Reinforcing the mucus: a new therapeutic approach for ulcerative colitis?

The treatment of active mucosal inflammation in ulcerative colitis remains challenging. Current therapies have limited efficacy and may be associated with clinically significant adverse effects. There is room for new therapeutic approaches. While nearly all of our current pharmacological approaches involve attacking various immune and inflammatory pathways in order to facilitate healing, few in the clinician’s current arsenal are directed towards enhancing or protecting the colonic epithelial barrier. Is this situation about to change?

The understanding of the pathogenesis of ulcerative colitis has not considerably progressed over the last decade. The proponents of concepts that the primary abnormalities lie within immune and inflammatory mechanisms have stumbled in attempts to explain the striking features of ulcerative colitis, such as the diffuse nature of the inflammation, its confinement to the mucosal compartment, and its distribution in the large bowel. The alternative concept that primary abnormalities lie within an abnormal epithelial barrier sits more comfortably with these characteristic features of the disease.¹ The barrier has regional differences in structure, composition, and function, which offer simple explanations for, for example, disease distribution. The nature of the inflammatory response in ulcerative colitis—intense polymorph infiltration and predominant antibody mediated (T_H2) responses with less prominent T cell activation—is most consistent with exposure of the immune system to large numbers of different “foreign” molecules.² Such events might be anticipated in a situation where multiple molecules are able to pass through a deficient epithelial barrier. This is in marked contrast with the situation in Crohn’s disease where the patchy inflammation involving deeper layers of the intestinal wall and draining lymph nodes, together with dominant T cell activation and T_H1-type cytokine profile that characterises the response to a limited number of antigens specifically taken up and presented to T cells via follicle associated epithelial cells.^2,3

Studies that date back more than 20 years have demonstrated that the colonic epithelium is abnormal in structure and function in patients with ulcerative colitis. The epithelium comprises cells that are metabolically abnormal (such as deficient β-oxidation⁴ or sulphation of phenols⁵), respond abnormally to stress (as shown by the response in vitro after its separation from the basement membrane^6,7), and have an abnormal cell membrane (such as abnormal permeability⁸). The mucus layer is abnormal, both in its thickness⁹ and composition (such as abnormal glycosylation of mucins^10–12 and abnormalities of the phospholipid component¹³). Many of these abnormalities are independent of the presence of mucosal inflammation, although whether they are primary abnormalities or secondary to other processes has never been definitively demonstrated. Why such abnormalities are present—whether autoimmune injury to the epithelium is occurring, whether there might be a genetic basis for epithelial structure or function, whether there are luminal factors that might induce abnormal behaviour of the epithelium, or a combination of any or all of these—has been the basis of much study and speculation without definitive answers.

Targeting the epithelial barrier in order to reduce the stimulus to inflammatory events, to enhance healing of actively inflamed mucosa, and to prevent relapse have also been the subject of much speculation and study. Approaches have ranged from improving the regenerative ability of the epithelium (such as the use of growth factors^14,15), to trying to enhance energy substrate supply (using butyrate enemas¹⁶), and to enhancing the mucous barrier (with trefoil peptides^15,17 or inhibition of bacterial sulphatases with bismuth¹⁸). None of these approaches has yet to be promoted from potential therapy to use in regular practice because, for example, the theoretical basis was misguided, efficacy was limited, or further development was hindered by funding difficulties and commercial realities. Will further attention to colonic mucus change this situation?

