World Journal of Urology

, Volume 22, Issue 3, pp 178–185

Mucus production after transposition of intestinal segments into the urinary tract

Authors

    • Academic Urology UnitUniversity of Aberdeen
    • Department of SurgeryThe Medical School
  • Jeffrey Pearson
    • Department of Physiological SciencesThe Medical School
  • David Neal
    • Department of SurgeryAddenbrooke’s Hospital
Topic Paper

DOI: 10.1007/s00345-004-0436-6

Cite this article as:
N’Dow, J., Pearson, J. & Neal, D. World J Urol (2004) 22: 178. doi:10.1007/s00345-004-0436-6

Abstract

Following transposition into the urinary tract, intestinal segments continue to produce mucus and problems related to excessive production do not to diminish with time. Currently, 20 human mucin genes have been described and their protein products partially or fully characterised. As the use of transposed intestinal segments in urology increases, there is now a need for a better understanding of mucins at the gene and protein levels. There is also a need for urologists to be aware of the many complications related to excess mucus production. Whilst effective therapeutic measures to reduce mucus production and its related complications remain elusive, it is now clear that without such effective mucoregulatory agents the quality of life of patients will continue to be less than satisfactory. This review describes the biology of mucus and the problems related to continued production following transposition of intestinal segments into the urinary tract. Difficulties related to quantification of urinary mucus, in addition to the structural and mucin gene changes that occur in transposed segments, are addressed.

Keywords

MucusMucin geneUrinary diversionBladder reconstructionNeobladder

Introduction

The use of transposed intestinal segments to reconstruct or replace the function of the lower urinary tract is now commonplace in urological practice. Whilst the main indication is bladder cancer, transposed intestinal segments are also used for benign aetiology such as neurogenic bladder dysfunction, idiopathic detrusor over-activity and chronic inflammatory conditions (such as interstitial cystitis, tuberculosis, schistosomiasis and post-radiation bladder contraction).

Although early reports suggested that intestinal segments transposed into the urinary tract maintained their secretomotor function in the long term [31], some reports have suggested that with time there is a reduction of mucus production [22]. As the popularity of bladder reconstruction and bladder replacement surgery has grown over the past 15 years, there is increasing evidence to suggest that for the majority of patients, mucus production by the transposed intestinal segment and its attendant problems do not diminish in time [1, 21, 31, 33].

This review concentrates on the problems caused by excess mucus production in patients who have had an intestinal segment transposed into the urinary tract.

Mucus structure, function and properties

Mucus is an epithelial cell secretion containing the glycoprotein mucin, lipid, protein, nucleic acid, salts and secretory IgA, and is about 95% water. In the intestine, mucus is secreted by goblet cells, and in the airways by goblet cells and submucosal glands. The surface mucosal cells and the mucus neck cells secrete mucus in the stomach [4]. The normal urothelium does not contain mucus secreting goblet cells and shows no evidence of a secreted mucus layer. On the surface of the intestine mucus forms a gel. To function effectively this gel must be able to flow and maintain a barrier while the mucosa moves during contraction of the underlying smooth muscle. The mucus gel, as well as flowing, must be able to reanneal if fractured. This is a unique property of mucus gels as they have the ability to flow and have rigidity. Therefore, mucus in its physical properties is somewhere between a rigid gel and an entangled system, with a storage modulus G′ that is always greater than G″ (the loss modulus). These moduli indicate solid and liquid behaviour respectively, and because a mucus gel has a small but significant G″ this means that it has some liquid like behaviour, which explains its ability to flow [46].

