- Email: [email protected]
Sepsis increases endocytosis of endotoxin into hepatocytes
Sepsis increases endocytosis of endotoxin into hepatocytes Abbeselom P. Ghermay, MD, Sandra Brady, BA, Richard J. Havel, MD, Hobart W. Harris, MD, and Joseph H. Rapp, MD, San Francisco, Calif.
Background. Chylomicrons bind endotoxins and accelerate their clearance from plasma to the liver. This results in reduced mortality from septic shock in a rodent model. We hypothesized that the clearance of the LPS-chylomicron (LPS-CM) complex by hepatocytes is due to receptor-mediated endocytosis and that sepsis up-regulates this process. Methods. Three groups of Sprague-Dawley rats; (1) contro# (2) pretreated with 10 l~g/kg LPS 24 hours before treatment; and (3) pretreated with 17-~-ethinyl estradiol (F~, 5 mg/kg subcutaneously for 3 days), were infused with labeled I125-LPS alone or with f125-LPS bound to chylomicron. Livers were removed 2.5, 15, and 30 minutes after LPS injection, and hepatic endosomes were isolated from the liver homogenates by serial ultracent~ifugation in sucrose gradients. Results. The injection of I125-LPS-CM complexes resulted in higher levels of endosomal I125-LPS in all groups, as compared with I125-LPS alone. In addition, the endosomal uptake of I125-LPS was markedly increased by both LPS and EE pretreatments. Conclusions. These data strongly suggest a primary role for receptor-mediated endocytosis in the increased clearance of I.,PS when bound to chylomicron. In addition, exposure to LPS appears to increase the accumulation of LPS in endosomes by a mechanism similar to that of Es which is known to up-regulate receptor-mediated lipoprotein uptake. This endogenous pathway for the catabolism of endotoxins may provide a teleological explanation for the hypertriglycaidemia observed du,ing sepsis. (Surgery 1996;120:389-94.) Prom the Department of Surgery, VA Medical Center of ,San Francisco, San Francisco General Hospital, and the Cardiovascular Research Institute and Department of Medicine, University of California, ,San Francisco.
GRAM-NEGATIVESEPSISaccounts for more than 200,000 deaths per year in the United States and remains a serious clinical problem despite the advent of potent antibiotics and sophisticated medical support. Endotoxin, a lipopolysaccharide (LPS) c o m p o n e n t of the cell wall of all gram-negative bacteria, initiates a pathologic host response mediated through the elaboration of various cytokines---tumor necrosis factor; interleukins-1, -6, and -8; and platelet-activating factor. Hepatic macrophages (Kupffer cells) play a central role in this LPS-induced response. All lipoproteins bind endotoxin avidly and thus reduce its toxicity. The specific mechanism by which lipoproteins detoxify endotoxin is not fully understood. However, it appears that the lipid A moiety of endotoxin binds to the phospholipid monolayer of lipoproteins. With the lipid A portion of the molecule hidden, the lipoprotein-bound endotoxin does not in-
Presented at the Fifty-seventhAnnual Meeting of the Societyof University Surgeons, Washington, D.C., Feb. 8-10, 1996. Reprint requests:Joseph Rapp, MD, San FranciscoVAMed Ctr. (112), 4150 Clement, San Francisco, C~. 94121. Copyright 9 1996 by Mosby-YearBook, Inc. 0039-6060/96/$5.00 + 0 11/6/73847
duce cytokine production and secretion by macrophages. 1,'a Furthermore, the clearance of the endotoxin b o u n d to chylomicron appears to mirror that of the chylomicron particle. Therefore the LPS is cleared by hepatocytes and shunted away from macrophages abrogating the host response. "~The binding of LPS to chylomicrons not only increases the delivery of endotoxin to hepatocytes, it also speeds its excretion into bile. 4 This alteration in endotoxin catabolism has p r o f o u n d metabolic effects. Chylomicron infusions can significantly improve survival in rats when given as either a bolus infusion after a lethal injection o f endotoxin or when chylomicron is administered in repeated infusions in a model of polymicrobial gram-negative sepsis, s, 5 The hepatic uptake of chylomicron has been studied extensively, and several candidate receptors have been identified. Several receptors bind chylomicron remnants, which are formed after lipoprotein lipase in the periphery depletes chylomicron of most o f its triglyceride content, including the low-density lipoprotein (LDL) receptor,6, 7 the LDL receptor-related protein, s and the lipolysis-stimulated receptor. 9 There is growing evidence that the LDL receptor plays the primary role in the endocytosis of chylomicron remnants in vivo. l~ We SURGERY 389
390
Ghermay et al.
hypothesized that (1) the accelerated hepatocellular uptake of chylomicron-bound LPS is caused by the receptor-mediated endocytosis of the LPS-chylomicron (LPS-CM) complex and (2) that this receptor directs endotoxin into endosomes and eventually into bile. To examine these hypotheses we isolated hepatocyte endosomes from normal rats, rats pretreated with 17-0-ethinyl estradiol (EE) (known to increase hepatocyte LDL receptors severalfold), and rats pretreated with LPS after injection with either I12"5-LPSor I125-LPS b o u n d to chylomicron.
