Digestive Diseases and Sciences, Vol. 37, No. 5 (May 1992), pp. 689-696
REVIEW ARTICLE
Defense System in the Biliary Tract Against Bacterial Infection J.Y. SUNG, J.W. COSTERTON, and E.A. SHAFFER
Bacteria can invade the biliary tract by ascending from the duodenum and via the hematogenous route from the hepatic portal venous blood. The sphincter of Oddi, situated at the junction of the biliary tract and the upper gastrointestinal tract, forms an effective mechanical barrier to duodenal reflux and ascending bacterial infection. Conversely, Kupffer cells and the tight junctions between hepatocytes help prevent bacteria and toxic metabolites from entering the hepatobiliary system from the portal circulation. The continuous flushing action of bile and the bacteriostatic effects of bile salts keeps the biliary tract sterile under normal conditions. Secretory immunoglobulin A (slgA), the predominant immunoglobulin in the bile, and mucus excreted by the biliary epithelium probably function as antiadherence factors, preventing microbial colonization. When barrier mechanisms break down, as in surgical or endoscopic sphincterotomy and with insertion of biliary stents, pathogenic bacteria enter the biliary system at high concentrations and take up residence on any foreign bodies. Intrabiliary pressure is a key factor in the development of cholangitis. Chronic biliary obstruction raises the intrabiliary pressure. This adversely influences the defensive mechanisms such as the tight junctions, Kupffer cell functions, bile flow, and slgA production in the system, resulting in a higher incidence of septicemia and endotoxemia in these patients. Knowledge of biliary defense against infection is still quite primitive. Unclear are the roles of slgA in the bile, mechanism of bacterial adhesion to the biliary epithelium, Kupffer cell function in biliary obstruction, and the antimicrobial activity of bile salts. KEY WORDS: defense mechanism; biliary tract; bacteria.
Human bile is sterile under normal conditions (1-3), yet bacteria exist on either side of the hepatobiliary system. The biliary tract exits into the duodenum, which is periodically colonized. The duodenal microorganisms are believed to be a major source of infection in cholangitis. Conversely, the liver, where bile forms, receives 70% of its blood supply from the portal vein, which also periodically delivers bacteria from the intestine. The portal vein hence constitutes another source of infection for the
biliary system. To guard against the invasion of bacteria, the biliary tract has been equipped with several defense mechanisms, which include (1) anatomical barriers (tight junctions between hepatocytes and the sphincter of Oddi), (2) physical mechanisms (bile flow and biliary mucus), (3) chemical factors (bile salts), and (4) immunological defense (Kupffer cells and immunoglobulin). This article is intended to review the effectiveness of these defense mechanisms in the biliary system.
Manuscript received June 5, 1991; revised manuscript received December 11, 1991; accepted December 16, 1991. From the Department of Biological Sciences and Department of Medicine, University of Calgary, Canada. Address for reprint requests: Dr. J.Y. Sung, Department of Biological Sciences, University of Calgary, Calgary, Alberta, Canada T2N 1N4.
BACTERIOLOGY OF THE BILIARY TRACT Microorganisms generally colonize the mucosal lining of the human gut and other hollow organs exposed to the environment. These bacteria are embedded in the glycoprotein covering of mucus as
Digestive Diseases and Sciences, Vol. 37, No. 5 (May 1992)
0163-2116/92/0500-0689506.50/09 1992PlenumPublishingCorporation
689
SUNG ET AL a sessile population (4, 5). The biliary tract of a healthy individual does not usually harbor bacteria. Studies of gallbladder bile aspirated from patients not suffering from biliary disease suggest that few planktonic bacteria exist in the biliary tract (1, 2). In the feline biliary tract, the bile duct mucus does not harbor microorganisms, except occasionally at the sphincter of Oddi (3), By implanting sterile foreign bodies in the gallbladder as a "bacterial trap," we have demonstrated that there is a transient passage of bacteria in the biliary tract of cats (6). Most reports on biliary organisms show a predominance of alimentary organisms such as the coliforms, Streptococci spp., and Klebsiella spp. (7-10). The importance of anaerobic organisms in biliary infections has received more attention in the past decade. Anaerobes have been associated with chronic biliary diseases and in biliary-intestinal anastomosis (1l, 12).
tissue that yielded positive cultures for bacteria (18). In acute cholangitis, some 40% of patients have bacteria isolated from the portovenous blood (19). Experimental infusion of specific bacteria into the portal circulation results in these bacteria appearing in bile (20, 21), These findings suggest that there is periodic delivery of bacteria from the intestine into the portal circulation; normally this hematogenous source is cleared by the phagocytes in the liver. The entry of bacteria via the portal blood seems to be facilitated by a raised intrabiliary pressure (21). Thus, it appears that bacteria may pass from the gastrointestinal tract into the biliary tract either by an ascending route, as in the case of sphincterotomy, or by the hematogenous route via the portal blood. To guard against bacterial invasion from duodenal reflux and from portal venous bacteraemia, the body has developed the following defense mechanisms.
SOURCE OF BACTERIA IN BILIARY INFECTION
MECHANICAL BARRIERS
The portal of entry of bacteria into the bile remains controversial but ascending infection has received most prominence: In patients with different biliary tract problems, positive bile cultures are most commonly found in those with ductal stones or carcinoma of the ampulla, which produce intermittent obstruction to bile flow (7). Another increasingly common mechanism for bacterial ascent results from disruption of the sphincter of Oddi following endoscopic sphincterotomy (13) or surgical choledochotomy (14). Here, bacteria presumably carried by duodenal reflux are frequently found in the common duct bile and persist for months after the procedures. Tumors obstructing the common bile duct are not usually complicated by cholangitis until after endoscopic or surgical interventions (15). The fixed obstruction presumably prevents the ascent of bacteria from the duodenum. The other potential route of bacterial invasion into the biliary tract is via the portal venous blood. In typhoid carriers, Salmoneila spp. harbored in the gallbladder enter the biliary tract as a consequence of transient portovenous bacteremia that occurs during the primary intestinal disease (!6). Transient portal vein bacteremia may even be a "normal" occurrence. Schatten et al, in a study of portal blood cultures, obtained positive results in 32% of patients undergoing upper abdominal surgery (17). In patients with nonbiliary diseases, 17% had liver 690
Tight Junctions Between Hepatocytes. Within the liver lobule, hepatocytes are arranged in plates bathed on each side by plasma. The lateral surfaces of the hepatocytes hold onto one another by desmosomes. Bile canaliculi are formed as blind-ended grooves between two adjacent hepatocytes. It is an irregular space with microvilli projecting from the side wall of the hepatocytes. The tight junctions situated on each side of the bile canaliculi seal these blind sacs from the liver sinusoids (Figure 1). The permeability of the tight junction is regulated by various factors such as bile acids (22), estrogen (23), and intrabiliary pressure (24). The integrity of the tight junction is important in preventing the entry of bacteria from the sinusoidal blood into the biliary tract. Electron microscopic studies have shown that an increased intrabiliary pressure causes morphological distortion of the tight junctions (25, 26). The cylindrical strands in the tight junction are decreased and large proteins, such as horseradish peroxidase, can penetrate the tight junction and easily move from blood to bile, reflecting disruption of the permeability barrier (26). When pressure in the biliary tract is raised to a certain critical level, tight junctions break down and the blood-bile barrier is compromised. From clinical experience (27) and animal experiments (28), an increased biliary pressure is an essential element leading to septicemia in cholangitis. Cholangiovenous reflux has been demonstrated in the presence of a high intrabiliary Digestive Diseases and Sciences, Vol. 37, No. 5 (May 1992)
BILIARY TRACT DEFENSE AGAINST INFECTION
~
id
ucts attaching to a surface, develops rapidly on these stents causing occlusion of the lumens. The importance of an intact sphincter of Oddi in preventing duodenal bacteria from invading the biliary tract is supported by clinical observations (7, 13, 14). PHYSICAL MECHANISMS
Sinusoid
Fig I. Single-cell-thick hepatocyte is separated from sinusoidal blood on either side by endothelial cells (E). Lining endothelial cells have a large fenestration (f) in the cytoplasm and lack a basal membrane, which allow free access of plasma to space of Disse (D). Kupffer cells (K) are situated at the sinusoidal surface of the endothelium acting as a phagocyte. The canalicular membranes of adjacent hepatocytes are joined by tight junctions (T), which seal the bile canaliculus (C) from the lateral intercellular space and sinusoidal space.
