• electron spin resonance
• experimental colitis
• nitroxide radical
• sulfhydryl compounds
• trinitrobenzene sulphonic acid
• transforming growth factor beta
• neuroendocrine tumour
• enteroendocrine cells
• menin
• Smad3
• primary biliary cirrhosis
• bile acids
• liver disease

Conjugated bile acids are water soluble amphipathic end products of cholesterol metabolism that promote lipid transport in the biliary tract and small intestine by forming mixed micelles.¹ Bile acids are formed in pericentral hepatocytes by a complex multienzyme process whose details have at last been largely elucidated.² After formation, their acidic group is linked (“conjugated”) with the amino group of glycine or taurine in an amide bond that is resistant to the proteolytic enzymes present in pancreatic secretion and on the surface of the enterocyte brush border. Conjugated bile acids differ from unconjugated bile acids in being membrane impermeable and water soluble at the pH conditions prevailing in the biliary tract and small intestine.

Efficient ileal conservation of bile acids results in the accumulation of a mass of bile acids termed the bile acid “pool”. Between meals, most of the pool is stored in the gall bladder; with meals, the gall bladder discharges bile into the small intestine where bile acids promote lipid absorption. Both bile acid synthesis and ileal conservation continue after a meal but the gall bladder does not increase in volume in proportion to the amount of bile acids it contains because of its continuous concentration of bile. The gall bladder appears early in vertebrate evolution and genes for gall bladder development appear to have evolved at the same time as genes for bile acid synthesis and intestinal conservation. Development of the enterohepatic circulation and gall bladder storage resulted in far more bile acids being available for digestion than those recently synthesised. Each bile acid molecule is used multiple times before it is lost to the large intestine.³

Feedback inhibition of bile acid biosynthesis in the hepatocyte is well established experimentally.⁴ Interruption of the enterohepatic circulation causes increased bile acid synthesis. This may be modest, for example, increases of 3–4 times are seen in patients taking bile acid sequestrants for hypercholesterolaemia; or it may be marked, for example, increases of 10–15 times are seen with an ileal resection causing severe bile acid malabsorption. Bile acid feeding of any of the natural bile acids occurring in human bile (cholic acid (CA), chenodeoxycholic acid (CDCA), deoxycholic acid (DCA)) suppresses bile acid synthesis, but the effect is relatively small (about a 50% decrease).

The mechanism by which the concentration of bile acids in the hepatocyte regulates bile acid synthesis has been elucidated only recently. Bile acids enter the nucleus and bind to a heterodimeric protein composed of two nuclear receptors, FXR and RXR.^5,6 Binding of the bile acid molecule to FXR changes its confirmation. This in turn leads to a complex sequence of events resulting ultimately in increased synthesis of one or more inhibitory proteins. The inhibitory protein(s) repress(es) the activity of the gene for cholesterol 7 alpha hydroxylase, the rate limiting enzyme in bile acid biosynthesis.⁷ FXR, the bile acid nuclear receptor, has now been crystallised, its structure determined by x ray crystallography, and the shape of the cavity that holds the conjugated bile acid elucidated in the last few months.^8,9

Transport of bile acids by the ileal enterocyte is also modulated in a homeostatic manner analogous to feedback inhibition of bile acid biosynthesis in the hepatocyte. Early studies of bile acid secretion at the Mayo Clinic reported that bile acid secretion increased only modestly or not at all when bile acids were fed,¹⁰ hinting at downregulation of ileal transport in response to bile acid feeding. The first convincing experimental evidence for feedback inhibition of bile acid transport was reported by Lillienau and colleagues¹¹ who performed experiments in the guinea pig. These workers measured total ileal absorptive capacity for conjugated bile acids by perfusing bile acids at such a high rate that the intraluminal concentration remained constant. This technique had been used previously in studies that defined the T_max for ileal transport in rats¹² and humans.¹³ Lillienau et al found that the ileal transport capacity for bile acids decreased after bile acid feeding and was increased by addition of cholestyramine to the diet. This finding was confirmed for the mouse,¹⁴ but using other experimental designs it was not confirmed in the rat (see Lanzini and colleagues¹⁵ ) or in the pig.¹⁶ Thus in this area of physiology there are marked species differences, a problem that continues to bedevil those who try to understand the intricacies of bile acid metabolism. The mechanism by which the concentration of bile acids in the ileal enterocyte modulates enterocyte transport is under active investigation at the moment. As in the hepatocyte, regulation is likely to involve interaction of bile acids with nuclear receptors such as FXR.¹⁷

Lillienau and colleagues¹¹ speculated that “patients with cholestatic liver disease are likely to inappropriately conserve endogenous dihydroxy bile acids such as CDCA and DCA, which are known to be hepatotoxic”. This speculation has now been confirmed in an important clinical study by Lanzini and colleagues¹⁵ in this issue of Gut [see page 1371]. These workers used ⁷⁵Se-SeHCAT, a selenium tagged homologue of taurocholate, whose metabolism was shown by Jazrawi et al to be essentially identical to that of taurocholate.¹⁸ Because SeHCAT is a gamma particle emitter, it can be used to visualise the enterohepatic circulation and has been used for this purpose to measure hepatic excretory function non-invasively in patients with cholestatic liver disease.¹⁹ SeHCAT has also been used to measure the efficiency of ileal conservation of bile acids in diarrhoeal conditions.²⁰

In the experiments reported by Lanzini et al, SeHCAT was used as a surrogate for taurocholate, and its turnover rate quantified by measuring gall bladder radioactivity daily for several days. The rate of decline in radioactivity with time gives the fractional turnover rate of the endogenous bile acid pool. The method used by Lanzini et al does not provide information on bile acid synthesis, which is the product of pool size and turnover rate.²¹