M E T H O D S A N D MATERIAL Chylomicron collection and CM-LPS complex preparation. The mesenteric lymph ducts of anesthetized male Sprague-Dawley rats (200 to 220 gm) were cannulated with Bolab (size V-3) tubing. "~Silicone rubber (inner diameter, 0.03 inches) tubing was used to cannulate the duodenum, which was secured with 3-0 silk sutures in a pursestring. The incision was then closed, and the animal was allowed to recover from anesthesia in a Bollman (Banton and Kingman, Inc., Fremont, Calif.) restraining cage. Mesenteric lymph production was promoted by a continuous infusion of a 0.15 m o l / L NaC1 solution containing 10% dextrose through the duodenal catheter at a rate of 10 m l / k g / h r . The lymph was collected in a sterile, endotoxin-free flask on ice for 4 to 6 hours, after which the animal was killed. To avoid contaminafion with exogenous endotoxin, all heat-stable materials used in the isolation, processing, and assay of solutions to be iniected into the rats were rendered sterile and free of detectable endotoxin (less than 5 to 10 pg/ml) as determined by a gel clot limulus assay. 11 Chylomicrons were separated from the lymph by ultracentrifugation (35,000 rpm x 1 hour) and then incubated with radiolabeled (1125) endotoxin at 37 ~ C for 1 hour. 4 Chylomicrons with bound LPS were then reisolated by ultracentrifugation to separate from u n b o u n d LPS. LPS was radiolabeled with 112"~ by the method of UleveitchJ ') Briefly, Escherichia coli (055:B5) endotoxin was derivatized by reacting with p-OH methylbenzimidate at alkaline pH and then labeled with Na-I 1'25by incubation for 30 hours with lactoperoxidase-coated beads. Male Sprague-Dawley rats (200 to 250 gin) were separated into three groups: normal (untreated), pretreated with 10 p g / k g LPS intraperitoneally 24 hours before experimentation, or pretreated with EE 5 m g / k g subcutaneously once a day for a total of 5 days before experimentation. Isolation o f endosomes. Endosomes were isolated from a low-density fraction of liver homogenates as described by Belcher et al. l's Sprague-Dawley rats in each of the previously described groups were injected through an exposed femoral vein with either I125-LPS alone or
Surgery August 1996 I125-LPS-CM complexes. 4 At specified times the livers were removed and homogenized in 0.25 m o l / L sucrose containing the following protease inhibitors: 0.11 m m o l / L antipain, 0.04 m m o l / L pepstain, 1.9 m m o l / L benzamidine, and 0.8 m g / m l bacitracin. After centrifugation of 10 minutes at 500 g, 15 minutes at 3500 g, and 20 minutes at 1200 g; the density of the remaining supernatant fraction was adjusted with Percoll (pH 7.4) and centrifuged in a Beckman (Beckman Instruments Inc., Fullerton, Calif.) 50.3 Ti rotor for 45 minutes at 29,900 g: The gradient was harvested down to marker beads of density 1.062 gm/ml. Two volumes of ice-cold 0.15 m o l / L NaC1 were added, and the fraction layered onto 2 ml 2.5 m o l / L sucrose. The tubes were then centrifuged at 17,800 g in a Beckman (Beckman Instruments Inc.) SW 41 rotor for 45 minutes. The white endosome bands were removed from the sucrose cushions, and the density was raised to 1.15 g m / m l with 2.5 m o l / L sucrose. A discontinuous sucrose gradient was prepared by successively layering 2.0 ml each of sucrose solutions with densities of 1.032, 1.074, 1.11, and 1.13 gm/ml. A portion of the endosome fraction was layered at the bottom of each tube, which was then centrifuged in a Beckman (Beckman Instruments Inc.) SW 41 at 197,500 gfor 90 minutes. A distinct white band was seen at each of the three gradient interfaces. Each of these fractions is a fairly pure concentration of early endosomes, late endosomes, and receptor recycling compartments as shown by electron microscopy./"s 1125in each of the three endosomal tiactions was then assayed by scintillation spectrometry. Statistics. Statistical significance was determined by Student's t test in all reported data. Reported data are presented as mean -+ SEM.
RESULTS Although Kupffer cells represent less than 3% of hepatocellular mass, they normally account for a much larger fiaction of the liver's clearance of endotoxin. To determine whether Kupffer cell endosomes accounted for a significant fraction of LPS in our endosomai preparations, we compared infusions of endotoxin and chylomicrons with and without pretreatment of rats with Gadolinium HCI, known to selectively eliminate Kupffer cells. We found that recovery of 1125 in endosomes was identical (1.79 -+ 0.32 and 1.84 -+ 0.27, respectively, n = 3) for each group. As shown in Fig. 1, the endosomal II25-LPS recovery of normal rats infused with LPS-CM complexes was lower 2.5 minutes after injection than that of normal rats infused with LPS alone. However, by 15 minutes after injection the rats infused with LPS-CM complexes contained more I125-LPS than those injected with LPS alone. This trend continued so that by 30 minutes endosomes from rats injected with LPSCM complexes contained more than twice as much II25-LPS (p -< 0.001) as those receiving LPS alone. The pattern of endosomal uptake
Surgery Volume 120, Number 2
Ghermay et al.