Mucin structure

Mucins are glycoproteins with a central protein core. Carbohydrate chains are attached to this core via O-glycosidic links between N-acetyl galactosamine and serine and/or threonine. Mucins are highly glycosylated and can contain up to 85% carbohydrate. Fucose, galactose, sialic acid, N-acetyl glucosamine and N-acetyl galactosamine are the sugars present. These carbohydrate chains express blood group activity [2]. Mucins also contain potential N-glycosylation sites, and N-linked sugar chains contain all of the sugars present in O-linked chains in addition to mannose. All mucins contain a region in the protein core that has a specific amino acid sequence rich in serine and/or threonine and proline that repeats. The number of repeats can vary so this region is called the variable tandem repeat region (VNTR). Fig. 1 shows the MUC 2 gene product where the VNTR occupies 40–50% of the core protein and when glycosylated the MUC 2 mucin is about 2×106 molecular weight [5].
Fig. 1

Diagrammatic representation of the MUC2 gene product. a is the VNTR, b is another repetitive region and putative O-glycosylation site rich in serine and theonine, c separates a and b and is rich in cysteine; d and e contain globular domains which have sequence homology with von Willebrand factor domains

Currently, 20 human mucin genes have been described and their protein products partially or fully characterised. Table 1 shows their tissue distribution and chromosomal location. Mucins can be divided into two main groups. The secreted gel forming mucins; MUCs 2, 5AC, 5B, 6 and MUC 19 and the transmembrane, membrane tethered mucins; MUCs 1, 3A, 3B, 4, 11, 12, 13, 15, 16, 17 and 18 [14, 34]. This leaves MUC 7, which, although it is secreted, is non-gel forming, and MUC 9 which codes for a membrane tethered mucin; but a secreted form may also be produced [29]. MUC 8 is still too poorly characterised to assign it to either group [43]. MUC 19, which has recently been described, identified from a genome wide search, codes for a protein core characteristic of a secreted gel forming mucin, however, the secreted gene product has not been isolated and directly characterised [12]. Mucins therefore provide two lines of epithelial defence, first the secreted gel, and second the membrane bound mucins on the epithelial cell apical surface. In the normal ileum and colon (the main segments used in transposed intestinal segment surgery) MUC 2 mucin forms a gel, and MUC 3, and to a lesser extent other membrane bound mucins, form the cell surface barrier. When transposed into the bladder these segments continue to express the above mucins. The normal urinary bladder does not express MUCs 5AC and B or 6 and probably only expresses MUC 2 when diseased [32]. Therefore, there is no evidence for a secreted gel layer protecting the umbrella cells of the bladder. MUC 4, 1 and MUC 3 are expressed to a lesser extent by the urinary bladder, and therefore form a membrane bound barrier at the cell surface [32].
Table 1

Tissue and chromosomal location of MUC genes

Mucin gene

Tissue expression

Chromosomal location

MUC 1

All epithelia, breast, pancreas, small intestine, urinary bladder

1q21

MUC 2

Colon, small intestine, airways

11p15.5

MUC 3B

Colon, small intestine, gall bladder, urinary bladder

7q22

MUC 3A

As MUC 3B, also heart, liver, thymus, pancreas

MUC 4

Airways, colon, small intestine, stomach, cervix, urinary bladder

3q29

MUC 5AC

Airways, stomach, cervix, middle ear

11p15.5

MUC 5B

Airways, submaxillary gland, cervix, gall bladder, middle ear

11p15.5

MUC 6

Stomach, gall bladder, cervix

11p15.5

MUC 7

Salivary glands, airways

4q13–21

MUC 8

Airways

12q24.3

MUC 9

Oviduct

1p13

MUC 10

MUC 11

Colon

7q22

MUC 12

Colon

7q22

MUC 13

GI tract, colon, airways (columnar and goblet cells)

3q13.3

MUC 15

Colon, airways, small intestine, spleen, prostate, breast etc

11p14.3

MUC 16

Ovarian epithelial cells

19 p13.3

MUC 17

Membrane tethered duodenum, stomach, colon

7q 22

MUC 18

Melanoma cell adhesion molecule (immunoglobulin gene superfamily, CD146), normal lung and breast