pressure (27). Relieving the intrabiliary pressure in obstructive jaundice is thus a critical measure to control infection in these patients (27, 29). Further, the entry of bacteria into the biliary tract via portovenous blood is facilitated by an increased intrabiliary pressure (20, 21). In chronic biliary obstruction, the liver also exhibits t r a n s c e l l u l a r regurgitation of bile into the systemic blood through the hepatocytes (30). Entry of bacteria may occur anywhere along the biliary epithelium; damage to the tight junctions allowing access via the paracellular pathway is likely the major factor causing the breakdown of the blood-bile barrier. Sphincter of Oddi. Motility studies using endoscopic manometry (31) and cineradiography (32) have shown that the sphincter of Oddi is more than just a passive conduit for the common bile duct to exit into the duodenum. The basal tone of the sphincter and any phasic contractions control bile delivery into the duodenum; this high pressure zone prevents reflux of duodenal contents into the biliary tract. The sphincter of Oddi, which separates the colonized duodenum from the uncolonized biliary tract acts as a mechanical barrier to microbial colonization (3). Animals experiments have repeatedly shown that disrupting the barrier by placing biliary stents across the sphincter into the lumen of duodenum would allow bacteria ascending into the biliary tract (33, 34). Bacterial biofilm, a complex association of microorganisms and microbial prodDigestive Diseases and Sciences, Vol. 37, No. 5 (May 1992)
Bile Flow. Canalicular bile flow results from the active secretion of solutes, followed osmotically by obligatory water flow. Bile salts, being the most abundant organic anions in the bile, are considered the major driving force in bile formation, the socalled "bile salt-dependent flow" (BSDF). A linear relationship exists between bile flow and bile salt secretion in the liver (35, 36). Extrapolation of the line relating bile flow to bile salt secretion defines a component of canalicular flow that theoretically would be present if no bile salts were secreted. This bile salt-independent flow (BSIF) represents about 50% of total canalicular bile flow in human. An average of 800-1000 ml of bile in humans effectively flushes the bile ducts throughout the day. Therefore, like the urine flushing the urinary tract continuously and menstrual flow cleansing the uterine cavity during every menstrual cycle, the physical movement of bile hinders bacteria from colonizing the biliary mucosa. In biliary obstruction, bile flow and bile salt secretion declined in parallel (37), indicating that a raised biliary pressure suppresses the BSDF. Conversely, gram-negative septicemia from urinary tract infection or other intraabdominal sepsis is often associated with intrahepatic cholestasis without biliary obstruction (38, 39). Lipopolysaccharide (LPS) components purified from the outer membrane of gram-negative bacteria exerts a cholestatic effect in the isolated perfused rat liver by inhibiting the BSIF of bile (40, 41). The decline in bile production due to suppression of BSDF and BSIF reduces the flushing effects of bile and hence might predispose to biliary infection in these patients. Mucus. The extrahepatic bile ducts are lined by a tall columnar mucus-secreting epithelium. Mucus forms a water-insoluble gel adherent to the mucosal surfaces, providing a stable unstirred layer of water between the gastrointestinal mucosa and the lumen. The gastrointestinal mucus plays an important role in excluding pathogenic microorganisms by forming a barrier to pathogens in the lumen, coaggregating the microorganisms, yet selecting the normal bac-
691
S U N G ET A L
fatty acids, leading to the formation of brown pigment stones (45, 46) or cholesterol microcrystals for cholesterol gallstone formation. CHEMICAL FACTORS
Fig 2. Transmission electron micrograph from a section of brown pigment stone showed that bacteria were surrounded by ruthenium red-stained material, compatible with bacterial glycocalyx, forming microcolonies. Bar = 1 ~m.
terial flora and retaining the secretory immunoglobulin A (slgA) within the mucus to permit specific immune protection (42). Although true for the intestine, the physiological role of mucus in the biliary tract is largely unexplored. Bile mucin is known to be involved in the formation of cholesterol gallstones (43, 44), but its antibacterial activity remains to be investigated. Mucin might also play an important role in the pathogenesis of brown pigment stones in the biliary tract. Morphological studies of these stones revealed that bacterial microcolonies are surrounded by highly hydrated carbohydrate material forming a bacterial biofilm (45, 46) (Figure 2). The carbohydrate coating on the surface of these bacteria, called glycocalyx, is doubtlessly bacterial in origin. Bacterial glycocalyx and mucin in the bile may even promote coaggregation of bacteria that could then act as a nidus for the crystallization of unconjugated bilirubin and
692
Bile Salts. Bile salts possess inhibitory effects on the proliferation of enteric microorganisms and thus might aid in preventing infections of the biliary tract (47) and control the bacterial flora of the gastrointestinal tract (48, 49). Both conjugated and unconjugated bile salts, at physiological concentrations, have been shown to inhibit E. coli, Klebsiella spp., and Enterococcus spp. in an in vitro experiment (47). In general, the inhibitory effect seems to be more significant for gram-positive bacteria than gram-negative bacteria (50, 51). The inhibitory effects of bile salts has been used in selective media to isolate specific organisms such as the SS medium for salmonellae and shigellae and the bacteroides bile esculin medium for Bacteroides fragilis (52, 53). The bacteriostatic activity of bile salt is related to the hydrophobicity and the detergent properties of these molecules. Therefore, bile salts with fewer hydroxyl group(s) and only a-hydroxylation are more potent than those with more hydroxyl group(s) and with [3-hydroxylation in suppressing the growth of bacteria (54). In more concentrated gallbladder bile, bile salts do not only exist as free ions but form mixed micelles with phospholipid and cholesterol. Vesicles of phospholipid and cholesterol also exist in the bile. As the hydrophobic component of bile salts are engaged in these complex micellar structures, the detergent activity of these molecules is altered. The antibacterial activity of even the most toxic bile salt in the micellar solution has been shown to be much attenuated (54). Hence, while the in vitro antibacterial effects of bile salts is well recognized, their role in defending against biliary infection in in vivo conditions might be overemphasized. IMMUNOLOGICAL DEFENSE Kupffer Cells. The functions of Kupffer cells and their role in the defense against biliary infection is complex (55). Kupffer cells represent about 30% of the sinusoidal cells in the liver and constitute approximately 80% of all fixed macrophages in the body (55). They are strategically located inside the liver sinusoids to filter toxic substances, endotoxins, and bacteria that come in through the portal circulation by phagocytosis (Figure 1). Besides actDigestive Diseases and Sciences, Vol. 37, No. 5 (May 1992)
BILIARY TRACT DEFENSE AGAINST INFECTION ing as scavengers, Kupffer cells also function as antigen-presenting cells for the induction of T-lymphocyte response (56). They are capable of releasing cytokines such as intefleukins, interferons, and tumor necrosis factor (55, 57). The reticuloendothelial system in the liver is impaired in chronic liver diseases (58) and in biliary obstruction (59-61). The integrity of Kupffer cell function is an important factor affecting the mortality from endotoxemia in biliary obstruction (60). There are at least three possible mechanisms leading to the failure of the reticuloendothelial system of the liver. First, in chronic liver diseases, the number of Kupffer cells in the liver is decreased (58). The reduced Kupffer cell population accounts for the depressed phagocytic activity in chronic liver failure. A second possible mechanism of impaired Kupffer cell function relates to fibronectin, a carbohydrate-binding protein produced primarily by hepatocytes and endothelial cells in the liver. Fibronecfin acts as an opsonin for a variety of targets, and fibronectin receptors are present on Kupffer cells. The hepatic production of lectin is reduced and so opsonin-mediated phagocytosis is suppressed in chronic liver disease (62). Finally, in biliary obstruction, clearance of both E. coli (55) and 99mTc-labeled sulfur colloid (60) injected into the circulation is significantly impaired. The proposed mechanism of depressed Kupffer cell function in this situation is due to an elevated plasma concentration of bile salts. In vitro studies have shown that both conjugated and unconjugated bile acids inhibit Kupffer cells phagocytosis (63, 64) and the binding of lipopolysaccharides (65). Differences likely exist depending upon the hydrophobichydrophilic balance of the bile salt molecule, but, this remains unexplored. Secretory ImmunologicalResponses. The predominant immunoglobulin in bile is secretory immunoglobulin A (sIgA) (66, 67). Immunoglobulin M (IgM) and immunoglobulin G (IgG) exist in much lower levels. In some animals (rats, mice, and rabbits), polymeric IgA produced by the lymphoid tissue of the gut enters the serum and is transported to bile by an active receptor-mediated mechanism (67). Secretory component (SC), a glycoprotein synthesized and expressed on the sinusoidal surface of the hepatocytes, binds polymeric IgA in the sinusoidal blood. This is followed by endocytosis and transportation into the bile canaliculi as sIgA (68, 69). In other animals (dog, guinea, sheep), as well as in humans, there is much less hepatic transport of Digestive Diseases and Sciences, Vot. 37, No. 5 (May 1992)
circulating IgA because of the lack of SC in their hepatocytes (70, 71). Transport of IgA into human bile occurs across the biliary epithelium where the SC is situated (72, 73). A SC-independent asialoglycoprotein receptor-mediated mechanism has been shown to provide an alternative IgA transport system across human hepatocytes (74, 75). Plasma cells forming minor glands in or adjacent to the wall of the human bile duct also synthesize IgA. The local production of IgA by these cells accounts for 50% of IgA in human bile (67). The biological functions of biliary IgA are unclear. In vitro studies have shown that the Fc receptors of sIgA are not involved in complement fixation, opsonization, or antibody-dependent cellmediated cytotoxicity (ADCC) reactions. Thus the source of IgA on mucosal surfaces does not function as a bactericidal factor. One important function of IgA in the intestine is to enhance the relative impermeability of the epithelium, by binding to antigens in the gastrointestinal lumen (76, 77). sIgA can bind and thus prevent bacteria (76, 77) and enteric virus (78) from attaching to and penetrating intestinal epithelial cells. In humans, the importance of slgA is largely unknown. Certainly, patients with selective IgA deficiency do not have an increased frequency of bacterial infection of the biliary tract. In the absence of sIgA, due to immaturity of the mucosal lymphoid system in infancy or specific IgA deficiencies, enteric antigens are found in the systemic circulation in high concentrations (79). In patients suffering from cholestasis due to biliary obstruction, the excretion of IgA by the liver is impaired and serum IgA levels elevated (80, 81). The serum IgA-immune complex (IgA-CIC) concentrations rise; this immune complex is deposited in the glomeruli in biliary obstruction after experimental ligation of the common bile duct (82) and in biliary atresia (83). Ohshio et al measured the serum levels of IgA-CIC and the corresponding endotoxin levels in 33 patients suffering from biliary obstruction and cholangitis (81). The endotoxin levels correlated with the amount of IgA-CIC existing in the blood. Impaired IgA transport might contribute to the entry of endotoxin into the systemic circulation and septicemic shock. When biliary obstruction is relieved, serum slgA returns to normal (81, 84). In a clinical study of 71 patients suffering from cholelithiasis, bile slgA levels of patients with brown pigment stones were significantly lower than the control group (85). Among the cholelithiasis pa-
693
SUNG ET AL TABLE 1. SUMMARYOF DEFENSEMECHANISMSIN B1LIARY SYSTEM Ascending infection (duodenal reflux)
Defense sphincter of Oddi mechanisms
Hematogenous infection (portal bacteraemia)
tight junction* Kupffer cell phagocytosis*
hydrophobic bile salts* flushing effect of bile* ? slgA* ? mucus *When the intrabiliary pressure is raised during biliary obstruction, this defense mechanism will be impaired. tients, bile s l g A values in t h o s e with active cholangitis w e r e l o w e r than those w i t h o u t infection. T h e a u t h o r suggested that the antibacterial a d h e r e n c e effect o f s l g A is i m p o r t a n t in avoiding microbial colonization o f the biliary epithelium. T h e f u n c t i o n o f this p r e d o m i n a n t s e c r e t o r y i m m u n o g l o b u l i n in the biliary tract n e e d s further investigation.
