Structure, Function, and Therapeutic Potential of the Trefoil Factor Family in the Gastrointestinal Tract

Abstract

Trefoil factor family peptides (TFF1, TFF2, and TFF3) are key players in protecting, maintaining, and repairing the gastrointestinal tract. Accordingly, they have the therapeutic potential to treat and prevent a variety of gastrointestinal disorders associated with mucosal damage. TFF peptides share a conserved motif, including three disulfide bonds that stabilize a well-defined three-loop-structure reminiscent of a trefoil. Although multiple functions have been described for TFF peptides, their mechanisms at the molecular level remain poorly understood. This review presents the status quo of TFF research relating to gastrointestinal disorders. Putative TFF receptors and protein partners are described and critically evaluated. The therapeutic potential of these peptides in gastrointestinal disorders where altered mucosal biology plays a crucial role in the underlying etiology is discussed. Finally, areas of investigation that require further research are addressed. Thus, this review provides a comprehensive update on TFF literature as well as guidance toward future research to better understand this peptide family and its therapeutic potential for the treatment of gastrointestinal disorders.

1. Introduction

Trefoil factor family (TFF) peptides are key players in maintaining and repairing the body’s mucosa. They are particularly well-known for their functions in the gastrointestinal tract, where they regulate gut homeostasis. Outside the gastrointestinal tract, TFF peptides are found in the respiratory tract, urinary tract, uterus, eyes, and salivary glands, where they have similar functions in mucosal homeostasis and repair.¹⁻⁷ They have also been detected in human breast milk and the brain.⁷⁻⁹ The mammalian TFF has three members: TFF1, TFF2, and TFF3. TFF1 (60 amino acid residues) and TFF3 (59 amino acid residues) contain a single highly conserved trefoil domain (formerly named P-type domain), whereas TFF2 (106 amino acids long) comprises two structurally homologous TFF domains (Figure 1).¹⁰ Originally, TFF1, TFF2, and TFF3 were named pS2, spasmolytic peptide, and intestinal trefoil factor, respectively, until 1997 when the current nomenclature was adopted at the Philippe Laudat Conference.¹¹

Figure 1.

3D NMR structure and trefoil domain sequence of the three mammalian members of the trefoil factor family. (A) 3D NMR structure of TFF1, TFF2, and TFF3. TFF domains are shown in blue. N- and C-terminals and linker region in TFF2 are shown in green. Disulfide bonds are highlighted in yellow. (B) Amino acid sequences of human TFF1, porcine TFF2, and human TFF3, highlighting the trefoil domain (gray) and the disulfide bonds inside the trefoil domain, as well as the extra disulfide bond between Cys⁶ and Cys¹⁰⁴ outside the domain in TFF2 (orange).

TFF peptides are expressed throughout the gastrointestinal tract, with TFF1 and TFF2 being mainly present in the stomach and TFF3 in the intestine.¹²⁻¹⁷ Their high concentration (micromolar) indicates that they are integral components of the mucosal barrier, where they regulate mucous gel viscosity and interact with proteins and receptors.^12,13,18,19 Thus, they hold therapeutic potential for disorders requiring epithelial repair, including inflammatory bowel diseases (IBD) and nonsteroidal anti-inflammatory drug (NSAID)-induced gastritis. This potential is supported by several preclinical²⁰⁻²³ and clinical studies.²⁴⁻²⁷ However, despite such potential, the mechanisms of action of TFF peptides remains largely unknown. Major knowledge gaps exist in understanding their pharmacokinetics, target receptors, and structure–activity relationships, with these limitations hampering therapeutic development.^7,18

This review focuses on the (patho)physiological roles and therapeutic potential of TFF peptides in the gastrointestinal tract, since this is where they are most abundantly expressed. It provides an update on the current literature, highlights existing knowledge gaps, and discusses mechanistic, structural, and therapeutic aspects to guide future research toward a better understanding of their mechanisms of action and clinical translation for the treatment of gastrointestinal disorders.

2. Structural Aspects and Complexity of the Trefoil Factor Family

The trefoil domain is defined by a highly conserved cysteine framework (CX_9–10C₉CX₄CCX₁₀C, Cys^I–V, Cys^II–IV, and Cys^III–VI), which yields a compact three-looped structure reminiscent of a trefoil (Figure 1).^25,28,29 This structure is maintained by its three disulfide bonds, making it relatively resistant to proteolytic degradation.^{10,25,28,30,31} Therefore, unlike many other bioactive peptides, TFF peptides are functional within the harsh environment of the gastrointestinal lumen. Disulfide bond connections were elucidated based on proteolytic/chemical digestion³² and X-ray crystallography³³ studies of TFF2 isolated from porcine pancreas. The same disulfide bond configuration was determined for recombinant and synthetic human TFF1^34,35 and recombinant TFF3³⁶ using NMR.

TFF1 and TFF3 have a seventh unpaired cysteine residue (Cys^57/58) that can form cross-links with other TFF peptides or proteins.^8,37−39 Monomeric,^8,40,41 homodimeric (TFF1–TFF1; TFF3–TFF3),⁸ and heterodimeric (TFF1 or TFF3 with a partner protein)^13,30,42 forms have been found, yet their physiological purposes remain elusive. Homodimers are often more potent than their monomeric counterparts. For example, subcutaneously injected TFF1 homodimer results in greater mucosal protective effects in an indomethacin-induced gastric injury rat model than does TFF1 monomer,⁴³ while intracolonically administered homodimeric, but not monomeric, TFF3 improves colitis in rats.⁴⁴

There are clear structural differences between TFF2 and homodimeric TFF1 and TFF3 (Figure 2).⁴⁵ TFF2 contains two trefoil domains in a single chain that are held in a fixed orientation by an extra disulfide bond linking the N- and C- termini (Figure 2).⁴⁶ By contrast, the two trefoil domains of the homodimeric TFF1 and TFF3 can rotate freely due to their long and flexible disulfide bond-connected C-termini.^38,47 Such a characteristic might impact the specificity of the TFF binding to their target protein and/or have implications on their mechanisms of action.^38,47

Figure 2.

3D NMR structures of homodimeric TFF peptides. (A) Human TFF1 (hTFF1) homodimer (PDB: 1HI7). (B) Porcine TFF2 (pTFF2) dimer (PDB: 1PCP). (C) Human TFF3 (hTFF3) homodimer (PDB: 1PE3).