11q 23

MUC 19

Salivary glands and tracheal submucosal glands

12

There are two gene clusters associated with mucins, all but one (MUC 19) of the secreted gel forming mucins are located at position p15.5 on chromosome 11 and five of the membrane bound mucins are clustered at 7q22 i.e. MUCs 3A and B, 11, 12 and 17. The possession of a VNTR is common to all mucins, however, the structure outside of the VNTR differs markedly between the two mucin families. Membrane bound mucins have the following in common, a short cytoplasmic tail with potential serine/tyrosine phosphorylation sites, a transmembrane domain and a highly glycosylated extracellular domain, the VNTR. All of the mucins localised at 7q22 have two EGF-like domains between the VNTR and the cell membrane. MUC 4 also has two EGF-like domains and MUC13 has three [14, 48]. These EGF-like domains are likely to play roles in cell adhesion, modulation, cell signaling, chemotaxis, wound healing and mucin/growth factor interactions. The four mucins located at 11p15.5, MUCs 2, 5AC, 5B and 6 all form gels at mucin concentrations between 30–100 mg/ml. MUC 2 mucin forms the gel layer of the small and large bowel whilst in the stomach the gel layer is formed by MUC 5AC and 6 mucins. These mucins all polymerise end-to-end via disulphide bridges formed between cysteine rich D domains (see Fig. 1) producing polymers of around 10×106 mol wt. These D domains are homologous to the D domains in the clotting factor, von Willebrand factor (vWF), a protein that also polymerises by disulphide bridge formation. MUCs 5B, 5AC and 2 all have D1, D2, D′ and D3 upstream of the VNTR and D4 downstream. Also downstream are the globular domains B, C and CK. Because of the structural similarities, it is believed that these gel forming mucins have evolved from a common ancestor of the vWF gene [15].

The surface mucus gel forms a continuous layer in the stomach and colon and consists of two layers, a shear compliant gel (sloppy) acting as a lubricant and a shear resistant (firm) gel providing the mucus barrier in vivo [45]. This sloppy gel is the home for the commensal bacteria present in the colon. The firm gel acts as a diffusion barrier preventing access of damaging molecules to the mucosa, e.g. proteases. In addition, in the stomach the firm gel also provides an unstirred layer allowing bicarbonate secreted from the mucosa to neutralise the acid diffusing through the gel from the lumen at the mucosal surface. This generates a pH gradient across the mucus layer of ~2 in the lumen and ~7 at the cell surface [35].

Biosynthesis and secretion of mucus

Biosynthesis of mucin has been studied for MUC 2 in rat colonic mucosal explants and human colonic epithelial cell lines. The first product of synthesis is an N-glycosylated apoprotein containing the N-acetyl galactosamine stubs, which are needed for O-glycosylation. This apoprotein is then dimerized via disulphide bridge formation at the C-terminal. As with vWF biosynthesis this process is believed to take place in the endoplasmic reticulum. Larger disulphide bridge mediated polymers are then produced by trimerization at the N-terminus. This trimerization occurs in the Golgi apparatus along with the completion of the O-linked carbohydrate chains [5, 20].

Mucus secretion occurs by compound exocytosis, both constitutive and regulated, mediated by muscarinic receptors. β adrenergic agents, secretin and prostaglandins stimulate secretion via cAMP mediated mechanisms and nitric oxide via cGMP. Cholinergic agonists act to increase secretion via IP3 and protein kinase C, which cause an increase in intracellular Ca++. The cytokine response to infection and inflammation also stimulates mucus secretion [44]. In addition, epithelial damaging agents, e.g. bile salts and free radicals, stimulate mucus secretion.

Urinary mucus production in patients with transposed intestinal segments

Although anecdotal reports had suggested that there is reduction in mucus production by intestinal segments over time [22], it is now well recognised that transposed intestinal segments maintain secretomotor function in the long-term with mucus production not decreasing in the majority of patients [31, 33].