SUMMARY T h e central elements fostering the d e v e l o p m e n t o f cholangitis are bacteria, biliary stasis, and an increase in intrabiliary pressure. A l t h o u g h bacteria are essential for the d e v e l o p m e n t o f cholangitis, not all patients with b a c t e r i a in the bile d e v e l o p cholangitis. I n t e r r u p t i o n o f bile flow and h e n c e the loss o f c o n t i n u o u s m e c h a n i c a l flushing o f the biliary s y s t e m is crucial in the p a t h o g e n e s i s o f cholangitis. W h e n the bile flow is o b s t r u c t e d , the raised intrabiliary p r e s s u r e s u p p r e s s e s the secretion o f bile salt and f u r t h e r r e d u c e s bile flow. T h e p r o t e c t i v e functions o f the tight j u n c t i o n s , K u p f f e r cells, and s e c r e t o r y i m m u n o g l o b u l i n will also be affected. A n i n c r e a s e d biliary p r e s s u r e is thus a k e y f a c t o r in the d e v e l o p m e n t o f septicemia in cholangitis, O b s t r u c tion to bile flow will result in biliary tract infection e v e n t h o u g h all o t h e r c o m p o n e n t s o f the defense (immunoglobulins, m u c u s and bile salts) a p p e a r to be intact.
ACKNOWLEDGMENTS This work is supported by Izaak Walton Killam Memorial Scholarship, The Croucher Foundation Fellowship for Scientific Research (J. Sung), The Medical Research Council of Canada (E. Shaffer), and Natural Sciences and Engineering Research Council of Canada O.W. Costerton).
REFERENCES 1. Csendes A, Fernandez M, Uribe P: Bacteriology of the gallbladder bile in normal subjects. Am J Surg 129:629-631, 1975
694
2. Scott AJ: Bacteria and disease of the biliary tract--progress report. Gut 12:4873, 1971 3. Sung JY, Olson ME, Leung JWC, Lundberg MS, Costerton JW: The sphincter of Oddi is a boundary for bacterial colonization in the feline biliary tract. Micro Eco Health Dis 3:199-207, 1990 4. Savage DC: Microbial ecology of the gastrointestinal tract. Annu Rev Microbiol 31:107-133, 1977 5. Costerton JW, Cheng KJ: Colonization of tissue surfaces by autochthonous bacteria. In Microbiology. D Schlessinger (ed). Washington, DC, ASM publication, 1982, pp 266-273 6. Sung JY, Leung JWC, Olson ME, Lundburg MS, Costerton JW: Demonstration of transient bacterobilia by foreign body implantation in the feline biliary tract. Dig Dis Sci 36:943948, 1991 7. Suzuki Y, Kobayashi A, Ohto M, Tsuchiya Y, Saisho H, Kimura K, Ono T, Okuda K: Bacteriological study of transhepatically aspirated bile: Relation to cholangiographic findings in 295 patients. Dig Dis Sci 29:109-115, 1984 8. Pitt HA, Postier RG, Cameron JL: Biliary bacteria: Significance and alterations after antibiotic therapy. Arch Surg 117:445-449, 1982 9. Keighley MRB: Microorganisms in the bile. Ann R Coll Surg Engl 59:328-334, 1977 10. Alexander-Williams J: Bacteriology of biliary disease. In Surgery of the Gallbladder and Bile Ducts. LW Way, CA Pellegrini (eds). New York, WB Saunders, 1987, pp 93-102 11. England DM, Rosenblatt JE: Anaerobes in human biliary tract. J Clin Microbiol 6:494-498, 1977 12. Finegold SM: Anaerobes in biliary tract infection. Arch Intern Med 139:1338, 1979 13. Gregg JA, Girolami PD, Carr-Locke DL: Effects of sphincteroplasty and endoscopic sphincterotomy on the bacteriologic characteristics of the common bile duct. Am J Surg 149:668-671, 1985 14. Feretis CB, Contou CT, Manoruas AJ, Apostolidis NS, Golematis BC: Long term consequences of bacterial colonization of the biliary tract after choledochotomy. Surg Gynecol Obstet 159:363-366, 1984 15. O'Conner MJ, Schwartz ML, McQuarrie DG, Summer HW: Cholangitis due to malignant obstruction of biliary flow. Ann Surg 193:341-345, 1981 16. Tynes BS, Utz JP: Factors influencing the cure of salmonella carriers. Ann Intern Med 57:871-882, 1962 17. Schatten WE, Desprez JD, Holden WD: Bacteriologic study of portal vein blood in man. Arch Surg 71:404-409, 1955 18. Dye M, Mcdonald A, Smith G: The bacterial flora of the biliary tract and the liver. Br J Surg 65:285-287, 1978 19. Ong GB: A study of recurrent pyogenic cholangitis. Arch Surg 84:199-255, 1962 20. Dineen P: The importance of the route of infection in experimental biliary tract obstruction. Surg Gynecol Obstet 19:1001-1008, 1964 21. Sung JY, Shaffer EA, Olson ME, Leung JWC, Lam K, Costerton JW: Bacterial invasion of the biliary system by way of the portal-venous system. Hepatology 14:313-317, 1991 22. LaGarde S, Elias E, Wade JB, et al: Structural heterogeneity of hepatocyte "tight junctions": A quantitative analysis. Hepatology 1:193-203, 1981 23. Forker EL: The effect of estrogen on bile formation in the rat. J Clin Invest 48:654-663, 1969 Digestive Diseases and Sciences, Vol. 37, No. 5 (May 1992)
BILIARY TRACT D E F E N S E AGAINST I N F E C T I O N 24. Easter DW, Wade JB, Boyer JL: Structural integrity of hepatocyte tight junctions. J Cell Biol 96:745-749, 1983 25. Robenek H, Herwig J, Themann H: The morphologic characteristics of intercellular junctions between normal human liver cells from patient with extrahepatic cholestasis. Am J Pathol 100:93-103, 1980 26. Devos R, Desmet VJ: Morphological changes of the junctional complex of the hepatocytes in rat liver after bile duct ligation. Br J Exp Pathol 59:220-227, 1978 27. Lygidakis NJ, Brummelkamp WH: The significance of intrabiliary pressure in acute cholangitis. Surg Gynecol Obstet 161:465-469, 1985 28. Huang T, Bass JA, Williams RD: The significance of biliary pressure in cholangitis. Arch Surg 98:629-632, 1969 29. Leung JWC, Chung SCS, Sung JY, Banez VP, Li AKC: Urgent endoscopic drainage for acute suppurative cholangitis. Lancet 1:1307-1309, 1990 30. Jones AL, Hradek GT, Schmucker DL, Underdown BJ: The fate of polymeric and secretory immunoglobulin A after retrograde infusion into the common bile duct in rats. Hepatology 4:1173-1183, 1984 31. Geenen JE, Hogan WJ, Dodds WJ, Stewart ET, Amdorfer RC: Intraluminal pressure recording from human sphincter of Oddi. Gastroenterology 78:317-324, 1980 32. Toouli J, Dodds WJ, Honda R, Sarua S, Hogan WJ, Komorowski RA, Linehan JH, Amdorfer RC: Motor function of the opossum sphincter of Oddi. J Clin Invest 71:208-220, 1983 33. Geoghegan JG, Branch MS, Costerton JW, Pappas TN, Cotton PB: Biliary stent occlude earlier if the distal tip is in the duodenum in dogs. Gastrointest Endosc 1991 (abstract) 37:257, 1991 34. Sung JY, Leung JWC, Shaffer EA, Lain K, Olson ME, Costerton JW: Ascending infection of the biliary tract after surgical sphincterotomy and biliary stenting. J Gastroenterol Hepatol (in press) 35. Scharschmidt BF, Van Dyke RW: Mechanisms of hepatic electrolyte transport. Gastroenterology 85:1199-1214, 1983 36. Boyer JL, Bloomer JP: Canalicular bile secretion in man: Studies utilizing the biliary clearance of 14C-mannitol. J Clin Invest 54:773-781, 1974 37. Straberg SM, Dorn BC, Small DM, Egdahl RH: The effect of biliary tract pressure on bile flow, bile salt secretion and bile salt synthesis in the primate. Surgery 70:140-146, 1971 38. Eley DF, Hargreaves T, Lamberth HP: Jaundice in severe infections. Br Med J 2:75-55, 1965 39. Miller DF, Irvin RW: Jaundice in acute appendicitis. Lancet 1:321-323, 1969 40. Utili R, Abernathy CO, Zimmerman HJ: Cholestatic effects of Escherichia coli endotoxin on the isolated perfused rat liver. Gastroenterology 20:248-253, 1976 41. Utili R, Abernathy CO, Zimmerman HJ: Studies on the effects of Escherichia coli endotoxin on canalicular bile formation in the isolated perfused rat liver. J Lab Clin Med 89:471-482, 1977 42. NeutraMR, Forstner JF: Gastrointestinal mucus: Synthesis, secretion and function. In Physiology of the Gastrointestinal Tract. LR Johnson (ed). New York, Raven Press, 1987, pp 975-1002 43. Carey MC, Cahalane MJ: Whither biliary sludge? Gastroenterology 95:508-523, 1988 Digestive Diseases and Sciences, Vol. 37, No. 5 (May 1992)
