manuscript; Z.M.S., B.R.K., J.M., and A.T.B. edited and revised manuscript; Z.M.S., B.R.K., J.M., and A.T.B. approved final version of manuscript.