Evidence from a randomised double blind placebo controlled crossover study

This study, for the first time, provided detailed information regarding the patients’ subjective views about urinary mucus production after transposition of an intestinal segment into the urinary tract (Fig. 2). Data from this study, using a well validated questionnaire (administered by an independent third party), confirmed that for the majority of patients (67%) mucus production did not decrease with time [33]; findings in keeping with the report by Murray et al [31]. The incidence of catheter or urostomy pouch blockage was high with about half of the patients reporting such episodes and up to a third having blockages on a daily basis. Patients capable of voiding naturally were also not spared, with more than half of them (53%) reporting intermittent obstruction of their stream by mucus plugs. The bladder reconstruction/replacement group would appear to have more bother from problems associated with excess mucus production than the ileal conduit group, and this is perhaps reflected by the number of patients who regret having had surgery as a consequence (13% vs 0%, respectively) [33]. The clinical trial involved the treatment of patients with reportedly effective mucoregulatory agents (N-acetyl-cysteine, aspirin, ranitidine, a combination of aspirin and ranitidine or placebo for 3 weeks with a 2 week wash-out period between treatments) [7, 10, 18, 19, 27, 36, 37, 38, 39]. The analysis using our post-treatment questionnaire failed to confirm any significant benefit of the four treatment schedules compared with simply treating patients with placebo. There was no benefit of active treatment over placebo in reducing the amount of mucus produced, in reducing the incidence of catheter or urostomy pouch blockage or the incidence of infection, and neither was the patient’s quality of life improved by active treatment as measured using the SF-36 questionnaire [33]. Non-dialysable material in ileal conduits and bladder reconstructions/replacements averaged 2.15±0.24 g and 1.99±0.16 g respectively, the vast majority not being mucin. Of this, the total PAS positive material amounted to 8.3% and 12.5% of the non-dialysable material in the ileal conduit and bladder reconstruction/replacement groups, respectively. Mucin contributed to about 39% and 23% of the total PAS positive material produced by ileal conduits and bladder reconstructions, respectively; mucin composing on average 3% of non-dialysable material in both groups of patients. Unlike reports of the effectiveness of aspirin and ranitidine in reducing mucin production by transposed intestinal segments [18, 38], our study failed to demonstrate significant benefit of using aspirin, ranitidine, a combination of aspirin and ranitidine or NAC over the use of a placebo [33]. Whilst the effectiveness of NAC instillation as a treatment of acute obstruction in the urinary tract due to mucus plugs was not addressed by our study, its role as an effective mucoregulatory agent at the dose used in this RCT is in question. Similarly, the role of aspirin and ranitidine as mucoregulatory agents in patients with transposed intestinal segments remains unproven. Perhaps the treatment period was too short or the dosages used were insufficient for a significant effect to become apparent for the different treatment schedules employed in the trial. It is also possible that the number of patients in the study group (a total of 43 patients) were insufficient for a treatment effect reaching conventional levels of significance to be detected.
Fig. 2

Patients’ views on mucus production following surgery (permission granted from Journal of Urology: N’Dow J et al. 2001 [33])

Technique of measuring mucin

The laboratory techniques used to quantify mucus production in patients with intestinal segments transposed into the urinary tract have in general been flawed for two reasons: first, the use of any carbohydrate based assay, including the PAS method, to measure mucin content are not specific for mucin [30, 42, 47], and therefore, in the presence of other glycoconjugates (as found in urine from patients following such procedures) cannot be made with any degree of accuracy without prior removal of contaminating glycoconjugates. Second, normal urine contains a large number of different glycoproteins, glycolpeptides and oligosaccharides [8], and therefore the accurate estimation of the mucin content in urine is not without problems. When these issues are not addressed in studies, then the amounts identified as mucus have been as high as 6.7 g/24 h [31]. The process of homogenisation, dialysis, freeze drying, papain digestion, gel chromatography and PAS assay is an accurate method of measuring urinary mucin content and the process did not lead to significant loss of mucin [33]. The anthrone method suffers from problems of interference from free protein and nucleic acid, has been shown to be three times less sensitive than the PAS assay [30], and like the PAS assay is not specific for mucin glycolprotein. The estimation of mucin glycoprotein in the presence of other glycolconjugates, as is the case in urine from patients following intestinal segment transposition, can therefore only be made with reasonable accuracy following some form of purification (i.e. gel chromatography, ion-exchange chromatography or equilibrium density gradient centrifugation).