391
5-
4"
3-
[] Endotoxin Alone
Endosomal Endotoxin
[] Endotoxin-Chylomicron
2-
1-
2.5 min
15 rain
30 min
Time Fig. 1. Endosomes isolated from normal rats 2.5, 15, and 30 minutes after infusion of either endotoxins alone or endotoxins preincubated with chylomicrons (n = 3 in each case). Percentage of endosomal endotoxin expressed as percentage of liver homogenate (p -< 0.001 at 30 minutes).
6 =
,-
]-
TI
4-
Endosomal Endotoxin
[]
NormalRats
[]
Septic Rats
3-
2"
0
I
!
2.5 rain
15 rain
30 rain
Time Fig. 2. Endosomes isolated from normal versus septic rats 2.5 (p -< 0.03), 15 (p < 0.04), and 30 minutes after infusion of endotoxin preincubated with chylomicrons (n = 3 in each case). Percentage of endosomal endotoxin expressed as percentage of liver homogenate. o f LPS b o u n d to chylomicron generally followed that seen with chylomicron alone, 14 in keeping with o u r hypothesis. To d e t e r m i n e whether this process is up-regulated by p r e t r e a t m e n t o f rats with a sublethal dose o f endotoxin, we c o m p a r e d the e n d o s o m a l recovery of II25-LPS in normal rats a n d in rats p r e t r e a t e d with LPS after infusion o f CM-LPS (Fig. 2). T h e e n d o s o m a l 1125LPS recovery o f the LPS-pretreated rats was significantly h i g h e r than in the n o r m a l rats at 2.5 minutes (p --- 0.03) a n d 15 minutes (p -< 0.04) after injection. By 30 minutes the endosomal recovery rates were equal in the two groups (Fig. 2). Similarly, rats p r e t r e a t e d with EE h a d a
significantly increased uptake of e n d o t o x i n as comp a r e d with u n t r e a t e d rats 2.5 a n d 15 minutes after injection, b u t the difference was insignificant at 30 minutes (Fig. 3). T h e e n d o s o m a l recovery o f I125-LPS p e a k e d by 15 minutes in both the LPS-pretreated a n d EE-treated rats. To c o m p a r e the uptake of free LPS a n d LPS-CM in animals previously exposed to endotoxin, we m e a s u r e d e n d o s o m a l uptake in both groups. As expected, there was a significant increase in the e n d o s o m a l recovery o f II~5-LPS when it was b o u n d to chylomicron (p ~ 0.004) in rats p r e t r e a t e d with LPS (Fig. 4).
392
Ghermay et al.
Surgery August 1996
-]-
8-
6-
Endosomal Endotoxin
[]
Normal Rats
[]
Ethinyl-Estradiol Treated
4-
2-
2.5 rain
15 rain
30 rain
Time Fig. 3, Endosomes isolated from normal rats versus rats pretreated with EE 2.5 (p -< 0.01), 15 (p ~ 0.01), and 30 minutes after infusion of endotoxin preincubated with chylomicrons (n = 3 in each case), Percentage of endosomal endotoxin expressed as percentage of liver homogenate.
Endosomal Endotoxin
2.5 min
15 min
[]
Endotoxin Infused
[]
Endotoxin-Chylomicmn
30 rain
Time Fig. 4. Endosomes isolated from septic rats 2.5, 15 (p -< 0.004), and 30 minutes (p <- 0.004) after infusion of either endotoxin alone or endotoxin preincubated with chylomicrons (n = 3 in each case). Percentage of endosomal endotoxin expressed as percentage of liver homogenate. DISCUSSION
There is growing evidence that lipoproteins serve as c o m p o n e n t s o f a nonspecific i m m u n e response to injury a n d infection. Previous studies 4' 5 have shown that the binding o f e n d o t o x i n to triglyceride-rich lipoproreins accelerates the clearance o f e n d o t o x i n from serum, shunts the e n d o t o x i n away from hepatic macrophages toward hepatocytes, a n d results in decreased levels o f circulating t u m o r necrosis factor. This prof o u n d alteration in LPS metabolism has been shown to
protect rats from death p r o d u c e d by b o t h LPS injections a n d polymicrobial sepsis. ~5 In this study we presented evidence that the hepatocyte clearance o f e n d o t o x i n b o u n d to chylomicrons is a r e c e p t o r - m e d i a t e d process. This process is up-regulated by LPS p r e t r e a t m e n t in a m a n n e r similar to that shown with ethinyl-estradiol. F u r t h e r studies are n e e d e d to d e t e r m i n e whether the up-regulation seen with LPS p r e t r e a t m e n t also results from an increase in hepatocyte LDL receptors. Studies of chylomicron metabolism have shown that