Very few TFF structure–activity relationship studies have been carried out to date. Cysteine-to-serine substitutions in TFF3 highlighted the importance of the disulfide framework for activity and stability since corresponding serine analogues are neither protease-resistant nor able to induce cell migration.²⁹ TFF3 homodimer exhibits antiapoptotic and cell-migration-promoting activity, while TFF3 monomer lacks the antiapoptotic activity but retains the motogenic activity.²⁹ Although no pharmacophore has yet been identified, it is speculated to sit within the trefoil loop structures that display different secondary structural motifs.³⁶ Linear peptides derived from loop 2 of TFF2 domain 1 (GFPGITSD) and domain 2 (GYPGIS) induce migration of protease-activated receptor 4 (PAR4) negative human adenocarcinoma gastric epithelial cells (AGS) upon expression of PAR4, albeit at considerably higher concentrations (300 μM) compared to those of TFF2 (200 nM).⁴⁸ Despite the high similarity in the trefoil domain among TFF peptides, some regions are clearly not conserved (Figure 3), suggesting that these sequences could mediate specific functions.

Figure 3.

Trefoil domain sequence alignment highlighting nonconserved residues (red). The trefoil domains display 70% similarity with 35% fully conserved residues. [*] identical residue, [:] strongly similar properties, [.] weakly similar properties.

2.1. TFF and Post-Translational Modifications

Little is known about the extent or importance of TFF post-translational modifications. Although glycosylated versions of hTFF2 have been observed,⁴⁹ a glycosylated version of TFF1 or TFF3 have not been identified. hTFF2 with an N-linked fucosylated N-diacetyllactosediamine (LacdiNAc) oligosaccharide at Asn15 was isolated from human gastric mucosa tissue (without neoplastic changes).⁴⁹ The functional significance of this glycosylation has not been determined, but it could regulate TFF2 half-life in serum similarly to other glycoprotein hormones.⁴⁹ Another hypothesis is that LacdiNAc mediates colonization of the human stomach by Helicobacter pylori (H. pylori), a bacterium found in the mucous layer that causes gastritis and duodenal ulcer disease in humans.⁵⁰ Complex glycosylations play a determinant role in binding H. pylori adhesins, and TFF2 levels are remarkably elevated in the serum of patients infected with the bacteria.^49,51,52 A similar post-translational modification at the Asn15 was also observed in recombinant TFF2 expressed in yeast. TFF2 expressed in this system yielded a large amount (34%) of hTFF2 glycosylated (mannose/N-acetyl glucosamine) heterogeneously via Asn15 N-linkage.⁵³ This glycosylated analogue was slightly more potent (1.3-fold) than the nonglycosylated hTFF2 when tested for protection against gastric damage in rats.³¹ hTFF1 and hTFF3, as well as porcine, rat, and murine TFF2, lack a consensus sequence for N-glycosylation (Asn-Xaa-Ser/Thr, where Xaa can be any amino acid except proline).⁵²

3. (Patho-)Physiological Roles of TFF in the Gastrointestinal Tract

3.1. TFF and Gastrointestinal Epithelia

The intestinal epithelium constitutes a major barrier protecting us against bacterial, immunogenic, or toxic luminal content.⁵⁴ Disruption of this barrier can favor abnormal antigen uptake and immune cell infiltration leading to uncontrolled inflammation and sometimes chronic gastrointestinal disorders, including IBD.⁵⁵⁻⁵⁷ Gastrointestinal mucosal defense is the most prominent known function of the TFF.^14,18,58 Accordingly, the three members of the TFF are mainly expressed in the gastrointestinal tract with distinct patterns of distribution.² TFF1 is predominantly produced by foveolar (surface mucus cells) and surface epithelial of the gastric mucosa.¹⁴⁻¹⁶ TFF2 is expressed by mucous neck cells, deep pyloric glands, and Brunner’s glands in the stomach and duodenum,¹⁶ while TFF3 is predominantly expressed by the mucus-secreting goblet cells in the small intestine and colon.¹⁷ Mucins are also expressed in a site-specific fashion, for example, MUC5AC is secreted in the stomach, MUC6 is secreted in the stomach and Brunner’s gland, and MUC2 is secreted in the intestine.⁵⁹ This might explain the association of each TFF peptide with a specific mucin in the different parts of the gastrointestinal tract (TFF1 with MUC5AC in stomach, TFF2 with MUC6 in stomach, and TFF3 with MUC2 in intestine).⁵⁹

A substantial body of evidence supports the specific roles of the TFF in gastrointestinal protection, with the presence of TFF1 and TFF3 in breast milk⁸ to play a role in the protection and development of the neonatal intestinal tract.⁸ TFF1 knockout (−/−) mice develop gastric carcinomas and adenomas,⁶⁰ whereas TFF2^–/– and TFF3^–/– mice are more susceptible to gastrointestinal inflammatory agents and less able to regenerate gastrointestinal epithelium following damage.^22,61

All TFF members promote epithelial repair by enhancing cell migration.⁵⁸ TFF1 and TFF3 also protects the epithelium by inhibiting apoptosis.⁶²⁻⁶⁵ Epithelial restitution, i.e., rapid repair of epithelial damage via cell migration toward the damaged area to cover the wound, is a complex process, and modulation of cell contact is a key step. TFF peptides modulate cell junctional complexes (structures that seal the paracellular space between the cells), contributing to the epithelial barrier function.^66,67 For example, TFF3 facilitates cell migration by disordering the E-cadherin/β-catenin complex⁶⁶ and by downregulating the expression⁶⁸ and increasing internalization of E-cadherin, a mediator of cell–cell contact.^66,69 TFF3 also modulates the expression of the tight junction proteins claudin-1 (increased) and claudin-2 (decreased), leading to decreased paracellular permeability.⁷⁰ Impairment of the mucosal epithelial barrier is a hallmark of IBD that comprises ulcerative colitis and Crohn’s disease.^56,71 Epithelial injury leads to dysregulation of the immune system, triggering an ongoing damage/inflammation in these disorders.^55,56,71 TFF levels increase upon epithelial injury, presumably to prevent further damage and disease progression.^{3,14,21,72−74} Interestingly, the TFF expression pattern can also change in response to injury, with TFF1 and TFF2 being upregulated in the intestine and TFF3 in the stomach.^59,74−78