COLONIC MUCUS

The normal colon is lined by a layer of mucus that is more than 100 μm thick.⁹ The mucus serves essential functions. It is a lubricant ensuring low friction between moving structures (luminal contents) and the epithelium. This conceptually simple but essential protective function is predominantly the responsibility of surface acting phospholipids.¹⁹ They form an oligolamellar lining that converts the hydrophilic epithelial surface into a hydrophobic one that interfaces with luminal contents. The tight packing together of fatty acid chains provides a good basis for a hydrophobic barrier. Indeed, the regions of the gastrointestinal tract with the most developed hydrophobicity are the stomach and colon, where the potential for injurious insults from luminal factors are the greatest.²⁰ Most attention in recent years has been paid to other mucous components that subserve complementary protective functions. Mucins are glycoproteins that function to exclude large molecules (polymers with a molecular weight >20 000) that are not glycoproteins, and to trap other molecules either non-specifically via general “stickiness” or more specifically via carbohydrate structures (lectin binding sites) that may be similar to those on the cell surface.²¹ Such trapped matter can be discarded by the constant removal of mucus. It is no surprise then that mucus secretion is increased in the face of threats and this may be mediated, at least in part, by immune events.²² The mucus environment also supports the presence of other protective proteins and peptides such as secretory IgA,²³ lactoferrin,²⁴ and trefoil peptides.²⁵

Of all of these components, it has not been clear what is or are the most important from functional, pathogenic, or treatment points of view. Much of the glamour and excitement have surrounded the mucin glycoproteins and other secreted proteins within the mucus. However, recent evidence suggests that the phospholipid component might be a critical factor that can be readily modulated when the mucous barrier is failing.

PHOSPHOLIPIDS

Phospholipids, the major lipid components of mucus, are amphiphilic molecules and contain a polar head group and non-polar hydrocarbon (fatty acid) tails. The major classes of phospholipids include phosphatidylcholine (PC), phosphatidylethanolamine, phosphatidylinostiol, and phosphatidylserine. In colonic mucus, PC and lysophosphatidylcholine (LPC) are the major species.¹³ LPC is an intermediate in the metabolism of PC but is also produced after the hydrolysis of PC by phospholipase A2.²⁶ Orientation of the lipophilic region of the phospholipid and the nature of the fatty acids characterise the hydrophobicity of the mucus gel layer.²⁰ The fatty acid tails extend into the lumen to form a “non-wettable” resistant layer.^20,27 They also extend from the mucosal cell side of the mucus gel.²⁷ In mucus, the PC species typically contain one saturated (palmitic acid 16:0 or stearic acid 18:0) and one unsaturated (oleic acid 18:1or linoleic acid 18:2) fatty acid with PC (that is, PC 16:0/18:1 and PC 18:0/18:2).¹³ This contrasts with the PC of pulmonary surfactant, dipalmitoylphosphatidylcholine, which contains two saturated fatty acids, palmitic acid (PC 16:0/16:0).¹⁹

The origin of PC in mucus has not been established and more research is required to understand how, when, and where these surface active phospholipids are synthesised, stored, and secreted. In animal studies, there is some evidence that PC is primarily secreted by the jejunum and ileum, suggesting that PC is delivered to the mucus via the lumen.¹³ In these studies, the contribution of PC produced via the colonic epithelium appeared to be minimal. It seems somewhat surprising and highly unlikely that a local source of PC secretion into the mucosal gel layer is not operational. Goblet cells are an obvious site for further investigation. Approaches that have been used to understand the role of surface active phospholipids in the gastric mucosa and as pulmonary surfactant could be readily employed to gain a greater understanding about the production of mucous PC in the colon. A great deal has been learnt by the use of special probes and stains specific for lipophilic areas and choline based phospholipids (see Lichtenberger²⁰), and their application to colonic mucus seems warranted.

MUCUS PHOSPHOLIPIDS IN ULCERATIVE COLITIS

As outlined above, mucus is abnormal in patients with ulcerative colitis. There is, however, limited knowledge regarding the phospholipid component of colonic mucus in ulcerative colitis. Recently, quantitatively less PC and LPC were reported in samples taken from the rectal mucosa of patients with ulcerative colitis than from healthy controls and patients with Crohn’s disease.¹³ This could be due to reduced production, increased breakdown, or both.