What happens to the segments following transposition?

Once segments are transposed into the urinary tract, structural changes occur over time with 57% of cases demonstrating profound villous atrophy and 36% having foci of metaplasia (colonic and pyloric types) in a recent study [32]. These changes were not related to time since surgery. More profound changes occur if the transposed segment also functions as a reservoir (as in the reconstructed bladder or orthotopic neobladder), resulting in more prolonged contact with urine rather than merely a conduit for urine (as in ileal conduit diversion). Possible aetiologies include a direct toxic effect of urine on the epithelium, disuse atrophy, or an adaptation to a different functional role of protection rather than absorption. Despite extensive morphological changes seen in the transposed ileal segments, MUC2 and MUC3 continue to be strongly expressed with a maturational gradient for MUC3 (cells in the crypts of Lieberkühn acquiring MUC3 expression as they migrate up toward the villous surface, Fig. 3) [32]. The expression of MUC2 in some segments was not restricted to goblet cells alone (as seen in normal ileum [11]) but there was also expression in absorptive cells, again a feature that was unrelated to the length of follow-up. There was upregulation of MUC1 and MUC4 in transposed segments, the two mucins predominantly expressed in the normal human bladder. Although the clinical implication of this finding is unclear at present, it is possible to speculate that the over-expression of MUC1 and MUC4 is a protective mechanism designed to reduce the amount of urinary solutes being presented to the absorptive cells for absorption or reducing the binding to the cell surface of bacteria and potential carcinogens present in the urine and hence representing another facet in the process of adaptation. There was aberrant expression of MUC5AC, 5B, 6 and 7 in foci of pyloric-type metaplasia in a patient with a transposed ileal segment in situ for 40 years. We also reported upregulation of MUC2, MUC3, MUC4 and MUC5AC in patients with clam enterocystoplasty in whom sections included the anastomotic junction between transposed ileum and bladder urothelium [32]. Whilst the implications of this are unclear presently, it could represent early stages of a premalignant change. We now know that mucin gene expression is altered in a variety of human cancers. In gastric cancer, for instance, there is decreased MUC5AC and MUC6 mRNA expression [23, 24], and in colon cancer there is increased expression of MUC5 and MUC6, genes which are only very weakly expressed if at all in the normal colon [28]. Buisine et al reported the aberrant expression of MUC5AC in 22 cases of rectosigmoid villous adenoma (RVA). MUC5AC is not normally expressed in the colon. The hybridization signal was stronger in RVAs with low grade dysplasia whilst the signal was weaker in high grade dysplasia and limited to a few cells only in RVAs of high grade dysplasia associated with invasive adenocarcinoma. The expression of MUC5AC in RVAs is postulated to be one of the early steps in a multistage process of tumourigenesis [9]. Despite such circumstantial evidence, the link between these mucin gene changes and the development of cancer in patients with a transposed intestinal segment remains far from proven, and just as valid is an argument for these gene changes merely representing early stages in the adaptation process.
Fig. 3 a

Normal ileum showing absence of signal following mock hybridization with omission of oligonucleotide probe; b expression of MUC2 mRNA in normal ileal controls showing restriction of signal to bases of goblet cells; c expression of MUC2 mRNA in transposed ileal segment demonstrating disorganised cellular architecture and gene expression; d MUC3 mRNA expression in transposed ileal segment with maturational gradient evident (Permission granted from Journal of Urology: N’Dow J et al. 2000 [32])