Surgery Volume 120, Number 2 a d i s t i n c t i o n m u s t b e m a d e b e t w e e n initial c l e a r a n c e o f r e m n a n t s by t h e liver a n d e n d o c y t o s i s i n t o h e p a t o cytes. 14 H e p a t i c lipase, g l y c o s a m i n o g l y c a n s , a n d o t h e r e x t r a c e l l u l a r m a c r o m o l e c u l e s c a p a b l e o f b i n d i n g chyl o m i c r o n s a r e all i n v o l v e d in t h e initial c l e a r a n c e o f c h y l o m i c r o n s f r o m p l a s m a . 15-17 T h e ability to isolate hepatocyte endosomes and determine their endotoxin c o n t e n t h a s e n a b l e d us to d i s t i n g u i s h b e t w e e n this initial c l e a r a n c e o f c o m p l e x e s f r o m t h e c i r c u l a t i o n by t h e liver a n d a c t u a l e n d o c y t o s i s by h e p a t o c y t e s . W h e t h e r infusions o f l i p o p r o t e i n s c a n serve as a n a d j u n c t in treating sepsis r e m a i n s unclear. E n d o g e n o u s lipoproteins could have a n ameliorating effect o n gram-negative sepsis. C h y l o m i c r o n s have a large capacity to inhibit e n d o t o x i n activity. E n d o t o x i n c o n c e n t r a t i o n s in t h e b l o o d o f patients with sepsis are r e p o r t e d to b e 80 to 250 p g / m l , as whereas t h e activity o f u p to 1000 p g o f e x o g e n o u s e n d o t o x i n c a n b e inhibited by 1 m g triglyceride o f chylomicrons in v i t r o ) ~ 2o However, t h e capacity o f t h e hepatocyte to clear the LPS-CM complexes may b e a critical factor in d e t e r m i n i n g the ultim a t e ability o f lipoproteins to r e m o v e LPS f r o m the body. T h e finding that LPS-CM complexes are cleared by a receptor-mediated process that is up-regulated by previous exposure to LPS is f u r t h e r evidence that the hypertriglyceridemia observed d u r i n g sepsis limits LPS toxicity.
REFERENCES 1. Flegel WA, Wolpl D, Mannel DN, Northoff H. hthibidon of endotoxin-induced activation of human monocytes by human lipoproteins. Infect Immun 1989;57:2237-45. 2. Cavaillon JM, Fitting C, Cawaillon NH, Warren H. Cytokine response by monocytes and macrophages to free and lipoprotein bound lipopolysaccharide. Infect Immun 1990;58:2375-82. 3. Harris HW, Grunteld C, Feingold KR, Read TE, Kane JP, Rapp JH. (3hylomicrons alter the fate of endotoxin, decreasing tumor necrosis factor release and preventing death. J Clin Invest 1993; 91:1028-34. 4. Read TE, Harris HW, Grunteld C, Feingold KR, KaneJP, Rapp JH. Cbylomicrons enhance endotoxin excretion in bile. Infect Immun 1993;61:3496-502. 5. Read TE, Grunfeld C, Feingold KR, Calhoun MC, Kumwenda Z, RappJH. Triglyceride-rich lipoproteins prevent septic death in rats. J Exp Med 1995;182:267-72. 6. Windler E, Greeve J, Daerr W, Greten H. Binding of rat chylomicrons and their remnants to the hepatic low-density-lipoprotein receptor and its role in remnant removal, BiochemJ 1988;252:55361. 7. Choi SY, Cooper AD. A comparison of the roles of the low density lipoprotein (LDL) receptor and the LDL receptor-related protein/alpha 2 macroglobulin receptor in chylomicron remnant removal in the mouse in vivo.J Biol Chem 1993;268:1580411. 8. Willnow TE, Goldstein SL, Orth K, Brown MS, HerzJ. Low density lipoprotein receptor-related protein and gp 330 bind similar ligands, including plasminogen activator-inhibitor complexes and lactoferrin an inhibitor of chylomicron remnant clearance. J Biol Cbem 1992;267:26172-80. 9. Yen FT, Mann CJ, Lydie MG, et al. Identification of a lipolysisstimulated receptor that is distinct from the LDL receptor and the LDL receptor-related protein. Biochemistry 1994;33:1172-80.
Ghermay et al.