44. Levy PF, Smith BF, LaMont JF: Human gallbladder mucin accelerates nucleation of cholesterol in artificial bile. Gastroenterology 87:270-275, 1984 45. Stewart L, Smith AL, Pellergrini CA, Motson RW, Way LW: Pigment gallstones form as a composite of bacterial microcolonies and pigment solids. Ann Surg 206:242-250, 1987 46. Leung JWC, Sung JY, Costerton JW: Bacteriological and electron microscopy examination of brown pigment stones. J Clin Microbiol 27:915-921, 1989 47. Stewart L, Pellergrini CA, Way LW: Antibacterial activity of bile acids against common biliary tract organisms. Surg Forum 37:157-159, 1986 48. Hill MJ, Crowther JS, Drasar BS, Hawkswroth G, Aries V, Williams REO: Bacteria and the aetiology of cancer of the large bowel. Lancet 1:95-97, 1971 49. Hill MJ: Factors controlling the microflora of the healthy upper gastrointestinal tract. In Human Microbial Ecology. MJ Hill, PD Marsh (eds). Boca Raton, Florida, CRC Press, 1989, pp 57-86 50. Floch MH, Gershengoren W, Elliott S, Spiro H: Bile acids inhibition of the intestinal microflora: A function for simple bile acids? Gastroenterology 61:228-233, 1971 51. Floch MH, Gershengoren W, Diamond S, Hersh T: Cholic acid inhibition of intestinal bacteria. Am J Clin Nutr 23:8-10, 1970 52. Leifson E: New culture media based on sodium deoxycholate for the isolation of intestinal pathogens and for enumeration of colon bacilli in milk and water. J Pathol Bacteriol 40:581-586, 1953 53. Nash P, Krenz MM: Culture media. In Manual of Clinical Microbiology, 5th ed. A Balows (ed). Washington, DC, American Society for Microbiology, 1991, pp 1226-1288 54. Sung JY, Costerton JW, Shaffer EA: Bacteriostatic activities of bile salts of different hydrophobicity and in the presence of phospholipid. Hepatology 14:262A, 1991 (abstract) 55. Laskin DL: Nonparenchymal cells and hepatotoxicity. Semin Liver Dis 10:293-304, 1990 56. Richman LK, Klingenstein J, Richman JA: The murine Kupffer cell I: Characterization of the cell serving accessory function in antigen-specific T cell proliferation. J Immunol 123:2602-2609, 1979 57. Kirn A, Koehren F, Steffan AM: Interferon synthesis in primary culture of Kupffer and endothelial cells from the rat liver. Hepatology 2:676, 1982 58. Manifold IH, Triger DR, Underwood JCE: Kupffer cell depletion in chronic liver disease: Implications for hepatic carcinogenesis. Lancet 2:431-433, 1983 59. Katz S, Grosfeld JL, Gross K, Plager DA, Ross D, Rosenthal RS, Hull M, Weber TR: Impaired bacterial clear. ance and trapping in obstructive jaundice. Ann Surg 199:1420, 1984 60. Tanaka N, Ryden S, Bergqvist L, Christensen P, Bengmark S: Reticulo-endothelial function in rats with obstructive jaundice. Br J Surg 72:946-949, 1985 61. Holman JM, Rikkers LF: Biliary obstruction in host defense failure. J Surg Res 32:208-213, 1982 62. Imawari M, Hughes RD, Gove CD, Williams R: Fibronectin and Kupffer cell function in fulminant hepatic failure. Dig Dis Sci 30:1028-1033, 1985
695
SUNG ET AL 63. Takiguchi S, Koga A: Effects of bile acids and endotoxin on the function and morphology of cultured hamster Kupffer cells. Virchow Arch B Cell Pathol 54:303-311, 1988 64. Tanikawa K, Sata M, Kumashiro R, Kawahara T: Kupffer cells function in cholestasis. In Cells of the Hepatic Sinusoid. E Wisse, DL Knook, K Decker (eds). Kupffer Cell Foundation, 1989, pp 288-292 65. Van Bossuyt H, Desmaretz C, Gaeta GB, Wisse E: The role of bile acids in the development of endotoxemia during obstructive jaundice in the rat. J Hepatol 10:274-279, 1990 66. Kagnoff MF: Immunology of the digestive system. In Physiology of the Gastrointestinal Tract. LR Johnson (ed). New York, Raven Press, 1987, pp 1699-1728 67. Brown WR, Kloppel TM: The liver and IgA: Immunological, cell biological and clinical implications. Hepatology 9:763784, 1989 68. Fisher MM, Nagy B, Bazin H, Underdown BJ: Biliary transport of IgA: Role of secretory component. Proc Natl Acad Sci USA 76:2008-2012, 1979 69. Orlans E, Peppard J, Fry JF, Hinton RH, Mullock BM: Secretory component as the receptor for polymeric IgA on rat hepatocytes. J Exp Med 150:1577-1581, 1979 70. Delacroix DL, Furtado-Barreira G, De Hemptinne B, Dive C, Vaerman JP: The liver in the IgA secretory immune system. Dogs but not rats and rabbits, are suitable models for human studies. Hepatology 3:980-988, 1983 71. Daniels CK, Schmucker DL: Secretory componentdependent binding of immunoglobulin A in the rat, monkey and human: A comparison of intestine and liver. Hepatology 7:517-521, 1987 72. Nagura H, Smith PD, Nakane PK, Brown WR: IgA in human bile and liver. J Immunol 126:587-595,1981 73. Sugura H, Nakanuma Y: Secretory component and immunoglobuling in the intrahepatic biliary tree and peribiliary gland in normal livers and hepatolithiasis. Gasiroenterol Jpn 24:308-314, 1989 74. Daniel CK, Schmuchker DL, Jones AL: Hepatic asialoglycoprotein receptor-mediated binding of human polymeric immnnoglobulin A. Hepatology 9:229-234, 1989
696
75. Brown WR, Kloppel TM: The role of the liver in translocation of IgA into the gastrointestinal tract. Immunol Invest 18:269-285, 1989 76. Freter R: Mechanisms of action of intestinal antibody in experimental cholera. II. Antibody-mediated antibacterial reaction at the mucosal surface. Infect Immun 2:556-562, 1970 77. Fubara ES, Freter R: Protection against enteric bacterial infection by secretory IgA antibodies. J Immunol 111:395403, 1973 78. Riepenhoff-Talty M, Bogger-Goren S, Li P, Carmody PJ, Barrett HJ, Ogra PL: Development of serum and intestinal antibody response to rotavirus after naturally acquired rotavirus infection in man. J Med Virol 8:215-222, 1981 79. Kleinman RE, Harmatz PR, Walker WA: The liver: An integral part of the enteric mucosal immune system. Hepatology 2:379-384, 1982 80. You XY: Secretory immunoglobulin A levels of the bile in cholelithiasis. Chung-Hua-Wai-Ko-Tsa-Chih 27:141-143, 1989 81. Ohshio G, Manabe T, Tobe T, Uoshioka H, Hamashima Y: Circulating immune complex endotoxin, and biliary infection in patients with biliary obstruction. Am J Surg 155:343-347, 1988 82. You XY: Clinical significance of serum secretory immunoglobulin A in cholelithiasis and biliary tract obstruction. Chung-Hua-Wai-Ko-Tsa-Chih 27:550-553, 1989 83. Emancipator SN, Gallo GR, Razaboni R, Lamm ME: Experimental cholestasis promotes the deposition of glomerular immunoglobulin A-immune complex. Am J Pathol 113:19-26, 1983 84. Abramowsky CR, Christiansen DM: Secretory immunoglobulin deposits in renal glomeruli of children with extrahepatic biliary atresia: Studies in a human counterpart of experimental ligation of the bile duct. Hum Pathol 18:1126-1131, 1987 85. Ohshio G, Mamabe T, Tamura K, Kudo H, Yoshioka H, Tobe T: Effects of percutaneous transhepatic biliary drainage on blood-bile permeability and selective IgA transport in patients with biliary obstruction. Ann Surg 211:428-432, 1990
Digestive Diseases and Sciences, Vol. 37, No. 5 (May 1992)
REVIEW ARTICLE
Defense System in the Biliary Tract Against Bacterial Infection J.Y. SUNG, J.W. COSTERTON, and E.A. SHAFFER
Bacteria can invade the biliary tract by ascending from the duodenum and via the hematogenous route from the hepatic portal venous blood. The sphincter of Oddi, situated at the junction of the biliary tract and the upper gastrointestinal tract, forms an effective mechanical barrier to duodenal reflux and ascending bacterial infection. Conversely, Kupffer cells and the tight junctions between hepatocytes help prevent bacteria and toxic metabolites from entering the hepatobiliary system from the portal circulation. The continuous flushing action of bile and the bacteriostatic effects of bile salts keeps the biliary tract sterile under normal conditions. Secretory immunoglobulin A (slgA), the predominant immunoglobulin in the bile, and mucus excreted by the biliary epithelium probably function as antiadherence factors, preventing microbial colonization. When barrier mechanisms break down, as in surgical or endoscopic sphincterotomy and with insertion of biliary stents, pathogenic bacteria enter the biliary system at high concentrations and take up residence on any foreign bodies. Intrabiliary pressure is a key factor in the development of cholangitis. Chronic biliary obstruction raises the intrabiliary pressure. This adversely influences the defensive mechanisms such as the tight junctions, Kupffer cell functions, bile flow, and slgA production in the system, resulting in a higher incidence of septicemia and endotoxemia in these patients. Knowledge of biliary defense against infection is still quite primitive. Unclear are the roles of slgA in the bile, mechanism of bacterial adhesion to the biliary epithelium, Kupffer cell function in biliary obstruction, and the antimicrobial activity of bile salts. KEY WORDS: defense mechanism; biliary tract; bacteria.