3.2. TFF and H. pylori

TFF peptides also play a role in H. pylori infection.⁷⁹⁻⁸⁴H. pylori infection triggers an increase in TFF expression in human gastric cancer cell lines.⁸⁴ However, it is unknown if this upregulation is a mechanism induced by the bacteria to facilitate its adhesion to the gastric surface or a host response to protect the mucosa.⁸⁴H. pylori can interact with homodimeric TFF1,^35,79 enabling attachment and colonization of the gastric mucosa.⁷⁹ This molecular interaction was investigated by flow cytometry and surface plasmon resonance⁷⁹ and appears to be pH-dependent⁸¹ and specific, as it is disrupted by anti-TFF1 antibodies or preincubation with TFF1 homodimer.⁷⁹ Binding of H. pylori to homodimeric TFF1 explains the bacterial tropism for colonizing gastric tissue and colocalization with MUC5AC.⁷⁹ By contrast, TFF1 suppresses in vivo and in vitro H. pylori induced inflammation and carcinogenesis.⁸² TFF2 does not interact with H. pylori, but it also seems to play a role in H. pylori infection.⁸¹H. pylori infected TFF2^−/− mice, for example, have accelerated progression of gastritis to dysplasia,⁸³ and H. pylori positive patients have higher levels of TFF2.⁸⁵ These results imply that TFF2 exerts a protective role against H. pylori induced injury probably by decreasing inflammatory responses.^14,83,85 TFF3 also interacts with H. pylori but to a lesser extent than does TFF1.⁸¹

3.3. TFF Modification of Immune Responses and Processes

TFF2 and TFF3 contribute to mucosal integrity not only by promoting epithelial restitution but also by modulating the immune response.⁸⁶⁻⁹⁰ They act as signaling molecules from the epithelium to the immune system, regulating cytokine expression and immunocyte migration and proliferation.⁸⁶⁻⁸⁹ Pretreatment of colitic mice induced by dextran sulfate sodium (DSS) with TFF2 reduces expression of vascular cell adhesion molecule 1 (VCAM-1), an adhesion molecule that mediates leukocyte recruitment,⁹⁰ while TFF2 deficiency upregulates the expression of pro-inflammatory mediators and increases macrophages cytokine secretion, IL-1β-induced T-cell proliferation and basal nuclear NF-κB activation.^88,89 TFF2 also decreases cytokine production, lymphocyte proliferation, and IL-1R signaling, a pathway particularly important in chronic inflammatory diseases.⁸⁸ In in vitro assays, TFF3 down-regulates proinflammatory cytokines (IL-8 and IL-6) and stimulates expression of human β-defensins, molecules related to the first line of defense against microorganisms.⁸⁶ TFF3 induces nitric oxide production, an important promoter of mucosal integrity maintenance and regulator of gut inflammatory responses^89,91 and decay-accelerating factor (also known as DAF), a mucosal defensive protein that protects injured tissue against autologous complement damage.⁸⁷

3.4. TFF and Gastrointestinal Cancer

TFF peptides protect the gastrointestinal tract by regulating cell death, proliferation, migration, and angiogenesis.^92,93 Deregulation of these functions, however, might contribute to the development of TFF-mediated oncogenesis.^92,93 TFF1 is considered a gastric tumor suppressor by regulating differentiation of cells, while TFF2 and TFF3 can increase tumor progression by supporting cellular invasion and metastasis.^60,92−94 TFF1^−/− mice develop antropyloric adenoma with 30% progressing to carcinoma.⁶⁰ The self-renewal of the gastric mucosa from stem and precursor cells is also impaired in TFF1^−/− mice, leading to accumulation of undifferentiated and nonfunctional cells.⁹⁴ TFF2 is a major marker of spasmolytic polypeptide expressing metaplasia (SPEM) cells, which is associated with the development of gastric carcinoma.^95,96 This lineage develops during gastric repair and plays a major role in wound healing following severe gastric ulceration.^95,96 Continuous inflammation conditions might promote SPEM to proliferative preneoplastic metaplasia.⁹⁷ Overexpression of TFF3 in gastric cancer is correlated with a highly aggressive phenotype, lymph node metastasis, and poor prognosis.^93,97,98 This suggests that TFF3 plays a role in gastric cancer development, progression and dissemination.^93,97,98

4. TFF Mechanisms of Action

TFF peptides were initially thought to signal via a mechanism analogous to that of growth factors. However, unlike growth factors, TFF2 and TFF3 promote epithelial restitution through a transforming growth factor (TGF-β)-independent pathway.⁵⁸ Overall, the mechanism of action of the TFF peptides is not completely understood, and several potential binding proteins and cell surface receptors have been proposed as described below, although more detailed and independent validation and mechanistic studies are required (Table 1).

Table 1. Summary of TFF Interactions and Functions.

Ligand	Putatative or identified receptor/protein partnera	Function (hypothetical (H) or validated (V))	Refs
TFF1	TFIZ1 (GKN2)	synergistic antiproliferative and pro-apoptotic (V)	(6,30,125)
	MUC2/MUC5AC	mucosal protection (H)	(101,130)
	RF-LPS	involved in Helicobacter pylori colonization of the gastric tissue (V)	(81)
	FCGBP	binding of microorganisms (H)	(42)
TFF2	LINGO3	mucosal integrity and tissue repair (V)	(114)
	DMBT1	mucosal immunity and protection (H)	(127,130)
	β-integrin	cell migration (H)	(127,130)
	PAR4	cell migration (V)	(48)
	CXCR4	cell migration (V)	(107,112,137)
	MUC6	mucosal protection (H)	(132,133,139)
TFF3	LINGO2	protection against colitis and gastrointestinal helminths (V)	(113)
	MUC2	mucosal protection (H)	(103)
	PAR2	downregulation of IL-6/8 and upregulation of BD2/4 (V)	(105)
	CXCR4/CXCR7	cell migration (V)	(106)
	FCGBP	TFF3–FCGBP considered reservoir (H)	(119)
	DMBT1	mucosal immunity and protection (H)	(128)

^aGKN2: gastrokine 2; MUC: mucin; RF-LPS: oligosaccharide portion of lipopolysaccharide; BD: beta-defensin; CXCR: C-X-C chemokine receptor; PAR: protease-activated receptors; DMBT1: deleted in malignant brain tumors 1; FCGBP: IgG Fc binding protein.