393
10. Herz J, Qui s, Oesterle A, Desilva HV, Shaft S, Havel RJ. Initial hepatic removal of chylomicron remnants is unaffected but endocytosis is delayed in mice lacking the low density lipoprotein receptor. Proc Nail Acad Sci 1995;92:4611-5. 11. LevinJ, Tomasulo PA, Oser RS. Detection ofendotoxin in human blood and demonstration of an inhibitor. J Lab Clin Med 1970; 75:903-11. 12. Uleveitch RJ. The preparation and characterization of radioiodinated bacterial lipopolysaccharide. Immunochemistry 1978;15: 157-64. 13. Belcher JD, Hamilton RL, Brady SE, et al. Isolation and characterization of three endosomal fractions from the liver of estradiol-treated rats. Proc Nail Acad Sci 1987;84:6785-9. 14. Havel RJ. Chylomicron remnants: hepatic receptors and metabolism. Curr Opin Lipidol 1995;6:312-6. 15. Mokuno H, Brady S, Kotite L, HerzJ, Havel R. Effect of 39 K-da receptor-associated protein on the hepatic uptake and endocytosis of chylomicron remnants and low density lipoproteins in rats. J Biol Chem 1994;269:13238-43. 16.Jackson RL, Busch SJ, Cardin AD. Glycosaminoglycans: molecular properties, protein interactions, and role in physiological processes. Physiol Rev 1991;71:481-539. 17. Shaft S, Brady SE, Bensadoun S, Havel RJ. Role of hepatic lipase in the uptake of chylomicron remnants in rat liver. J Lipid Res 1994;35:709-19. 18. Obayashi T. Addition of perchloric acid to blood samples for colorimetric limulus test using chromogenic substrate: comparison with conventional procedures and clinical applications. J Lab Clin Med 1984;104:321-30. 19. Harris HW, Eichbaum EB, KaneJP, RappJH. Detection ofendotoxin in triglyceride-rich lipoproteins in vitro. J Lab Clin Med 1991;118:186-93. 20. Eichbaum EB, Harris HW, KaneJP, RappJH. Chylomicrons can inhibit endotoxin activity in vitro. J Surg Res 1991;51:413-6.
DISCUSSION Dr. Timothy R. Billiar (Pittsburgh, Pa.). Can you give us a little more information on how you determined the content of the endosome? Are you sure that the endotoxin is really in there, or has the label somehow been transferred to some other lipid moiety associated with the chylomicron? Second, in the past Dr. Rapp's group has shown that chylomicrons protect animals from sepsis and reduce cytokine release. Could this be a mechanism of endotoxin tolerance? Could you c o m m e n t o n the possibility of the increased endotoxin clearance as a mechanism of endotoxin tolerance? Finally, do you have any more information on the importance of the LDL receptor in this process? We know that LDL receptors can act as scavenger receptors for endotoxin in a n u m b e r of cell types. Do you know whether the LDL receptor is really involved? Dr. Ghermay. I shall start by answering your last question first. In terms of the LDL receptor, it is one of the receptors that has been identified as probably the physiologically important receptor among many receptors that are involved in the uptake of chylomicron remnants. We are in the process of trying to determine which receptor is determinant in the uptake of the LPS-CM complex. At this time I can say that it does not appear to be the LDL receptor in our early, preliminary studies. To answer the question of whether the radiolabel sticks to another lipid moiety in terms of going in, we radiolabeled endotoxin and then b o u n d that to the chylomicrons. The ability of radiolabel to j u m p from one moiety to another moiety is a
394
Ghermay et al.
potential. That is something that we should probably investigate; however, that does n o t appear to be the case with several previous experiments we have done in determining this. Dr. David L. Duma (Minneapolis, Minn.). First, could you describe the effect of this agent on mortality? Does this agent in a treatment mode prevent endotoxic mortality in this model? Second, have you attempted to use this agent to prevent mortality in a bacteremic or an infection-type model with or without antibiotics? Dr. Ghermay. We have looked at chylomicron infusions in both cecal ligafion and puncture with significant decrease in mortality in this model with infusions up to 30 minutes after the cecal ligafion and puncture in serial infusions. In addition, we have looked at it with lethal endotoxin injections, and we have shown that there is again significant survival benefit to this in the rat model. H u m a n studies are yet to be done by us, and antibiotics have not b e e n initiated in rats.
Surgery August 1996 Dr. Paul E. Bankey (Dallas, Texas). Do you have any information on whether Kupffer cells or liver macrophages are involved in this clearance of the lipid particles and endotoxins? We have been experimenting with liposomes, and there seem to be more liposomes taken up by the liver macrophages rather than the hepatocytes. In sepsis it seems that most of the hypertriglyceridemia is due to very low density lipoproteins. Have you looked at very low density lipoprotein particles as having a similar effect? Dr. Ghermay. In terms of very low density lipoprotein and LDL, both have been studied a n d b o t h have been shown to have similar survival benefits, both by our group and other groups. The question about the Kupffer cells was one that occurred to us during the design of the experiment. Gadolinium will selectively inhibit the function of Kupffer cells. Although not presented, we did do this a n d there was no significant change to the endosomal data with gadolinium-treated livers.