Human bile is sterile under normal conditions (1-3), yet bacteria exist on either side of the hepatobiliary system. The biliary tract exits into the duodenum, which is periodically colonized. The duodenal microorganisms are believed to be a major source of infection in cholangitis. Conversely, the liver, where bile forms, receives 70% of its blood supply from the portal vein, which also periodically delivers bacteria from the intestine. The portal vein hence constitutes another source of infection for the
biliary system. To guard against the invasion of bacteria, the biliary tract has been equipped with several defense mechanisms, which include (1) anatomical barriers (tight junctions between hepatocytes and the sphincter of Oddi), (2) physical mechanisms (bile flow and biliary mucus), (3) chemical factors (bile salts), and (4) immunological defense (Kupffer cells and immunoglobulin). This article is intended to review the effectiveness of these defense mechanisms in the biliary system.
Manuscript received June 5, 1991; revised manuscript received December 11, 1991; accepted December 16, 1991. From the Department of Biological Sciences and Department of Medicine, University of Calgary, Canada. Address for reprint requests: Dr. J.Y. Sung, Department of Biological Sciences, University of Calgary, Calgary, Alberta, Canada T2N 1N4.
BACTERIOLOGY OF THE BILIARY TRACT Microorganisms generally colonize the mucosal lining of the human gut and other hollow organs exposed to the environment. These bacteria are embedded in the glycoprotein covering of mucus as
Digestive Diseases and Sciences, Vol. 37, No. 5 (May 1992)
0163-2116/92/0500-0689506.50/09 1992PlenumPublishingCorporation
689
SUNG ET AL a sessile population (4, 5). The biliary tract of a healthy individual does not usually harbor bacteria. Studies of gallbladder bile aspirated from patients not suffering from biliary disease suggest that few planktonic bacteria exist in the biliary tract (1, 2). In the feline biliary tract, the bile duct mucus does not harbor microorganisms, except occasionally at the sphincter of Oddi (3), By implanting sterile foreign bodies in the gallbladder as a "bacterial trap," we have demonstrated that there is a transient passage of bacteria in the biliary tract of cats (6). Most reports on biliary organisms show a predominance of alimentary organisms such as the coliforms, Streptococci spp., and Klebsiella spp. (7-10). The importance of anaerobic organisms in biliary infections has received more attention in the past decade. Anaerobes have been associated with chronic biliary diseases and in biliary-intestinal anastomosis (1l, 12).
tissue that yielded positive cultures for bacteria (18). In acute cholangitis, some 40% of patients have bacteria isolated from the portovenous blood (19). Experimental infusion of specific bacteria into the portal circulation results in these bacteria appearing in bile (20, 21), These findings suggest that there is periodic delivery of bacteria from the intestine into the portal circulation; normally this hematogenous source is cleared by the phagocytes in the liver. The entry of bacteria via the portal blood seems to be facilitated by a raised intrabiliary pressure (21). Thus, it appears that bacteria may pass from the gastrointestinal tract into the biliary tract either by an ascending route, as in the case of sphincterotomy, or by the hematogenous route via the portal blood. To guard against bacterial invasion from duodenal reflux and from portal venous bacteraemia, the body has developed the following defense mechanisms.
SOURCE OF BACTERIA IN BILIARY INFECTION
MECHANICAL BARRIERS
The portal of entry of bacteria into the bile remains controversial but ascending infection has received most prominence: In patients with different biliary tract problems, positive bile cultures are most commonly found in those with ductal stones or carcinoma of the ampulla, which produce intermittent obstruction to bile flow (7). Another increasingly common mechanism for bacterial ascent results from disruption of the sphincter of Oddi following endoscopic sphincterotomy (13) or surgical choledochotomy (14). Here, bacteria presumably carried by duodenal reflux are frequently found in the common duct bile and persist for months after the procedures. Tumors obstructing the common bile duct are not usually complicated by cholangitis until after endoscopic or surgical interventions (15). The fixed obstruction presumably prevents the ascent of bacteria from the duodenum. The other potential route of bacterial invasion into the biliary tract is via the portal venous blood. In typhoid carriers, Salmoneila spp. harbored in the gallbladder enter the biliary tract as a consequence of transient portovenous bacteremia that occurs during the primary intestinal disease (!6). Transient portal vein bacteremia may even be a "normal" occurrence. Schatten et al, in a study of portal blood cultures, obtained positive results in 32% of patients undergoing upper abdominal surgery (17). In patients with nonbiliary diseases, 17% had liver 690
Tight Junctions Between Hepatocytes. Within the liver lobule, hepatocytes are arranged in plates bathed on each side by plasma. The lateral surfaces of the hepatocytes hold onto one another by desmosomes. Bile canaliculi are formed as blind-ended grooves between two adjacent hepatocytes. It is an irregular space with microvilli projecting from the side wall of the hepatocytes. The tight junctions situated on each side of the bile canaliculi seal these blind sacs from the liver sinusoids (Figure 1). The permeability of the tight junction is regulated by various factors such as bile acids (22), estrogen (23), and intrabiliary pressure (24). The integrity of the tight junction is important in preventing the entry of bacteria from the sinusoidal blood into the biliary tract. Electron microscopic studies have shown that an increased intrabiliary pressure causes morphological distortion of the tight junctions (25, 26). The cylindrical strands in the tight junction are decreased and large proteins, such as horseradish peroxidase, can penetrate the tight junction and easily move from blood to bile, reflecting disruption of the permeability barrier (26). When pressure in the biliary tract is raised to a certain critical level, tight junctions break down and the blood-bile barrier is compromised. From clinical experience (27) and animal experiments (28), an increased biliary pressure is an essential element leading to septicemia in cholangitis. Cholangiovenous reflux has been demonstrated in the presence of a high intrabiliary Digestive Diseases and Sciences, Vol. 37, No. 5 (May 1992)
BILIARY TRACT DEFENSE AGAINST INFECTION
~
id
ucts attaching to a surface, develops rapidly on these stents causing occlusion of the lumens. The importance of an intact sphincter of Oddi in preventing duodenal bacteria from invading the biliary tract is supported by clinical observations (7, 13, 14). PHYSICAL MECHANISMS
Sinusoid
Fig I. Single-cell-thick hepatocyte is separated from sinusoidal blood on either side by endothelial cells (E). Lining endothelial cells have a large fenestration (f) in the cytoplasm and lack a basal membrane, which allow free access of plasma to space of Disse (D). Kupffer cells (K) are situated at the sinusoidal surface of the endothelium acting as a phagocyte. The canalicular membranes of adjacent hepatocytes are joined by tight junctions (T), which seal the bile canaliculus (C) from the lateral intercellular space and sinusoidal space.