4.1. TFF Interactions with Mucins

Mucins, a family of high-molecular-weight glycoproteins produced by epithelial cells, make up most of the mucus, where they bestow protective functions to the gut barrier. TFF peptides are proposed to interact with polysaccharide chains of mucins, thereby cross-linking mucins to increase mucosal gel viscosity and providing better protection to the underlying epithelium.^{13,58,99−101} The exact cross-linking mechanism, however, remains to be established.^100,102 The hypothesis that the TFF mechanism of action involves binding to mucins is furthermore supported by the observation that each TFF member colocalizes with a unique mucin-type in normal gastrointestinal mucosa, as well as in ulcer-associated cell lineages (TFF1 associates with MUC5AC, TFF2 associates with MUC6, and TFF3 associates with MUC2, respectively).⁵⁹ TFF2 and homodimeric, but not monomeric, TFF1 or TFF3 increase mucus gel viscosity and elasticity.^13,99 Combinations of mucins and TFF2 or TFF3 enhance protective effects¹⁰⁰ and restitution capacities⁵⁸ in in vitro models of epithelial injury. Yeast two-hybrid assay screening against a stomach and duodenum complementary DNA expression library revealed interactions between mouse TFF1 (mTFF1) and murine MUC2 (intestinal mucin) or MUC5AC (gastric mucin).¹⁰¹ TFF1 interacts with these two mucins through their cysteine-rich von Willebrand factor C domains 1 and 2 (WFC1 and VWFC2), which are involved in the polymerization of mucin monomers and therefore in mucus viscosity.¹⁰¹ TFF3 and MUC2 are bound in intestinal mucus in rats.¹⁰³

4.2. TFF Interactions with Membrane Receptors

Current consensus is that TFF peptides can also act via membrane receptors on the epithelial cells. Several biological activities mediated by TFF peptides, including antiapoptotic activity⁶⁵ and cell migration,⁵⁸ are generally regulated by receptor activation. Radioligand binding assays show the high-affinity binding of TFF3 to proteins on the surface of intestinal epithelial cell lines HT-29, Caco-2, and IEC-6.¹⁰⁴ It is important to note that these might not be the best models to study the TFF mechanism of action since HT-29 and Caco-2 are carcinoma epithelial cell lines and IEC-6 is a rodent nontransformed cell line.

TFF peptides also activate signaling pathways.²⁹ Multiple putative protein partners and receptors have been proposed, yet evidence of downstream signaling upon TFF binding has been demonstrated for only a few of them including chemokine receptor type 4 and 7 (CXCR4 and CXCR7) and PAR2 and PAR4.^48,105−107 Functions mediated by many of these interactions remain unknown. Studies with ¹²⁵I-labeled TFF suggest that these receptors are located at the basolateral domain of gastrointestinal cell lines.¹⁰⁸ Intravenously administered ¹²⁵I-TFF2 binds basolaterally to mucous neck cells and the pyloric glands in the stomach and Brunner’s glands and the Paneth cells in the small intestine and crypt cells in the colon.¹⁰⁹ Similar distributions are seen for intravenously administered ¹²⁵I-TFF1 or ¹²⁵I-TFF3.¹⁰⁸ Basolateral, but not apical, stimulation of the colonic epithelial cells T84 triggers TFF3-induced protective effects, further supporting the presence of a TFF receptor on the basolateral side.⁸⁷ The presence of such a receptor at the basolateral surface of epithelial cell implies that TFF peptides can only access this receptor during mucosal damage.⁸⁷ Mechanism of action via the apical side has also been suggested due to the synergic effects of TFF2/3 and mucin glycoproteins.¹¹⁰

4.2.1. TFF Interaction with CXCR4 and CXCR7

CXCR4 is a receptor related to cell migration, recruitment of immune cells to inflamed tissues, and neuronal activity.¹¹¹ TFF2 triggers downstream signaling (activation of mitogen-activated protein or AKT kinases) in AGS cells engineered to express CXCR4, but not in parental cells.¹⁰⁷ Neutralization of CXCR4 dampens the intracellular signaling (Ca²⁺ flux and AKT kinase phosphorylation) induced by TFF2 in Jurkat cells, an immortalized line of T lymphocyte cells.¹⁰⁷ Similarly, ERK1/2 activation upon TFF2 stimulation was abrogated after treatment with AMD3100, a CXCR4 antagonist.¹⁰⁷ The concentration of TFF2 (500–600 nM) required to achieve maximum Ca²⁺ response upon binding to CXCR4 is ∼50 times higher than that of SDF-1-α (stroma cell derived factor 1-α; 12.5 nM), a well-established CXCR4 agonist.¹⁰⁷ Competitive interactions between TFF2 and SDF-1-α (CXCL12) might inhibit the inflammatory cell recruitment and survival induced by SDF-1-α, thus decreasing inflammation in the gastrointestinal mucosa.¹⁰⁷ In line with this hypothesis, TFF2 reduces the expression of adhesion molecules and leukocyte recruitment to the inflamed intestine in a mouse model of intestinal inflammation.⁹⁰ In addition, endogenous or exogenous TFF2 increases the cell migration of lymphoblastic jurkat cells and competitively attenuates SDF-1α-mediated migration.¹⁰⁷ TFF2–CXCR4’s role in epithelial repair was also verified using a gastric organoid system, which is an in vitro model that closely reflects the native tissue.¹¹² This organoid model also demonstrated that TFF2 promotes Ca²⁺ mobilization via CXCR4.¹¹² CXCR4 is not involved in TFF2-enhanced cell migration in HT-29 cells that endogenously express CXCR4, since the CXCR4 antagonist AMD3100 abolished the SDF-1-α-induced migration while not affecting TFF2-induced migration.⁴⁸ However, HT-29 is a colonic cell line and TFF2 is not naturally present in the colon.

TFF3 interaction with CXCR4 and CXCR7 expressed on ocular surface tissues has also been described.¹⁰⁶ This interaction mediates cell migration which involves an ERK1/2-independent signaling pathway.¹⁰⁶ Cell migration induced by TFF3 is dependent on both receptors, since neutralizing either one of the receptors abolishes the migratory effects. By contrast, TFF3’s cell proliferation effect is independent of CXCR4 and CXCR7.¹⁰⁶ Therefore, it seems that TFF3 mediates a complex cellular response by activating multiple receptors.¹⁰⁶

4.2.2. TFF Interaction with LINGO

TFF2 and TFF3 have been described as natural ligands for the leucine-rich repeat and immunoglobulin-like domain-containing Nogo receptor-interacting protein (LINGO).^113,114 The LINGO family is better described in the nervous system where they play a role in axonal regeneration, neuronal survivor, oligodendrocyte differentiation, and myelination.^113,114 Thus, LINGO has gained considerable interest as a new target for neurodegenerative diseases.¹¹⁵ LINGO also exists in the gastrointestinal tract, but its function there is poorly characterized. LINGO2 is highly expressed in patients with advanced gastric cancer and involved in cell motility, tumorigenic ability, and angiogenesis.¹¹⁶ LINGO activation usually occurs via homotypic or heterotypic interaction with other membrane proteins such as the Nogo-A receptor and p75 neurotrophin receptor interaction with LINGO1 and EGF interaction with LINGO3.^114,115