Background. Chylomicrons bind endotoxins and accelerate their clearance from plasma to the liver. This results in reduced mortality from septic shock in a rodent model. We hypothesized that the clearance of the LPS-chylomicron (LPS-CM) complex by hepatocytes is due to receptor-mediated endocytosis and that sepsis up-regulates this process. Methods. Three groups of Sprague-Dawley rats; (1) contro# (2) pretreated with 10 l~g/kg LPS 24 hours before treatment; and (3) pretreated with 17-~-ethinyl estradiol (F~, 5 mg/kg subcutaneously for 3 days), were infused with labeled I125-LPS alone or with f125-LPS bound to chylomicron. Livers were removed 2.5, 15, and 30 minutes after LPS injection, and hepatic endosomes were isolated from the liver homogenates by serial ultracent~ifugation in sucrose gradients. Results. The injection of I125-LPS-CM complexes resulted in higher levels of endosomal I125-LPS in all groups, as compared with I125-LPS alone. In addition, the endosomal uptake of I125-LPS was markedly increased by both LPS and EE pretreatments. Conclusions. These data strongly suggest a primary role for receptor-mediated endocytosis in the increased clearance of I.,PS when bound to chylomicron. In addition, exposure to LPS appears to increase the accumulation of LPS in endosomes by a mechanism similar to that of Es which is known to up-regulate receptor-mediated lipoprotein uptake. This endogenous pathway for the catabolism of endotoxins may provide a teleological explanation for the hypertriglycaidemia observed du,ing sepsis. (Surgery 1996;120:389-94.) Prom the Department of Surgery, VA Medical Center of ,San Francisco, San Francisco General Hospital, and the Cardiovascular Research Institute and Department of Medicine, University of California, ,San Francisco.
GRAM-NEGATIVESEPSISaccounts for more than 200,000 deaths per year in the United States and remains a serious clinical problem despite the advent of potent antibiotics and sophisticated medical support. Endotoxin, a lipopolysaccharide (LPS) c o m p o n e n t of the cell wall of all gram-negative bacteria, initiates a pathologic host response mediated through the elaboration of various cytokines---tumor necrosis factor; interleukins-1, -6, and -8; and platelet-activating factor. Hepatic macrophages (Kupffer cells) play a central role in this LPS-induced response. All lipoproteins bind endotoxin avidly and thus reduce its toxicity. The specific mechanism by which lipoproteins detoxify endotoxin is not fully understood. However, it appears that the lipid A moiety of endotoxin binds to the phospholipid monolayer of lipoproteins. With the lipid A portion of the molecule hidden, the lipoprotein-bound endotoxin does not in-
Presented at the Fifty-seventhAnnual Meeting of the Societyof University Surgeons, Washington, D.C., Feb. 8-10, 1996. Reprint requests:Joseph Rapp, MD, San FranciscoVAMed Ctr. (112), 4150 Clement, San Francisco, C~. 94121. Copyright 9 1996 by Mosby-YearBook, Inc. 0039-6060/96/$5.00 + 0 11/6/73847
duce cytokine production and secretion by macrophages. 1,'a Furthermore, the clearance of the endotoxin b o u n d to chylomicron appears to mirror that of the chylomicron particle. Therefore the LPS is cleared by hepatocytes and shunted away from macrophages abrogating the host response. "~The binding of LPS to chylomicrons not only increases the delivery of endotoxin to hepatocytes, it also speeds its excretion into bile. 4 This alteration in endotoxin catabolism has p r o f o u n d metabolic effects. Chylomicron infusions can significantly improve survival in rats when given as either a bolus infusion after a lethal injection o f endotoxin or when chylomicron is administered in repeated infusions in a model of polymicrobial gram-negative sepsis, s, 5 The hepatic uptake of chylomicron has been studied extensively, and several candidate receptors have been identified. Several receptors bind chylomicron remnants, which are formed after lipoprotein lipase in the periphery depletes chylomicron of most o f its triglyceride content, including the low-density lipoprotein (LDL) receptor,6, 7 the LDL receptor-related protein, s and the lipolysis-stimulated receptor. 9 There is growing evidence that the LDL receptor plays the primary role in the endocytosis of chylomicron remnants in vivo. l~ We SURGERY 389
390
Ghermay et al.
hypothesized that (1) the accelerated hepatocellular uptake of chylomicron-bound LPS is caused by the receptor-mediated endocytosis of the LPS-chylomicron (LPS-CM) complex and (2) that this receptor directs endotoxin into endosomes and eventually into bile. To examine these hypotheses we isolated hepatocyte endosomes from normal rats, rats pretreated with 17-0-ethinyl estradiol (EE) (known to increase hepatocyte LDL receptors severalfold), and rats pretreated with LPS after injection with either I12"5-LPSor I125-LPS b o u n d to chylomicron.