pressure (27). Relieving the intrabiliary pressure in obstructive jaundice is thus a critical measure to control infection in these patients (27, 29). Further, the entry of bacteria into the biliary tract via portovenous blood is facilitated by an increased intrabiliary pressure (20, 21). In chronic biliary obstruction, the liver also exhibits t r a n s c e l l u l a r regurgitation of bile into the systemic blood through the hepatocytes (30). Entry of bacteria may occur anywhere along the biliary epithelium; damage to the tight junctions allowing access via the paracellular pathway is likely the major factor causing the breakdown of the blood-bile barrier. Sphincter of Oddi. Motility studies using endoscopic manometry (31) and cineradiography (32) have shown that the sphincter of Oddi is more than just a passive conduit for the common bile duct to exit into the duodenum. The basal tone of the sphincter and any phasic contractions control bile delivery into the duodenum; this high pressure zone prevents reflux of duodenal contents into the biliary tract. The sphincter of Oddi, which separates the colonized duodenum from the uncolonized biliary tract acts as a mechanical barrier to microbial colonization (3). Animals experiments have repeatedly shown that disrupting the barrier by placing biliary stents across the sphincter into the lumen of duodenum would allow bacteria ascending into the biliary tract (33, 34). Bacterial biofilm, a complex association of microorganisms and microbial prodDigestive Diseases and Sciences, Vol. 37, No. 5 (May 1992)
Bile Flow. Canalicular bile flow results from the active secretion of solutes, followed osmotically by obligatory water flow. Bile salts, being the most abundant organic anions in the bile, are considered the major driving force in bile formation, the socalled "bile salt-dependent flow" (BSDF). A linear relationship exists between bile flow and bile salt secretion in the liver (35, 36). Extrapolation of the line relating bile flow to bile salt secretion defines a component of canalicular flow that theoretically would be present if no bile salts were secreted. This bile salt-independent flow (BSIF) represents about 50% of total canalicular bile flow in human. An average of 800-1000 ml of bile in humans effectively flushes the bile ducts throughout the day. Therefore, like the urine flushing the urinary tract continuously and menstrual flow cleansing the uterine cavity during every menstrual cycle, the physical movement of bile hinders bacteria from colonizing the biliary mucosa. In biliary obstruction, bile flow and bile salt secretion declined in parallel (37), indicating that a raised biliary pressure suppresses the BSDF. Conversely, gram-negative septicemia from urinary tract infection or other intraabdominal sepsis is often associated with intrahepatic cholestasis without biliary obstruction (38, 39). Lipopolysaccharide (LPS) components purified from the outer membrane of gram-negative bacteria exerts a cholestatic effect in the isolated perfused rat liver by inhibiting the BSIF of bile (40, 41). The decline in bile production due to suppression of BSDF and BSIF reduces the flushing effects of bile and hence might predispose to biliary infection in these patients. Mucus. The extrahepatic bile ducts are lined by a tall columnar mucus-secreting epithelium. Mucus forms a water-insoluble gel adherent to the mucosal surfaces, providing a stable unstirred layer of water between the gastrointestinal mucosa and the lumen. The gastrointestinal mucus plays an important role in excluding pathogenic microorganisms by forming a barrier to pathogens in the lumen, coaggregating the microorganisms, yet selecting the normal bac-
691
S U N G ET A L
fatty acids, leading to the formation of brown pigment stones (45, 46) or cholesterol microcrystals for cholesterol gallstone formation. CHEMICAL FACTORS
Fig 2. Transmission electron micrograph from a section of brown pigment stone showed that bacteria were surrounded by ruthenium red-stained material, compatible with bacterial glycocalyx, forming microcolonies. Bar = 1 ~m.
terial flora and retaining the secretory immunoglobulin A (slgA) within the mucus to permit specific immune protection (42). Although true for the intestine, the physiological role of mucus in the biliary tract is largely unexplored. Bile mucin is known to be involved in the formation of cholesterol gallstones (43, 44), but its antibacterial activity remains to be investigated. Mucin might also play an important role in the pathogenesis of brown pigment stones in the biliary tract. Morphological studies of these stones revealed that bacterial microcolonies are surrounded by highly hydrated carbohydrate material forming a bacterial biofilm (45, 46) (Figure 2). The carbohydrate coating on the surface of these bacteria, called glycocalyx, is doubtlessly bacterial in origin. Bacterial glycocalyx and mucin in the bile may even promote coaggregation of bacteria that could then act as a nidus for the crystallization of unconjugated bilirubin and
692
Bile Salts. Bile salts possess inhibitory effects on the proliferation of enteric microorganisms and thus might aid in preventing infections of the biliary tract (47) and control the bacterial flora of the gastrointestinal tract (48, 49). Both conjugated and unconjugated bile salts, at physiological concentrations, have been shown to inhibit E. coli, Klebsiella spp., and Enterococcus spp. in an in vitro experiment (47). In general, the inhibitory effect seems to be more significant for gram-positive bacteria than gram-negative bacteria (50, 51). The inhibitory effects of bile salts has been used in selective media to isolate specific organisms such as the SS medium for salmonellae and shigellae and the bacteroides bile esculin medium for Bacteroides fragilis (52, 53). The bacteriostatic activity of bile salt is related to the hydrophobicity and the detergent properties of these molecules. Therefore, bile salts with fewer hydroxyl group(s) and only a-hydroxylation are more potent than those with more hydroxyl group(s) and with [3-hydroxylation in suppressing the growth of bacteria (54). In more concentrated gallbladder bile, bile salts do not only exist as free ions but form mixed micelles with phospholipid and cholesterol. Vesicles of phospholipid and cholesterol also exist in the bile. As the hydrophobic component of bile salts are engaged in these complex micellar structures, the detergent activity of these molecules is altered. The antibacterial activity of even the most toxic bile salt in the micellar solution has been shown to be much attenuated (54). Hence, while the in vitro antibacterial effects of bile salts is well recognized, their role in defending against biliary infection in in vivo conditions might be overemphasized. IMMUNOLOGICAL DEFENSE Kupffer Cells. The functions of Kupffer cells and their role in the defense against biliary infection is complex (55). Kupffer cells represent about 30% of the sinusoidal cells in the liver and constitute approximately 80% of all fixed macrophages in the body (55). They are strategically located inside the liver sinusoids to filter toxic substances, endotoxins, and bacteria that come in through the portal circulation by phagocytosis (Figure 1). Besides actDigestive Diseases and Sciences, Vol. 37, No. 5 (May 1992)
BILIARY TRACT DEFENSE AGAINST INFECTION ing as scavengers, Kupffer cells also function as antigen-presenting cells for the induction of T-lymphocyte response (56). They are capable of releasing cytokines such as intefleukins, interferons, and tumor necrosis factor (55, 57). The reticuloendothelial system in the liver is impaired in chronic liver diseases (58) and in biliary obstruction (59-61). The integrity of Kupffer cell function is an important factor affecting the mortality from endotoxemia in biliary obstruction (60). There are at least three possible mechanisms leading to the failure of the reticuloendothelial system of the liver. First, in chronic liver diseases, the number of Kupffer cells in the liver is decreased (58). The reduced Kupffer cell population accounts for the depressed phagocytic activity in chronic liver failure. A second possible mechanism of impaired Kupffer cell function relates to fibronectin, a carbohydrate-binding protein produced primarily by hepatocytes and endothelial cells in the liver. Fibronecfin acts as an opsonin for a variety of targets, and fibronectin receptors are present on Kupffer cells. The hepatic production of lectin is reduced and so opsonin-mediated phagocytosis is suppressed in chronic liver disease (62). Finally, in biliary obstruction, clearance of both E. coli (55) and 99mTc-labeled sulfur colloid (60) injected into the circulation is significantly impaired. The proposed mechanism of depressed Kupffer cell function in this situation is due to an elevated plasma concentration of bile salts. In vitro studies have shown that both conjugated and unconjugated bile acids inhibit Kupffer cells phagocytosis (63, 64) and the binding of lipopolysaccharides (65). Differences likely exist depending upon the hydrophobichydrophilic balance of the bile salt molecule, but, this remains unexplored. Secretory ImmunologicalResponses. The predominant immunoglobulin in bile is secretory immunoglobulin A (sIgA) (66, 67). Immunoglobulin M (IgM) and immunoglobulin G (IgG) exist in much lower levels. In some animals (rats, mice, and rabbits), polymeric IgA produced by the lymphoid tissue of the gut enters the serum and is transported to bile by an active receptor-mediated mechanism (67). Secretory component (SC), a glycoprotein synthesized and expressed on the sinusoidal surface of the hepatocytes, binds polymeric IgA in the sinusoidal blood. This is followed by endocytosis and transportation into the bile canaliculi as sIgA (68, 69). In other animals (dog, guinea, sheep), as well as in humans, there is much less hepatic transport of Digestive Diseases and Sciences, Vot. 37, No. 5 (May 1992)
circulating IgA because of the lack of SC in their hepatocytes (70, 71). Transport of IgA into human bile occurs across the biliary epithelium where the SC is situated (72, 73). A SC-independent asialoglycoprotein receptor-mediated mechanism has been shown to provide an alternative IgA transport system across human hepatocytes (74, 75). Plasma cells forming minor glands in or adjacent to the wall of the human bile duct also synthesize IgA. The local production of IgA by these cells accounts for 50% of IgA in human bile (67). The biological functions of biliary IgA are unclear. In vitro studies have shown that the Fc receptors of sIgA are not involved in complement fixation, opsonization, or antibody-dependent cellmediated cytotoxicity (ADCC) reactions. Thus the source of IgA on mucosal surfaces does not function as a bactericidal factor. One important function of IgA in the intestine is to enhance the relative impermeability of the epithelium, by binding to antigens in the gastrointestinal lumen (76, 77). sIgA can bind and thus prevent bacteria (76, 77) and enteric virus (78) from attaching to and penetrating intestinal epithelial cells. In humans, the importance of slgA is largely unknown. Certainly, patients with selective IgA deficiency do not have an increased frequency of bacterial infection of the biliary tract. In the absence of sIgA, due to immaturity of the mucosal lymphoid system in infancy or specific IgA deficiencies, enteric antigens are found in the systemic circulation in high concentrations (79). In patients suffering from cholestasis due to biliary obstruction, the excretion of IgA by the liver is impaired and serum IgA levels elevated (80, 81). The serum IgA-immune complex (IgA-CIC) concentrations rise; this immune complex is deposited in the glomeruli in biliary obstruction after experimental ligation of the common bile duct (82) and in biliary atresia (83). Ohshio et al measured the serum levels of IgA-CIC and the corresponding endotoxin levels in 33 patients suffering from biliary obstruction and cholangitis (81). The endotoxin levels correlated with the amount of IgA-CIC existing in the blood. Impaired IgA transport might contribute to the entry of endotoxin into the systemic circulation and septicemic shock. When biliary obstruction is relieved, serum slgA returns to normal (81, 84). In a clinical study of 71 patients suffering from cholelithiasis, bile slgA levels of patients with brown pigment stones were significantly lower than the control group (85). Among the cholelithiasis pa-
693
SUNG ET AL TABLE 1. SUMMARYOF DEFENSEMECHANISMSIN B1LIARY SYSTEM Ascending infection (duodenal reflux)
Defense sphincter of Oddi mechanisms
Hematogenous infection (portal bacteraemia)
tight junction* Kupffer cell phagocytosis*
hydrophobic bile salts* flushing effect of bile* ? slgA* ? mucus *When the intrabiliary pressure is raised during biliary obstruction, this defense mechanism will be impaired. tients, bile s l g A values in t h o s e with active cholangitis w e r e l o w e r than those w i t h o u t infection. T h e a u t h o r suggested that the antibacterial a d h e r e n c e effect o f s l g A is i m p o r t a n t in avoiding microbial colonization o f the biliary epithelium. T h e f u n c t i o n o f this p r e d o m i n a n t s e c r e t o r y i m m u n o g l o b u l i n in the biliary tract n e e d s further investigation.