TFF2 interaction with LINGO3 triggers ERK signaling, mediating tissue repair at the mucosal interface.¹¹⁴In vitro this interaction is supported by immunoprecipitation and colocalization studies in human rectosigmoid and respiratory tissues. In addition, LINGO3^–/– mice display a phenotype that resembles TFF2 deficiency, such as impairment of mucosal regeneration and accumulation of immune cells secreting inflammatory cytokines even without stimuli.¹¹⁴ TFF3 interaction with LINGO2 disrupts LINGO2/EGFR interactions, enhancing activation of the EGFR pathway; thus, this framework mediates epithelial repair. A LINGO2–TFF3 interaction is supported by immunoprecipitation and colocalization studies at the IEC cell surface.¹¹³

4.2.3. TFF Interaction with PAR

TFF2 and TFF3 activate PARs, a family of four receptors (PAR1–4) belonging to the G protein-coupled receptor (GPCR) family associated with several inflammatory diseases, including gastrointestinal disorders.^117,118 PAR activation by TFF peptides is interesting since these receptors, unlike other GPCRs, are not directly activated by endogenous extracellular agonists.¹¹⁷ Activation occurs via proteolytic enzymes that cleave the extracellular N-terminus to generate a new N-terminus that folds back and activates the receptor.¹¹⁷ TFF2 (200 nM) induces migration of HT-29 cells; however, this effect is inhibited by the knockdown of PAR4 with small interfering RNA.⁴⁸ This suggests that PAR4 is involved in the TFF2 mucosal healing effect. PAR1, when stimulated with a PAR1 agonist (PAR1-AP), abrogates Bombina maxima TFF2 (Bm-TFF2; isolated from frog skin secretion), induced human platelet aggregation, and Ca²⁺ mobilization.⁴⁸ These data suggest that Bm-TFF2 induces platelet activation by activation of PAR1.⁴⁸

TFF3 activates PAR2, but not PAR1, inducing cytosolic Ca²⁺ activity in HT-29 colonic cells.¹⁰⁵ Western blotting and coimmunoprecipitation further support the PAR2–TFF3 interaction.¹⁰⁵ Intracellular Ca²⁺ mobilization measurements demonstrate that TFF3 (∼1.5 μM) activates PAR2 in HT-29 intestinal epithelial cells, causing downregulation of proinflammatory cytokines and upregulation of defensin expression (hBD2 and hBD4).¹⁰⁵ TFF-PAR interactions have been reproduced by different groups.^105,117 However, further characterization of PAR–TFF interactions needs to be performed to truly validate PAR as a TFF target. This includes characterization via radioligand displacement assays, second messenger signaling in PAR overexpressed cells blocked with PAR antagonists or in vivo assays.

4.3. TFF Interaction with FCGBP

TFF3 has been described to be mostly connected to the IgG Fc binding protein via a disulfide bond (FCGBP) in the intestine¹¹⁹ as well as in saliva.¹²⁰ TFF3–FCGBP complex was purified from human colonic tissue and saliva by gel filtration and FPLC and analyzed by mass spectrometry.^119,120 FCGBP is a constituent of the mucosal immunological defense system which inhibits complement-mediated reactions in the intestinal wall and regulates pathogen attachment and disease progression in mucosal surfaces.^120,121In vitro assays demonstrated that hydrogen sulfide, an important gas promoting injury repair and resolution of inflammation,¹²² reduces the disulfide bond connecting the TFF3–FCGBP complex, thereby slowly releasing TFF3. This suggests that this gas can control the intestinal levels of TFF3 monomer.¹¹⁹ Therefore, the TFF3–FCGBP heterodimer is thought to be a TFF3 reservoir, although no direct evidence supporting this hypothesis has been shown. In addition, the purified TFF3–FCGBP complex lacks motogenic activity.¹¹⁹ Heterodimerization of FCGBP and TFF3 might also have synergic antimicrobial effect since TFF3 as a lectin could bind to microbial glycans.¹²⁰

4.4. TFF Interaction with Gastrokine

TFIZ1 (trefoil factor interactions 1; also called gastrokine 2), a protein from the gastrokine family, is another TFF1-interacting protein.^13,30,123 Gastrokines are expressed in healthy gastric surfaces, and although their functions in the stomach are not completely understood, they seem to play a role in both gastric epithelium homeostasis and tumor suppression.^{13,30,123,124} The TFF1–TFIZ1 heterodimer was identified by immunoprecipitation approaches and accounts for the predominant form of TFF1 in the stomach.^13,30,123 TFF1–TFIZ1 seems to regulate the biological function of TFF1, since the relative abundance of TFIZ1 controls the unique molecular form of TFF1 produced.¹²³ For example, reduced expression of TFIZ1 leads to increased formation of TFF1 homodimers, which induce migration of tumor cells.¹²³ In addition, the interaction between TFF1 and TFIZ1 has synergistic antiproliferative and proapoptotic effects in gastric cancer; therefore, the interaction between these molecules represents a mechanism of cancer suppression.¹²⁵

Blottin, the murine orthologue of TFIZ1, was identified as a mTFF2-binding protein using alkaline-phosphatase-labeled mTFF2 to probe interacting proteins in 2D blots.¹²⁶ However, these results contradict a second study, which demonstrated that TFIZ1 interacts specifically with TFF1, but not TFF2 or TFF3 in human gastric cells.¹²³ Both TFIZ1 and TFF1, but not TFF2, are expressed in the same cell types. The interaction between blottin and TFF2 would, therefore, depend on the paracrine secretion of this binding protein to the mucus where TFF2 is found.^123,126

4.5. TFF Interaction with DMBT1

DMBT1 (deleted in malignant brain tumours 1 protein, also called CRF-Ductin) and fibronectin, present in mucosal membrane homogenate from porcine stomach scrapings, were purified by gel affinity using pTFF2 as a bait and identified by mass spectrometry sequencing after tryptic digestion.¹²⁷ DMBT1 is involved in cell differentiation and a role in tumor suppression and innate immune defense has also been suggested.^128,129 DMBT1 is upregulated in inflamed IBD tissues and associated with regenerative functions in the gastrointestinal tract.^128,129 A solid-phase ELISA assay showed binding of homodimeric (but not monomeric) hTFF3, hTFF2, and glycosylated human TFF2 to a purified variant of human DMBT1^gp340 (glycoprotein 340) in a Ca²⁺-dependent manner.¹²⁸ It remains an open question if DMBT1 from different tissues can interact with distinct TFF peptides.¹²⁸ The roles of TFF–DMBT1 heterodimer have not been elucidated; however, it has been hypothesized that this heterodimer is involved in microbial defense.¹²⁰

4.6. TFF Interaction with Integrins

Integrins are involved in cell adhesion and recognition, being key players in cell migration processes. Porcine pancreatic TFF2 binds noncovalently to integrin β1.^127,130 This interaction was found by TFF2 affinity chromatography. Elution of integrin β1 using a buffer containing a high concentration of TFF2 confirmed that binding was noncovalent;¹²⁷ however, this interaction has not been further characterized.