M E T H O D S A N D MATERIAL Chylomicron collection and CM-LPS complex preparation. The mesenteric lymph ducts of anesthetized male Sprague-Dawley rats (200 to 220 gm) were cannulated with Bolab (size V-3) tubing. "~Silicone rubber (inner diameter, 0.03 inches) tubing was used to cannulate the duodenum, which was secured with 3-0 silk sutures in a pursestring. The incision was then closed, and the animal was allowed to recover from anesthesia in a Bollman (Banton and Kingman, Inc., Fremont, Calif.) restraining cage. Mesenteric lymph production was promoted by a continuous infusion of a 0.15 m o l / L NaC1 solution containing 10% dextrose through the duodenal catheter at a rate of 10 m l / k g / h r . The lymph was collected in a sterile, endotoxin-free flask on ice for 4 to 6 hours, after which the animal was killed. To avoid contaminafion with exogenous endotoxin, all heat-stable materials used in the isolation, processing, and assay of solutions to be iniected into the rats were rendered sterile and free of detectable endotoxin (less than 5 to 10 pg/ml) as determined by a gel clot limulus assay. 11 Chylomicrons were separated from the lymph by ultracentrifugation (35,000 rpm x 1 hour) and then incubated with radiolabeled (1125) endotoxin at 37 ~ C for 1 hour. 4 Chylomicrons with bound LPS were then reisolated by ultracentrifugation to separate from u n b o u n d LPS. LPS was radiolabeled with 112"~ by the method of UleveitchJ ') Briefly, Escherichia coli (055:B5) endotoxin was derivatized by reacting with p-OH methylbenzimidate at alkaline pH and then labeled with Na-I 1'25by incubation for 30 hours with lactoperoxidase-coated beads. Male Sprague-Dawley rats (200 to 250 gin) were separated into three groups: normal (untreated), pretreated with 10 p g / k g LPS intraperitoneally 24 hours before experimentation, or pretreated with EE 5 m g / k g subcutaneously once a day for a total of 5 days before experimentation. Isolation o f endosomes. Endosomes were isolated from a low-density fraction of liver homogenates as described by Belcher et al. l's Sprague-Dawley rats in each of the previously described groups were injected through an exposed femoral vein with either I125-LPS alone or
Surgery August 1996 I125-LPS-CM complexes. 4 At specified times the livers were removed and homogenized in 0.25 m o l / L sucrose containing the following protease inhibitors: 0.11 m m o l / L antipain, 0.04 m m o l / L pepstain, 1.9 m m o l / L benzamidine, and 0.8 m g / m l bacitracin. After centrifugation of 10 minutes at 500 g, 15 minutes at 3500 g, and 20 minutes at 1200 g; the density of the remaining supernatant fraction was adjusted with Percoll (pH 7.4) and centrifuged in a Beckman (Beckman Instruments Inc., Fullerton, Calif.) 50.3 Ti rotor for 45 minutes at 29,900 g: The gradient was harvested down to marker beads of density 1.062 gm/ml. Two volumes of ice-cold 0.15 m o l / L NaC1 were added, and the fraction layered onto 2 ml 2.5 m o l / L sucrose. The tubes were then centrifuged at 17,800 g in a Beckman (Beckman Instruments Inc.) SW 41 rotor for 45 minutes. The white endosome bands were removed from the sucrose cushions, and the density was raised to 1.15 g m / m l with 2.5 m o l / L sucrose. A discontinuous sucrose gradient was prepared by successively layering 2.0 ml each of sucrose solutions with densities of 1.032, 1.074, 1.11, and 1.13 gm/ml. A portion of the endosome fraction was layered at the bottom of each tube, which was then centrifuged in a Beckman (Beckman Instruments Inc.) SW 41 at 197,500 gfor 90 minutes. A distinct white band was seen at each of the three gradient interfaces. Each of these fractions is a fairly pure concentration of early endosomes, late endosomes, and receptor recycling compartments as shown by electron microscopy./"s 1125in each of the three endosomal tiactions was then assayed by scintillation spectrometry. Statistics. Statistical significance was determined by Student's t test in all reported data. Reported data are presented as mean -+ SEM.
RESULTS Although Kupffer cells represent less than 3% of hepatocellular mass, they normally account for a much larger fiaction of the liver's clearance of endotoxin. To determine whether Kupffer cell endosomes accounted for a significant fraction of LPS in our endosomai preparations, we compared infusions of endotoxin and chylomicrons with and without pretreatment of rats with Gadolinium HCI, known to selectively eliminate Kupffer cells. We found that recovery of 1125 in endosomes was identical (1.79 -+ 0.32 and 1.84 -+ 0.27, respectively, n = 3) for each group. As shown in Fig. 1, the endosomal II25-LPS recovery of normal rats infused with LPS-CM complexes was lower 2.5 minutes after injection than that of normal rats infused with LPS alone. However, by 15 minutes after injection the rats infused with LPS-CM complexes contained more I125-LPS than those injected with LPS alone. This trend continued so that by 30 minutes endosomes from rats injected with LPSCM complexes contained more than twice as much II25-LPS (p -< 0.001) as those receiving LPS alone. The pattern of endosomal uptake
Surgery Volume 120, Number 2
Ghermay et al.
391
5-
4"
3-
[] Endotoxin Alone
Endosomal Endotoxin
[] Endotoxin-Chylomicron
2-
1-
2.5 min
15 rain
30 min
Time Fig. 1. Endosomes isolated from normal rats 2.5, 15, and 30 minutes after infusion of either endotoxins alone or endotoxins preincubated with chylomicrons (n = 3 in each case). Percentage of endosomal endotoxin expressed as percentage of liver homogenate (p -< 0.001 at 30 minutes).
6 =
,-
]-
TI
4-
Endosomal Endotoxin
[]
NormalRats
[]
Septic Rats
3-
2"
0
I
!