SUMMARY T h e central elements fostering the d e v e l o p m e n t o f cholangitis are bacteria, biliary stasis, and an increase in intrabiliary pressure. A l t h o u g h bacteria are essential for the d e v e l o p m e n t o f cholangitis, not all patients with b a c t e r i a in the bile d e v e l o p cholangitis. I n t e r r u p t i o n o f bile flow and h e n c e the loss o f c o n t i n u o u s m e c h a n i c a l flushing o f the biliary s y s t e m is crucial in the p a t h o g e n e s i s o f cholangitis. W h e n the bile flow is o b s t r u c t e d , the raised intrabiliary p r e s s u r e s u p p r e s s e s the secretion o f bile salt and f u r t h e r r e d u c e s bile flow. T h e p r o t e c t i v e functions o f the tight j u n c t i o n s , K u p f f e r cells, and s e c r e t o r y i m m u n o g l o b u l i n will also be affected. A n i n c r e a s e d biliary p r e s s u r e is thus a k e y f a c t o r in the d e v e l o p m e n t o f septicemia in cholangitis, O b s t r u c tion to bile flow will result in biliary tract infection e v e n t h o u g h all o t h e r c o m p o n e n t s o f the defense (immunoglobulins, m u c u s and bile salts) a p p e a r to be intact.
ACKNOWLEDGMENTS This work is supported by Izaak Walton Killam Memorial Scholarship, The Croucher Foundation Fellowship for Scientific Research (J. Sung), The Medical Research Council of Canada (E. Shaffer), and Natural Sciences and Engineering Research Council of Canada O.W. Costerton).
REFERENCES 1. Csendes A, Fernandez M, Uribe P: Bacteriology of the gallbladder bile in normal subjects. Am J Surg 129:629-631, 1975
694
2. Scott AJ: Bacteria and disease of the biliary tract--progress report. Gut 12:4873, 1971 3. Sung JY, Olson ME, Leung JWC, Lundberg MS, Costerton JW: The sphincter of Oddi is a boundary for bacterial colonization in the feline biliary tract. Micro Eco Health Dis 3:199-207, 1990 4. Savage DC: Microbial ecology of the gastrointestinal tract. Annu Rev Microbiol 31:107-133, 1977 5. Costerton JW, Cheng KJ: Colonization of tissue surfaces by autochthonous bacteria. In Microbiology. D Schlessinger (ed). Washington, DC, ASM publication, 1982, pp 266-273 6. Sung JY, Leung JWC, Olson ME, Lundburg MS, Costerton JW: Demonstration of transient bacterobilia by foreign body implantation in the feline biliary tract. Dig Dis Sci 36:943948, 1991 7. Suzuki Y, Kobayashi A, Ohto M, Tsuchiya Y, Saisho H, Kimura K, Ono T, Okuda K: Bacteriological study of transhepatically aspirated bile: Relation to cholangiographic findings in 295 patients. Dig Dis Sci 29:109-115, 1984 8. Pitt HA, Postier RG, Cameron JL: Biliary bacteria: Significance and alterations after antibiotic therapy. Arch Surg 117:445-449, 1982 9. Keighley MRB: Microorganisms in the bile. Ann R Coll Surg Engl 59:328-334, 1977 10. Alexander-Williams J: Bacteriology of biliary disease. In Surgery of the Gallbladder and Bile Ducts. LW Way, CA Pellegrini (eds). New York, WB Saunders, 1987, pp 93-102 11. England DM, Rosenblatt JE: Anaerobes in human biliary tract. J Clin Microbiol 6:494-498, 1977 12. Finegold SM: Anaerobes in biliary tract infection. Arch Intern Med 139:1338, 1979 13. Gregg JA, Girolami PD, Carr-Locke DL: Effects of sphincteroplasty and endoscopic sphincterotomy on the bacteriologic characteristics of the common bile duct. Am J Surg 149:668-671, 1985 14. Feretis CB, Contou CT, Manoruas AJ, Apostolidis NS, Golematis BC: Long term consequences of bacterial colonization of the biliary tract after choledochotomy. Surg Gynecol Obstet 159:363-366, 1984 15. O'Conner MJ, Schwartz ML, McQuarrie DG, Summer HW: Cholangitis due to malignant obstruction of biliary flow. Ann Surg 193:341-345, 1981 16. Tynes BS, Utz JP: Factors influencing the cure of salmonella carriers. Ann Intern Med 57:871-882, 1962 17. Schatten WE, Desprez JD, Holden WD: Bacteriologic study of portal vein blood in man. Arch Surg 71:404-409, 1955 18. Dye M, Mcdonald A, Smith G: The bacterial flora of the biliary tract and the liver. Br J Surg 65:285-287, 1978 19. Ong GB: A study of recurrent pyogenic cholangitis. Arch Surg 84:199-255, 1962 20. Dineen P: The importance of the route of infection in experimental biliary tract obstruction. Surg Gynecol Obstet 19:1001-1008, 1964 21. Sung JY, Shaffer EA, Olson ME, Leung JWC, Lam K, Costerton JW: Bacterial invasion of the biliary system by way of the portal-venous system. Hepatology 14:313-317, 1991 22. LaGarde S, Elias E, Wade JB, et al: Structural heterogeneity of hepatocyte "tight junctions": A quantitative analysis. Hepatology 1:193-203, 1981 23. Forker EL: The effect of estrogen on bile formation in the rat. J Clin Invest 48:654-663, 1969 Digestive Diseases and Sciences, Vol. 37, No. 5 (May 1992)
BILIARY TRACT D E F E N S E AGAINST I N F E C T I O N 24. Easter DW, Wade JB, Boyer JL: Structural integrity of hepatocyte tight junctions. J Cell Biol 96:745-749, 1983 25. Robenek H, Herwig J, Themann H: The morphologic characteristics of intercellular junctions between normal human liver cells from patient with extrahepatic cholestasis. Am J Pathol 100:93-103, 1980 26. Devos R, Desmet VJ: Morphological changes of the junctional complex of the hepatocytes in rat liver after bile duct ligation. Br J Exp Pathol 59:220-227, 1978 27. Lygidakis NJ, Brummelkamp WH: The significance of intrabiliary pressure in acute cholangitis. Surg Gynecol Obstet 161:465-469, 1985 28. Huang T, Bass JA, Williams RD: The significance of biliary pressure in cholangitis. Arch Surg 98:629-632, 1969 29. Leung JWC, Chung SCS, Sung JY, Banez VP, Li AKC: Urgent endoscopic drainage for acute suppurative cholangitis. Lancet 1:1307-1309, 1990 30. Jones AL, Hradek GT, Schmucker DL, Underdown BJ: The fate of polymeric and secretory immunoglobulin A after retrograde infusion into the common bile duct in rats. Hepatology 4:1173-1183, 1984 31. Geenen JE, Hogan WJ, Dodds WJ, Stewart ET, Amdorfer RC: Intraluminal pressure recording from human sphincter of Oddi. Gastroenterology 78:317-324, 1980 32. Toouli J, Dodds WJ, Honda R, Sarua S, Hogan WJ, Komorowski RA, Linehan JH, Amdorfer RC: Motor function of the opossum sphincter of Oddi. J Clin Invest 71:208-220, 1983 33. Geoghegan JG, Branch MS, Costerton JW, Pappas TN, Cotton PB: Biliary stent occlude earlier if the distal tip is in the duodenum in dogs. Gastrointest Endosc 1991 (abstract) 37:257, 1991 34. Sung JY, Leung JWC, Shaffer EA, Lain K, Olson ME, Costerton JW: Ascending infection of the biliary tract after surgical sphincterotomy and biliary stenting. J Gastroenterol Hepatol (in press) 35. Scharschmidt BF, Van Dyke RW: Mechanisms of hepatic electrolyte transport. Gastroenterology 85:1199-1214, 1983 36. Boyer JL, Bloomer JP: Canalicular bile secretion in man: Studies utilizing the biliary clearance of 14C-mannitol. J Clin Invest 54:773-781, 1974 37. Straberg SM, Dorn BC, Small DM, Egdahl RH: The effect of biliary tract pressure on bile flow, bile salt secretion and bile salt synthesis in the primate. Surgery 70:140-146, 1971 38. Eley DF, Hargreaves T, Lamberth HP: Jaundice in severe infections. Br Med J 2:75-55, 1965 39. Miller DF, Irvin RW: Jaundice in acute appendicitis. Lancet 1:321-323, 1969 40. Utili R, Abernathy CO, Zimmerman HJ: Cholestatic effects of Escherichia coli endotoxin on the isolated perfused rat liver. Gastroenterology 20:248-253, 1976 41. Utili R, Abernathy CO, Zimmerman HJ: Studies on the effects of Escherichia coli endotoxin on canalicular bile formation in the isolated perfused rat liver. J Lab Clin Med 89:471-482, 1977 42. NeutraMR, Forstner JF: Gastrointestinal mucus: Synthesis, secretion and function. In Physiology of the Gastrointestinal Tract. LR Johnson (ed). New York, Raven Press, 1987, pp 975-1002 43. Carey MC, Cahalane MJ: Whither biliary sludge? Gastroenterology 95:508-523, 1988 Digestive Diseases and Sciences, Vol. 37, No. 5 (May 1992)