Lanzini et al found that the fractional turnover rate of 14 women with primary biliary cirrhosis (PBC) was, on average, one half that of 14 age matched healthy women. The t½12 (equal to 0.69 divided by the fractional turnover rate) was correspondingly increased. Thus in these patients with all stages of PBC, bile acids were inappropriately retained. The simplest interpretation of this novel finding is that the ileum has sensed a lowered intraluminal bile acid concentration and reacted by increasing its efficiency of bile acid conservation. However, a sensing of the elevated plasma level of bile acids might also contribute. In health, the ileum efficiently downregulates transport in response to increased bile acid loads thereby protecting the liver. When the bile acid pool is lost, as in acute diarrhoeal disease, the ileum upregulates to regenerate the bile acid pool as quickly as possible. In cholestatic liver disease, the signal of decreased intraluminal bile acid concentration acts to mislead the ileal transport system, which cannot know that bile acids are being retained in the hepatocyte because of biliary ductule obstruction. Inappropriate ileal conservation in cholestatic liver disease is homeostasis gone awry.

Lanzini et al made a second important observation. Inappropriate ileal conservation of bile acids was abolished by administration of ursodiol at the usual dose of 15 mg/kg/day. Although ursodiol is fairly well absorbed, it does not suppress endogenous bile acid synthesis because it does not interact with the nuclear receptor FXR.⁶ Thus in patients receiving ursodiol, the enterohepatic circulation has an additional input (probably 10–12 mg/kg/day) of exogenous bile acids, far exceeding endogenous bile acid synthesis (3–5 mg/kg/day). Presumably, ursodiol conjugates secreted by the liver compete for active ileal transport, thus preventing the inappropriate conservation of endogenous bile acids and restoring the fractional turnover rate to normal. Ursodiol is non-cytotoxic and has multiple effects on the hepatocyte that appear to decrease the injurious effects of retained endogenous bile acids and to promote hepatic excretory function.²²

A major question remaining for the hepatologist is whether downregulation of ileal bile acid transport to its normal level by ursodiol therapy is optimal therapy in cholestatic liver disease, or whether it is desirable to decrease the efficiency of ileal conservation to a still greater degree, thereby reducing the return of bile acids to the hepatocyte that is already impacted with bile acids.

Historically, bile acid drainage was used to treat the pruritus of cholestatic liver disease.^23,24 When cholestyramine was introduced, it was also shown to decrease pruritus that, then and still now, is considered by many to arise from increased plasma levels of bile acids.²⁵ Emerick and Whitiington have treated intractable pruritus in children by partial biliary diversion which prevents a fraction of secreted bile acids from reaching the ileum.²⁶ Another surgical approach reported to be successful is ileal bypass which should have the same effect as partial biliary diversion.²⁷ The technique of extracorporeal albumin dialysis removes plasma bile acids and also decreases pruritus.²⁸ A new bile acid sequestrant, colesevalem, has binding properties for bile acids that are superior to those of cholestyramine and has been reported to be more effective than cholestyramine in treating cholestatic pruritus in open label studies.²⁹ The majority of these cholestatic patients were already receiving ursodiol so that these adjuvant therapeutic approaches appear to add efficacy to that achievable by ursodiol therapy alone. All of these approaches will result in less absorption of endogenous cytotoxic bile acids so that the input of bile acids to the liver will be enriched in the recently ingested ursodiol.

The last approach to be considered is inhibition of asbt, the apical transporter of the ileal enterocyte. Ileal absorption of bile acids begins with transport into the enterocyte mediated by the apical sodium dependent transporter (asbt) that has been cloned and characterised in the laboratory of Dawson.³⁰ Development of a potent inhibitor of asbt has been the goal of several pharmaceutical companies.³¹ The target disease for such an inhibitor of bile acid transport was not cholestatic liver disease, but hypercholesterolaemia, a far more prevalent problem. The rationale for the development of such inhibitors was the observation that addition of a bile acid sequestrant to a statin potentiates its hypocholesterolaemic effect by still further upregulating LDL receptor activity.³² Sequestrants are known to induce only mild bile acid malabsorption, suggesting that a potent asbt inhibitor (together with a statin) should be still more effective therapy for hypercholesterolaemia. Although these agents have been promising in animal studies, it is not clear that they will reach the market. Newer more potent statins are quite effective without adjuvant therapy; and older statins will soon become available as generic drugs. In addition, bile acid malabsorption caused by ileal blockade appears to induce diarrhoea in humans because of the cathartic effect of malabsorbed bile acids. This adverse effect has dampened the enthusiasm of the drug development groups. From a commercial standpoint, cholestatic liver disease is unlikely to ever be a target of drug development by “big pharma” because the market is tiny. Let us hope, none the less, that these new potent ileal uptake blockers will be made available to hepatologists so that their value, if any, in treating cholestatic liver disease can be assessed rigorously. The side effect of diarrhoea observed in hypercholesterolaemic patients might be less of a problem in cholestatic patients as the compensatory increase in bile acid synthesis might be dampened because of liver disease.

The paper of Lanzini et al is an importance advance in our understanding of the pathophysiology of cholestatic liver disease. The enterohepatic circulation of bile acids arose in vertebrate evolution to promote nutrition, not to deal with the problem of cholestatic liver disease. Ursodiol therapy corrects the defect in inappropriate conservation. Whether this is enough or whether we should further reduce ileal transport can be tested if the newly developed asbt inhibitors become available to the liver community. Still, all of the approaches discussed above are palliative and we must continue to seek therapeutic approaches that deal with the fundamental aetiology of these conditions, which is likely to be infectious and/or autoimmune.