4.7. TFF Interaction with Carbohydrates

Interactions with carbohydrate moieties (lectin-like activities) have also been described for the TFFs.^52,131,132 Homodimeric TFF1 interaction with the core oligosaccharide of H. pylori low-molecular-mass lipopolysaccharide appears to be involved in gastric mucus colonization by these bacteria.¹³¹ TFF2 interacts with the GlcNAcα1 → 4Galβ1 → R moiety present in MUC6. This interaction is modulated by Ca²⁺ and might affect the mucus viscoelastic properties.^132,133 In addition, it is speculated that TFF peptides could bind to the potential receptors previously described via their carbohydrate moieties, explaining the low affinity binding.⁹²

4.8. Unidentified TFF Protein Partners

Rat TFF3 (rTFF3) forms a complex of ∼45 kDa in membrane protein preparations of breast (MCF-7), colonic (colony-29), and rat gastrointestinal cells as shown by immunoprecipitation.¹³⁴ Analysis of this complex by immunoblotting suggested that the protein partner under reducing conditions has a molecular weight of ∼28 kDa and is connected via disulfide bond to rTFF3. Binding of hTFF2 to protein partners was also investigated and yielded a complex of identical size; however, TFF2 and TFF3 binding proteins were not further characterized.¹³⁴ Interestingly, densitometry analysis revealed that the rTFF3-binding partner complex is more abundant in colonic cells (colony-29) wounded by scraping a pipet tip across a confluent cellular layer than it is in intact cells.¹³⁴ This result is in line with findings that binding of parenterally administrated ¹²⁵I-mTFF3 is higher in colitic than in healthy mice.⁷³ Likewise, a putative receptor in the small intestine of rats was described.¹³⁵ Using recombinant, biotinylated rTFF3 fusion protein (biotin-TFF3) in a ligand blot assay, the authors identified a ∼50 kDa glycosylated protein in the solubilized membrane homogenate from small intestinal tissue of rat that was absent in cytosolic fractions.¹³⁵ Despite the molecular weight divergence between the binding proteins identified by these two studies,^134,135 it is probable that both studies identified the same protein and that the shift in mass derives from proteolytic degradation. Given that protein sequences were not identified in these studies, the precise nature and significance of the interactions between this putative receptor and the TFF remain unknown.

4.9. Concluding Remarks about TFF Protein Partners

The numerous biological functions described for TFF suggest that these peptides signal through different receptors, which aligns with the description of several interaction partners (Table 1). However, most of the interactions between the TFF and these protein partners/receptors have not been reproduced by independent work, and more studies are needed to confirm them as TFF targets and to understand how the interactions modify TFF function. LINGO is the most convincing target family described so far since in vivo as well as in vitro studies underpin this protein as a bona fide receptor for TFF2 and TFF3.^113,114 At least four research groups have independently demonstrated that TFF2 or TFF3 interaction with CXCR triggers an intracellular signal¹⁰⁷ or a biological response such as proliferation or cell migration.^106,136,137 Despite these pieces of evidence supporting TFF peptides signaling through CXCR4 and TFF3-RFP/CXCR4-GFP, double transfectants failed to demonstrate colocalization of TFF3-RFP with CXCR4, although CXCR4 canonical ligand SDF1 tagged with RFP clearly colocalizes with CXCR4-GFP.¹¹³ In addition, another limitation of the CXCR4–TFF3 axis is the lack of evidence that it is promoted in vivo. Further studies are therefore required to confirm this interaction, clarify its physiological significance, and to determine whether TFF1 is also a CXCR4 ligand. TFF’s interaction with PARs, including TFF3–PAR2¹⁰⁵ and TFF2–PAR4,⁴⁸ has been demonstrated by two independent groups so far. TFF3 mediates an anti-inflammatory response via PAR2 activation, downregulating the levels of IL-6 and IL-8. An unexpected result since activation of PAR usually promotes inflammation.¹³⁸ Altogether, further investigation is required to understand the molecular mechanism of TFF and PARs in health and disease.

5. Therapeutic Potential and Preclinical Studies

The TFF has been suggested to hold great therapeutic potential for gastrointestinal disorders since they target processes that reduce epithelial permeability¹⁴⁰ and regulate inflammation, thus maintaining epithelial homeostasis.^{21,86,141,142} Increased gastrointestinal epithelial permeability is observed in many gastrointestinal disorders, such as IBD, irritable bowel syndrome (IBS), and celiac disease.⁵⁶ In these disorders, dysregulation of the immune system is linked to epithelial barrier damage and leads to ongoing damage/inflammation.^55,56,71 Although the etiology of these disorders is not completely understood, a growing body of evidence supports gastrointestinal wound healing as a promising therapeutic strategy.^143−145 Aside from gastrointestinal disorders, the TFF could hold therapeutic potential for other diseases (Table 2), particularly if they are related to mucosal injuries, including corneal wounds,¹⁴⁶ dry eye disease,¹⁴⁷ and asthma.¹⁴⁸

Table 2. Pathological Disorders Characterized by Altered TFF Expression and Potential Therapeutic Effect.

Location	Pathologic disorder	Therapeutic effect	Refs
Gastrointestinal tract	Inflammatory bowel diseases, intestinal and gastric damage, oral mucositis, diabetes	Gastrointestinal epithelium repair and protection, β-cells proliferation	(77,78,99,140,141,149−158)
Ocular system	Dry eye disease, corneal wounds	Ocular surface damage repair	(146,159)
Urinary tract	Chronic kidney disease, nephrolithiasis	Kidney damage repair and calcium oxalate crystal growth inhibitor	(1,160−163)
Respiratory tract	Asthma, nasal allergic mucosa, mustard lung, chronic obstructive pulmonary disease, chronic rhinosinusitis, bronchopathy	Airway remodelling limitation, mucus viscosity modulation, immune regulation, lung protection and repair.	(164−174)
Nervous system	Depression	Antidepressant	(175,176)

The most convincing evidence for the TFF’s physiological function and therapeutic benefit derive from in vivo studies (Table 3). The therapeutic potential of all three TFF peptides has been validated in several in vivo models, including those induced by DSS,^20,90,158 dinitrobenzenesulfonic acid (DNBS),²¹ alcohol,²³ chemotherapy,¹⁷⁷ radiation,^156,177 and nonsteroid anti-inflammatory drugs (NSAIDs).^23,31,178 These studies demonstrated beneficial effects on mucosal protection and restitution following administration by either parenteral (intravenous, subcutaneous, or intraperitoneal) or luminal (oral, intracolonic, intragastric, and rectal) routes (Table 3).