If indeed goblet cells do contribute to the PC content of colonic mucus, they might play more than a passive role in PC depletion of mucus. Goblet cell depletion is a more prominent pathological feature of ulcerative colitis than for Crohn’s colitis. While it might just reflect excessive stimulation of the goblet cells to discharge their mucus, the cells might also be defective in their ability to incorporate PC into the mucus, offering another candidate for the primary abnormality in ulcerative colitis.

Control of PC biosynthesis is a key in the apoptotic programme so that agents that induce apoptosis turn off the biosynthesis of PC.²⁸ Most epithelial cell death in ulcerative colitis appears to follow an apoptotic pathway.^29,30 Thus the increased apoptosis that is occurring in the epithelium may deplete the pool of PC by inhibiting its biosynthesis.

PC can be destroyed within the epithelium, thereby depleting the pool of PC available for secretion into mucus, or can be destroyed within the mucus itself. Mucosal phospholipase A2 activity is increased in patients with ulcerative colitis or Crohn’s disease^31,32 and this activity resides in the epithelium.³³ Protein kinase C, which is involved in several signal transduction pathways linked with inflammatory responses (see Brown and colleagues³⁴), activates a PC specific phospholipase C in the plasma membrane with subsequent breakdown of PC.³⁵ Indeed, colonic mucosa from patients with ulcerative colitis has significantly elevated activity of protein kinase C in the particulate fraction compared with that in normal mucosal samples.³⁶ Insertion of fluorescent analogues of PC into the plasma membrane of cells followed by activation of protein kinase C with phorbol esters has been used to follow the movement of PC and its metabolites via fluorescent microscopy.³⁷ A similar approach could be used to gain a greater understanding about the fate of PC in the colonic mucus in ulcerative colitis and in healthy individuals.

Once phospholipids enter the mucus they remain vulnerable to the action of phospholipases, which may be of epithelial origin or derive from mucosa associated bacteria. In the stomach, Helicobacter pylori colonises the mucous layer in part by producing phospholipases A1, A2, and C, and can extract the host phospholipids for its own protective coating.²⁰H pylori can also generate high concentrations of ammonium ion that competes with phospholipids for negatively charged glycoprotein binding sites.²⁰ It is intriguing to speculate about the possible role of a H pylori-like bacterium that might be responsible (in part at least) for the breakdown of the colonic mucous barrier in ulcerative colitis. Large numbers of bacteria are found within the depleted mucous layer of these patients,^38,39 but whether such bacteria are efficient producers of phospholipases is not known.

The net result of reduced biosynthesis and increased breakdown of phospholipids within the mucosa might be a PC starved system in ulcerative colitis with subsequent depletion of PC available for mucus. Indeed, the observation that plasma phospholipid levels are low in patients with severe ulcerative colitis supports such a concept.⁴⁰ If this situation were combined with excessive destruction of PC within the mucus layer itself, the ability of PC to offer adequate barrier and lubricant function might be severely compromised. Could correction of phospholipid deficiency of the mucosa and/or mucus be of therapeutic value in ulcerative colitis?

PHOSPHOLIPIDS AS A THERAPY

Phospholipids are readily taken up by mucus, so that the mucous content can be substantially increased by their topical administration. In experimental animals, where there is no abnormality of mucus, topical PC protects the mucosa from luminal insults in the form of acetic acid and trinitrobenzenesulphonic acid, which usually induce colitis.^41,42 It might be anticipated that the protective and lubricant function of mucus would be considerably enhanced by topical application of PC in a situation such as ulcerative colitis where PC content is reduced.¹³ Such a simple concept has recently been evaluated⁴³ and the results of a randomised controlled trial are reported in this issue of Gut (see page 966). Delivery to the large bowel lumen was achieved by coating of PC enriched phospholipids with Eudragit-S 100. Release of phospholipids from this pH dependent coating would be expected to predominantly occur in the terminal ileum and proximal colon, a situation mimicking the putative main source of phospholipids in the colonic mucus. Details of the origin (for example, from soy or egg), and the type of phospholipids and fatty acid species used in this study were not described. Nevertheless, PC was successfully incorporated into rectal mucus confirming that the delivery system works.