44. Levy PF, Smith BF, LaMont JF: Human gallbladder mucin accelerates nucleation of cholesterol in artificial bile. Gastroenterology 87:270-275, 1984 45. Stewart L, Smith AL, Pellergrini CA, Motson RW, Way LW: Pigment gallstones form as a composite of bacterial microcolonies and pigment solids. Ann Surg 206:242-250, 1987 46. Leung JWC, Sung JY, Costerton JW: Bacteriological and electron microscopy examination of brown pigment stones. J Clin Microbiol 27:915-921, 1989 47. Stewart L, Pellergrini CA, Way LW: Antibacterial activity of bile acids against common biliary tract organisms. Surg Forum 37:157-159, 1986 48. Hill MJ, Crowther JS, Drasar BS, Hawkswroth G, Aries V, Williams REO: Bacteria and the aetiology of cancer of the large bowel. Lancet 1:95-97, 1971 49. Hill MJ: Factors controlling the microflora of the healthy upper gastrointestinal tract. In Human Microbial Ecology. MJ Hill, PD Marsh (eds). Boca Raton, Florida, CRC Press, 1989, pp 57-86 50. Floch MH, Gershengoren W, Elliott S, Spiro H: Bile acids inhibition of the intestinal microflora: A function for simple bile acids? Gastroenterology 61:228-233, 1971 51. Floch MH, Gershengoren W, Diamond S, Hersh T: Cholic acid inhibition of intestinal bacteria. Am J Clin Nutr 23:8-10, 1970 52. Leifson E: New culture media based on sodium deoxycholate for the isolation of intestinal pathogens and for enumeration of colon bacilli in milk and water. J Pathol Bacteriol 40:581-586, 1953 53. Nash P, Krenz MM: Culture media. In Manual of Clinical Microbiology, 5th ed. A Balows (ed). Washington, DC, American Society for Microbiology, 1991, pp 1226-1288 54. Sung JY, Costerton JW, Shaffer EA: Bacteriostatic activities of bile salts of different hydrophobicity and in the presence of phospholipid. Hepatology 14:262A, 1991 (abstract) 55. Laskin DL: Nonparenchymal cells and hepatotoxicity. Semin Liver Dis 10:293-304, 1990 56. Richman LK, Klingenstein J, Richman JA: The murine Kupffer cell I: Characterization of the cell serving accessory function in antigen-specific T cell proliferation. J Immunol 123:2602-2609, 1979 57. Kirn A, Koehren F, Steffan AM: Interferon synthesis in primary culture of Kupffer and endothelial cells from the rat liver. Hepatology 2:676, 1982 58. Manifold IH, Triger DR, Underwood JCE: Kupffer cell depletion in chronic liver disease: Implications for hepatic carcinogenesis. Lancet 2:431-433, 1983 59. Katz S, Grosfeld JL, Gross K, Plager DA, Ross D, Rosenthal RS, Hull M, Weber TR: Impaired bacterial clear. ance and trapping in obstructive jaundice. Ann Surg 199:1420, 1984 60. Tanaka N, Ryden S, Bergqvist L, Christensen P, Bengmark S: Reticulo-endothelial function in rats with obstructive jaundice. Br J Surg 72:946-949, 1985 61. Holman JM, Rikkers LF: Biliary obstruction in host defense failure. J Surg Res 32:208-213, 1982 62. Imawari M, Hughes RD, Gove CD, Williams R: Fibronectin and Kupffer cell function in fulminant hepatic failure. Dig Dis Sci 30:1028-1033, 1985
695
SUNG ET AL 63. Takiguchi S, Koga A: Effects of bile acids and endotoxin on the function and morphology of cultured hamster Kupffer cells. Virchow Arch B Cell Pathol 54:303-311, 1988 64. Tanikawa K, Sata M, Kumashiro R, Kawahara T: Kupffer cells function in cholestasis. In Cells of the Hepatic Sinusoid. E Wisse, DL Knook, K Decker (eds). Kupffer Cell Foundation, 1989, pp 288-292 65. Van Bossuyt H, Desmaretz C, Gaeta GB, Wisse E: The role of bile acids in the development of endotoxemia during obstructive jaundice in the rat. J Hepatol 10:274-279, 1990 66. Kagnoff MF: Immunology of the digestive system. In Physiology of the Gastrointestinal Tract. LR Johnson (ed). New York, Raven Press, 1987, pp 1699-1728 67. Brown WR, Kloppel TM: The liver and IgA: Immunological, cell biological and clinical implications. Hepatology 9:763784, 1989 68. Fisher MM, Nagy B, Bazin H, Underdown BJ: Biliary transport of IgA: Role of secretory component. Proc Natl Acad Sci USA 76:2008-2012, 1979 69. Orlans E, Peppard J, Fry JF, Hinton RH, Mullock BM: Secretory component as the receptor for polymeric IgA on rat hepatocytes. J Exp Med 150:1577-1581, 1979 70. Delacroix DL, Furtado-Barreira G, De Hemptinne B, Dive C, Vaerman JP: The liver in the IgA secretory immune system. Dogs but not rats and rabbits, are suitable models for human studies. Hepatology 3:980-988, 1983 71. Daniels CK, Schmucker DL: Secretory componentdependent binding of immunoglobulin A in the rat, monkey and human: A comparison of intestine and liver. Hepatology 7:517-521, 1987 72. Nagura H, Smith PD, Nakane PK, Brown WR: IgA in human bile and liver. J Immunol 126:587-595,1981 73. Sugura H, Nakanuma Y: Secretory component and immunoglobuling in the intrahepatic biliary tree and peribiliary gland in normal livers and hepatolithiasis. Gasiroenterol Jpn 24:308-314, 1989 74. Daniel CK, Schmuchker DL, Jones AL: Hepatic asialoglycoprotein receptor-mediated binding of human polymeric immnnoglobulin A. Hepatology 9:229-234, 1989
696
75. Brown WR, Kloppel TM: The role of the liver in translocation of IgA into the gastrointestinal tract. Immunol Invest 18:269-285, 1989 76. Freter R: Mechanisms of action of intestinal antibody in experimental cholera. II. Antibody-mediated antibacterial reaction at the mucosal surface. Infect Immun 2:556-562, 1970 77. Fubara ES, Freter R: Protection against enteric bacterial infection by secretory IgA antibodies. J Immunol 111:395403, 1973 78. Riepenhoff-Talty M, Bogger-Goren S, Li P, Carmody PJ, Barrett HJ, Ogra PL: Development of serum and intestinal antibody response to rotavirus after naturally acquired rotavirus infection in man. J Med Virol 8:215-222, 1981 79. Kleinman RE, Harmatz PR, Walker WA: The liver: An integral part of the enteric mucosal immune system. Hepatology 2:379-384, 1982 80. You XY: Secretory immunoglobulin A levels of the bile in cholelithiasis. Chung-Hua-Wai-Ko-Tsa-Chih 27:141-143, 1989 81. Ohshio G, Manabe T, Tobe T, Uoshioka H, Hamashima Y: Circulating immune complex endotoxin, and biliary infection in patients with biliary obstruction. Am J Surg 155:343-347, 1988 82. You XY: Clinical significance of serum secretory immunoglobulin A in cholelithiasis and biliary tract obstruction. Chung-Hua-Wai-Ko-Tsa-Chih 27:550-553, 1989 83. Emancipator SN, Gallo GR, Razaboni R, Lamm ME: Experimental cholestasis promotes the deposition of glomerular immunoglobulin A-immune complex. Am J Pathol 113:19-26, 1983 84. Abramowsky CR, Christiansen DM: Secretory immunoglobulin deposits in renal glomeruli of children with extrahepatic biliary atresia: Studies in a human counterpart of experimental ligation of the bile duct. Hum Pathol 18:1126-1131, 1987 85. Ohshio G, Mamabe T, Tamura K, Kudo H, Yoshioka H, Tobe T: Effects of percutaneous transhepatic biliary drainage on blood-bile permeability and selective IgA transport in patients with biliary obstruction. Ann Surg 211:428-432, 1990
Digestive Diseases and Sciences, Vol. 37, No. 5 (May 1992)