Table 3. Therapeutic Effects of TFF Peptides in Animal Models of Gastrointestinal Injury.

Molecule	Model	Administration	Effect and notes	Refs
TFF1	DSS-induced colitis	Systemic	Homodimer effective and acts in synergy in with EGF; monomer ineffective	(158)
	DSS-induced colitis or IL-10–/– mouse	Intragastric administration of TFF secreting L. lactis; rectal or oral administration	Improved chronic colitis in IL-10–/– mouse; prevention and healing of chronic DSS-induced colitis; In situ synthesis by L. lactis is more effective than rectally and orally administered.	(20)
	Radiation-induced oral mucositis	Mouth rinse formulation of TFF1-secreting L. lactis	Safe and efficient for prevention and treatment of oral mucositis	(156)
TFF2	Indomethacin-induced gastric damage	Subcutaneous or oral	Cytoprotection subcutaneously administrated but ineffective via oral administration	(31)
	Aspirin-induced gastric injury	Intragastric	Prevents damage but fails to stimulate early restitution	(178)
	Ethanol- or indomethacin-induced gastric injury	Gastric lavage or intraperitoneal	Protection via oral but not intraperitoneal	(23)
	DSS-induced colitis	Subcutaneous or intracolonic	Ameliorated the clinical course of colitis in a preventive mode but not for treatment; topical administration superior to systemic administration	(90)
	DNBS-induced colitis	Intrarectal	Enhancement of colonic epithelial repair and reduction of inflammation	(21)
	DSS- or mitomycin-induced colitis	Intracolonic or subcutaneous	Luminal positive effect only on DSS-induced colitis; injected aggravated colitis in both models	(44)
	DSS-induced colitis or IL-10–/– mouse	Intragastric administration of TFF-secreting L. lactis	Improved chronic colitis in IL-10–/– mouse; effective in the prevention and healing of chronic DSS-induced colitis	(20)
	Indomethacin-induced gastric damage or mercaptamine-induced duodenal ulcers	Subcutaneous infusion or oral	Accelerated gastric ulcer healing but aggravated duodenal ulcer	(179)
TFF3	Acetic acid-induced colitis	Rectal instillation	Enhanced epithelial migration and reconstitution of normal healing	(22)
	Chemotherapy- or radiation-induced intestinal mucositis	Oral gavage	Improved mucolitis and mucosal permeability; reduced ongoing apoptosis	(177)
	Ethanol- or indomethacin-induced gastric injury	Gastric lavage or intraperitonial	Preventive	(23)
	DSS-induced colitis	Systemic	Moderate effect attenuating the disease	(73)
	DSS- or mitomycin-induced colitis	Intracolonic or subcutaneous	Injected aggravated colitis; luminal administration of dimer improved both models; monomer ineffective in both models	(44)
	DSS-induced colitis and IL-10–/– mouse	Intragastric administration of TFF-secreting L. lactis	Improved chronic colitis in IL-10–/– mouse; effective in the prevention and healing of chronic DSS-induced colitis	(20)
	Ethanol-induced gastric mucosal damage	Gastric gavage	Prevented/ameliorated gastric mucosal injury	(40)
	Burn-induced intestinal mucosa injury	Gene therapy	Alleviated and accelerated reconstruction of the damaged mucosa	(180)

All of the TFF peptides have beneficial effects when tested against rodent DSS-induced colitis, a well-established model resembling ulcerative colitis.⁹⁰ Homodimeric, but not monomeric, TFF1 reduces epithelial damage when subcutaneously administered before induction of colitis.¹⁵⁸ TFF2 has a preventive effect when subcutaneously or intracolonically administered, with luminal administration being superior to the systemic route.⁹⁰ Contradictory results have been described regarding the beneficial effect of TFF2 on existing DSS-induced injury. One study found that intracolonically administered TFF2 did not modify disease severity,⁹⁰ while another study observed a beneficial effect.⁴⁴ Intracolonic administration of TFF3 is also effective in treating DSS-induced colitis in rats and is more effective than TFF2 administrated via the same route.⁴⁴ Parenteral administration of TFF3 also yielded contradictory results. In one study, subcutaneous administration of TFF3 aggravated the disease to a minor degree,⁴⁴ whereas another study demonstrated that systemic administration (intraperitoneal and subcutaneous) had a positive effect.⁷³ These discrepancies remain to be clarified. Although these studies did not compare administration routes in detail, they do suggest that luminal administration is superior to systemic administration.

TFF2 and TFF3 have also been tested in other models of colitis, such as DNBS/ethanol, mitomycin C, and acetic acid induced injury. TFF2 delivered intrarectally during the active phase of the disease accelerates healing and reduces inflammation indexes in rats with DNBS/ethanol-induced colitis, a model that mimics the damage and repair profiles of Crohn’s disease in humans.²¹ Luminal or systemic administration of TFF2 has beneficial effects in the treatment of mitomycin C induced colitis in rats.⁴⁴ Interestingly, mitomycin C induces colitis following intraperitoneal injection, producing an inflammatory response that is primarily localized in the mucosa, leaving the surface epithelial barrier almost intact.⁴⁴ This is distinct from other models that induce colitis from the luminal side. The disease activity in the mitomycin C model is also aggravated upon subcutaneous injection of the TFF3 monomer, which reduces the extent of the disease when administered luminally.⁴⁴ In acetic acid induced mucosal injury, rectal instillation of TFF3 restores restitution in TFF3-deficient mice.²²

TFF1/2 are also effective in ameliorating gastric injury in models where the damage has been induced by NSAIDs. These drugs inhibit the cyclooxygenase enzyme leading to a deficiency in prostaglandins, molecules that mediate inflammatory responses.¹⁸¹ Subcutaneous infusion of TFF1 has preventive effects against indomethacin-induced gastric damage in rats. Monomeric TFF1 significantly reduces the injury only at the highest dose, reducing the damage by about 30%, while its homodimeric counterpart produces an even greater reduction in damage of around 70%.⁴³ Luminal and systemic administration of TFF2 accelerates ulcer healing in indomethacin-induced gastric damage in rats but aggravates mercaptamine-induced duodenal ulcers.¹⁷⁹ Studies comparing luminal versus systemic administration of TFF2 in the prevention of gastric mucosal damage are contradictory.^23,31 TFF2 subcutaneously but not orally administrated prevents gastric damage elicited by an indomethacin/restraint rat model.³¹ Conversely, another group demonstrated that oral TFF2 prevents ethanol- and indomethacin-induced gastric injury in rats, while intraperitoneal TFF2 injection had no protective effect in ethanol-induced gastric injury models and very little effect in indomethacin models.²³ Disagreements between these studies could be explained by the different methods used, such as the methodology used to induce the gastric damage and different routes of systemic administration.^23,31 In an aspirin-induced gastric injury model in rats, luminal TFF2 prevented damage but failed to stimulate epithelial restitution.¹⁷⁹ Finally, intragastrically delivered TFF3 prevents ethanol- and indomethacin-induced gastric injury in rats.^23,40

Combined, these results from animal studies are highly promising and underpin TFF’s therapeutic potential for the treatment of gastrointestinal disorders. An improved understanding of their fundamental mechanisms of action and target proteins are however required to accelerate this translation.