Role of mucus in infection recurrence and persistence

It is well recognised that transposition of intestinal segments into the urinary tract predisposes to urinary tract infections. In a recent prospective cohort study, Escherischia coli was reported as the most frequently isolated organism, responsible for 59% (62 of 105) of monobacterial infections [26]. Other bacteria isolated included Klebsiella species, Proteus species and Enterococcus faecalis. Community acquired E. coli strains were more likely to carry multiple determinants for particular adhesins (P and S fimbriae) and toxins ([alpha]-hemolysin and cytotoxic necrotizing factor) than faecal strains. The most important finding, however, was that E. coli strains persisted for prolonged periods often despite various antibiotic treatments [26]. Whilst a recent report suggests that the recurrence or persistence of urinary tract infections in the normal native bladder is due in some cases to bacteria taking refuge inside umbrella cells protecting themselves from antibiotics [6], another reason in patients with transposed intestinal segments would be urinary mucus. The mucus gel provides a suitable microenvironment and nutrient source for endogenous bacteria to survive in the colon [3, 25]. Microbial organisms express surface adhesins, which recognise host cells or secreted molecules. They bind peptide or carbohydrate receptors. and evidence from the human airway confirms that mucins possess receptors for a number of microorganisms [16, 40, 41]. In the human airway, once microorganisms bind to mucus they can be removed with the sloughed mucus, thereby functioning as a protective barrier and preventing bacteria from binding with cell surface receptors because the bacteria act as though they have bound to the cell surface instead. Non-mucin components of the mucus secretion like secretory IgA antibodies also assist in the bacterial adherence processes by reducing hydrophobicity and enhancing hydrophilic interactions with mucin oligosaccharides [13, 17]. Whilst some patients with a reconstructed bladder or bladder replacement are capable of voiding naturally, other patients must perform a clean intermittent self catheterisation as the mechanics of emptying are not as efficient as in the native human bladder, resulting in some patients maintaining significant residual urine volumes. It is now well recognised that a significant percentage of such patients report blockage of their catheters or their flow by mucus plugs when they void naturally [33]. Although not proven to date, we would hypothesise that mucus plugs are heavily colonised with bacteria, which not only prevents antibiotics reaching the bacteria at the desired concentrations but also facilitates persistence of infection as reported recently [26]. Some 63% of patients in our randomised trial had suffered from a symptomatic urine infection, and of these 59% reported that mucus production had increased during such an incident [33]. Whilst this is an expected protective response of the intestinal segment to irritation/inflammation, the resultant excessive mucus produced potentially contributes to the persistence of the microorganisms.

Quite apart from the risks of metabolic, nutritional and gastrointestinal complications associated with the transposition of intestinal segments to the urinary tract, patients should always be warned of the possibility not only of the continued production of mucus, but also the possibility of troublesome complications directly related to excess mucus production, such as the recurrence or persistence of infection and impaired bladder emptying. Perhaps with such preoperative counselling we can reduce the level of dissatisfaction experienced by a few patients.

The future

The search for an effective mucoregulatory agent is long overdue if the quality of life of the cohort of patients with transposed intestinal segments is to be improved. We would hope that patients be better informed about the potential complications associated with excess mucus production in the interim, while we strive to find the elusive ideal mucoregulatory agent that not only reduces the production of mucin but also prevents mucus plug formation with its attendant problems. It may be that this will never be achievable without affecting the beneficial protective functions of mucus as outlined in this review. The field of transposed intestinal segment surgery will undoubtedly continue to expand for the foreseeable future and, although progress has been made in our understanding of short-term and long-term complications, the risk of developing cancer as a direct consequence of this type of surgery remains poorly defined. Whilst further studies into the altered expression of mucin genes in the transposed intestinal segments and the upregulation of certain mucin genes in the urothelium close to the anastomotic junction with the transposed segment may help to clarify this risk, it remains doubtful whether these changes in mucin expression predate any obvious histopathological evidence of a malignant process, and if so for how long.

Copyright information

© Springer-Verlag 2004