In a population of 60 patients with chronically active disease, the efficacy of the PC was astoundingly good over a three month treatment period. Compared with a response of 10% in the placebo arm, 90% of the phospholipid treated group responded and 53% were in remission after three months of therapy. No clinically significant side effects were noted. Is this too good to be true?

There were problems with the study design. Despite being described as a single centre study, end points were scored in multiple centres for practical reasons. However, as blinding and randomisation seemed appropriate, heterogeneity of assessment is unlikely to be a significant factor in skewing the results. There were also no data on the time course of efficacy, although a comment in the discussion indicated a slow onset of effect over weeks. The dearth of adverse events in a three month study is somewhat surprising to those used to dealing with randomised controlled drug trials. This suggests laxity in documenting every minor event but again does not detract from the efficacy demonstrated. The placebo treated patients faired badly. A 10% response (and remission) rate is at the lower end of what might be anticipated in most trials, except that patients with chronically active disease, as selected, might have a lower placebo response. The choice of placebo, Eudragit-S 100 coated cellulose, may not have been appropriate. Cellulose would be delivered to the colonic lumen in an identical fashion to that of the phospholipids and, since it is not fermented by colonic bacteria, it might be considered benign and inert. However, cellulose might potentially be abrasive to the mucosa, this being a postulated mechanism by which non-fermentable fibre stimulates epithelial proliferation in fibre starved atrophic colon in otherwise healthy rats.⁴⁴ The placebo therefore might have worsened the outcome. On the other hand, there is no evidence that non-fermented fibre is detrimental to the course of ulcerative colitis. Whatever the case, it is still reasonable to say that three months of therapy with PC rich phospholipids delivered to the colonic lumen were convincingly efficacious in inducing remission in patients with chronically active ulcerative colitis.

What about the mechanism of action? It would be precarious to assign this efficacy to the improved hydrophobicity and barrier function of the mucus without confirmatory data. The hydrophobicity of the mucosal layer can be quantified from the “contact angle” after a drop of saline is placed on the mucosal surface.²⁷ Whether this simple test can be carried out on biopsy specimens or even at colonoscopy is uncertain, but such an assessment in patients with ulcerative colitis seems worthwhile. Alternatively, effects on mucus might only be a minor player in the efficacy. Replenishment of the epithelial pool of PC, with its potential positive effects on improving epithelial health, limiting epithelial destruction, and suppressing its involvement with inflammatory mechanisms, as discussed above, might also be an important mechanism of action.

FUTURE DIRECTIONS

The findings that treatment with PC rich phospholipids permits healing of otherwise difficult to treat ulcerative colitis opens a new approach to the treatment of ulcerative colitis. There is the important need to confirm the results using a more appropriate placebo (perhaps free unsaturated fatty acid such as linoleic acid in place of the cellulose), to determine the appropriate dose, and to define the time course of effect. Additional questions are raised by the complex nature of the phospholipid mixture used. For example, is the effect observed indeed due to the PC component or is the active moiety a minor part of the phospholipid mixture, such as the unsaturated fatty acid component? The question of how this phospholipid mixture is achieving its efficacy, particularly whether it is acting via reinforcement of the mucus, as seems logical, or by other mechanisms, needs to be addressed.

While there has been much previous focus on the mucin glycoproteins, this work shows that more attention needs to be drawn to the less glamorous lipid component, which is not simply playing an inert structural role in the mucus gel layer but rather is a dynamic component of a complex barrier system. The putative efficacy of PC might be enhanced by better distribution using spreading agents. Perhaps liposomes of PC could be used to introduce other therapies such as anti-inflammatory drugs or even antisense RNA that will assist with epithelial healing—two for the price of one! There is also the potential to introduce phospholipids and fatty acid species, such as arachidonic and butyric acids that may have cytoprotective properties. Welcome phospholipids to the cutting edge of ulcerative colitis!