6. Considerations Regarding Route of Administration

The optimal route of delivery is still under debate, and it might depend on whether the use is geared toward protection or repair as well as the site of action. Intravenously administered TFF2 and TFF3 monomer and homodimer distribute to peripheral tissues and clear from circulation within 2–3 hours.²⁵ Also, systemically administered TFF2 and TFF3 homodimer in healthy mice or rats are taken up by the gastrointestinal tract and secreted into the lumen.¹⁸² These studies suggest the presence of a basolateral receptor in epithelial cells and that intravenously injected TFF peptides reach the inner part of the mucus layer similarly to endogenously expressed TFF peptides.^108,109,182 Of note, porcine TFF2 intravenously injected in rats was not degraded by the kidney or liver and was excreted almost unaltered in the urine.¹⁰⁹

An oral route of administration might not be suitable for the treatment of the colon, since TFF peptides do not reach this site,^25,179 even though they are regarded as highly stable against proteases and low pH.³¹ Potential reasons for not reaching the colon could be interactions with mucins or microbial degradation during transit through the gastrointestinal tract.²⁵ Although rectal administration is effective in promoting epithelial repair and protection in animal models,²⁰⁻²² it is not the preferred administration method in humans. In mice, this limitation was elegantly addressed by using TFF(1, 2, or 3)-secreting gut bacteria Lactococcus lactis (L. lactis; LL-TFF).²⁰ This approach displayed prophylactic and therapeutic effects in acute and chronic colitis in mice.²⁰ Also, intragastrically administrated LL-mTFF1 was more effective in the prevention or treatment of DSS-induced colitis than rectally administrated purified TFF1 to the intestinal tract. Of note, the amount of rectally administered mTFF1 was 1200-fold higher than the amount of mTFF1 secreted by the bacteria.²⁰

7. Clinical Trials

Animal studies demonstrated variation in the efficacy of TFF peptide treatment depending on the damage model and administration route, yet this has not prevented their investigation in a number of clinical trials. In a phase I/II double-blind randomized placebo-controlled study, enemas containing recombinant hTFF3 were given in combination with orally delivered mesalazine (5-aminosalicylic acid; 5-ASA, a well-established agent for colitis therapy) to 16 patients with mild-to-moderate left-sided ulcerative colitis.²⁴ All patients (TFF3 and placebo group) received an active treatment (mesalazine). TFF3 enemas (10 mg/mL in 75 mL of 0.9% saline for 14 days) were well-tolerated, and no adverse effects were described.²⁴ However, no additional improvement was observed when compared to 5-ASA treatment alone, and the study was not progressed to phases II/III.²⁴ While these results were disappointing, it should be considered that the number of patients was small, and the weak effects may have been a result of a suboptimal administration route. Thus, more promising results may be observed in differently designed clinical trials.^24,25

An oral rinse formulation containing L. lactis secreting TFF1 (AG013) was safe and efficacious for the treatment of oral mucositis in an animal model.¹⁵⁶ On the basis of these preclinical results, AG013 was moved forward to phase 1b to assess safety and tolerability in head and neck cancer patients receiving induction chemotherapy cancer.²⁶ A preliminary efficacy analysis was performed in 19 subjects (14 patients received AG013 treatment and 5 received a placebo) divided into three dosing schedules (one, three, and six times daily) receiving a total dose of either 2.0 × 10¹¹, 6.0 × 10¹¹, or 1.2 × 10¹² colony-forming units per day.²⁶ Once and three times daily doses were well-tolerated and had high compliance rates, whereas six times daily doses impaired tolerability and compromised compliance. The thrice-daily dosing of AG013 was safe, well-tolerated, and had promising efficacy in reducing ulcerative oral mucositis in this preliminary test.²⁶ AG013 has been granted Orphan Drug status in the European Union and received the Fast Track designation by the USA Food and Drug Administration, which will facilitate further development of AG013 by the company Oragenics for the treatment of oral mucositis, a currently unmet clinical need.

Further promising results were obtained in a phase II trial testing the safety and efficacy of an oral spray formulation of homodimeric hTFF3 in preventing oral mucositis in colorectal cancer patients receiving fluorouracil-based chemotherapy.²⁷ The patients were divided into three groups (33 subjects/group) receiving 10 mg/mL TFF3, 80 mg/mL TFF3, or placebo via an oral spray. This topically administered formulation was well-tolerated and reduced the incidence and severity of debilitating chemotherapy-induced oral mucositis.

While the therapeutic potential of TFF for mucositis looks highly promising, more research is needed toward an effective TFF delivery into the gut for the treatment of gastrointestinal disorders. It would certainly be interesting to investigate L. lactis secreting TFF in humans considering its promising results in animal models.²⁰ In addition, strategies to enhance delivery into the gastrointestinal tract could be applied to the TFF peptides, such as controlled release, prodrugs, pH-specific coating, matrices, and nanocarriers.¹⁸³

8. Conclusions

Trefoil factor family peptides comprise an intriguing class of peptides that has received substantial interest from both academia and industry due to their remarkable stability, low toxicity, fundamental role in gut homeostasis, and promising therapeutic potential for gastrointestinal and other mucosal disorders. Despite these advantages, several gaps remain regarding their mechanisms of action, target proteins/receptors, and (patho)physiological significance that hampers therapeutic development and clinical trial design. Identification and validation of TFF targets, establishment of robust bioassays to evaluate TFF activities and systematic structure–activity relationship studies are required to advance the field and improve the development of efficient therapeutic options. Directed efforts in TFF probe development and proteomic approaches toward target identification will prove valuable in this regard. TFF’s optimal route of administration remains unclear, and more systematic and disease-specific pharmacokinetic studies are required. Taken together, TFF offers exciting fundamental as well as translational research opportunities with promising therapeutic potential for gastrointestinal and mucosal disorders.

The authors declare no competing financial interest.