2.5 rain
15 rain
30 rain
Time Fig. 2. Endosomes isolated from normal versus septic rats 2.5 (p -< 0.03), 15 (p < 0.04), and 30 minutes after infusion of endotoxin preincubated with chylomicrons (n = 3 in each case). Percentage of endosomal endotoxin expressed as percentage of liver homogenate. o f LPS b o u n d to chylomicron generally followed that seen with chylomicron alone, 14 in keeping with o u r hypothesis. To d e t e r m i n e whether this process is up-regulated by p r e t r e a t m e n t o f rats with a sublethal dose o f endotoxin, we c o m p a r e d the e n d o s o m a l recovery of II25-LPS in normal rats a n d in rats p r e t r e a t e d with LPS after infusion o f CM-LPS (Fig. 2). T h e e n d o s o m a l 1125LPS recovery o f the LPS-pretreated rats was significantly h i g h e r than in the n o r m a l rats at 2.5 minutes (p --- 0.03) a n d 15 minutes (p -< 0.04) after injection. By 30 minutes the endosomal recovery rates were equal in the two groups (Fig. 2). Similarly, rats p r e t r e a t e d with EE h a d a
significantly increased uptake of e n d o t o x i n as comp a r e d with u n t r e a t e d rats 2.5 a n d 15 minutes after injection, b u t the difference was insignificant at 30 minutes (Fig. 3). T h e e n d o s o m a l recovery o f I125-LPS p e a k e d by 15 minutes in both the LPS-pretreated a n d EE-treated rats. To c o m p a r e the uptake of free LPS a n d LPS-CM in animals previously exposed to endotoxin, we m e a s u r e d e n d o s o m a l uptake in both groups. As expected, there was a significant increase in the e n d o s o m a l recovery o f II~5-LPS when it was b o u n d to chylomicron (p ~ 0.004) in rats p r e t r e a t e d with LPS (Fig. 4).
392
Ghermay et al.
Surgery August 1996
-]-
8-
6-
Endosomal Endotoxin
[]
Normal Rats
[]
Ethinyl-Estradiol Treated
4-
2-
2.5 rain
15 rain
30 rain
Time Fig. 3, Endosomes isolated from normal rats versus rats pretreated with EE 2.5 (p -< 0.01), 15 (p ~ 0.01), and 30 minutes after infusion of endotoxin preincubated with chylomicrons (n = 3 in each case), Percentage of endosomal endotoxin expressed as percentage of liver homogenate.
Endosomal Endotoxin
2.5 min
15 min
[]
Endotoxin Infused
[]
Endotoxin-Chylomicmn
30 rain
Time Fig. 4. Endosomes isolated from septic rats 2.5, 15 (p -< 0.004), and 30 minutes (p <- 0.004) after infusion of either endotoxin alone or endotoxin preincubated with chylomicrons (n = 3 in each case). Percentage of endosomal endotoxin expressed as percentage of liver homogenate. DISCUSSION
There is growing evidence that lipoproteins serve as c o m p o n e n t s o f a nonspecific i m m u n e response to injury a n d infection. Previous studies 4' 5 have shown that the binding o f e n d o t o x i n to triglyceride-rich lipoproreins accelerates the clearance o f e n d o t o x i n from serum, shunts the e n d o t o x i n away from hepatic macrophages toward hepatocytes, a n d results in decreased levels o f circulating t u m o r necrosis factor. This prof o u n d alteration in LPS metabolism has been shown to
protect rats from death p r o d u c e d by b o t h LPS injections a n d polymicrobial sepsis. ~5 In this study we presented evidence that the hepatocyte clearance o f e n d o t o x i n b o u n d to chylomicrons is a r e c e p t o r - m e d i a t e d process. This process is up-regulated by LPS p r e t r e a t m e n t in a m a n n e r similar to that shown with ethinyl-estradiol. F u r t h e r studies are n e e d e d to d e t e r m i n e whether the up-regulation seen with LPS p r e t r e a t m e n t also results from an increase in hepatocyte LDL receptors. Studies of chylomicron metabolism have shown that
Surgery Volume 120, Number 2 a d i s t i n c t i o n m u s t b e m a d e b e t w e e n initial c l e a r a n c e o f r e m n a n t s by t h e liver a n d e n d o c y t o s i s i n t o h e p a t o cytes. 14 H e p a t i c lipase, g l y c o s a m i n o g l y c a n s , a n d o t h e r e x t r a c e l l u l a r m a c r o m o l e c u l e s c a p a b l e o f b i n d i n g chyl o m i c r o n s a r e all i n v o l v e d in t h e initial c l e a r a n c e o f c h y l o m i c r o n s f r o m p l a s m a . 15-17 T h e ability to isolate hepatocyte endosomes and determine their endotoxin c o n t e n t h a s e n a b l e d us to d i s t i n g u i s h b e t w e e n this initial c l e a r a n c e o f c o m p l e x e s f r o m t h e c i r c u l a t i o n by t h e liver a n d a c t u a l e n d o c y t o s i s by h e p a t o c y t e s . W h e t h e r infusions o f l i p o p r o t e i n s c a n serve as a n a d j u n c t in treating sepsis r e m a i n s unclear. E n d o g e n o u s lipoproteins could have a n ameliorating effect o n gram-negative sepsis. C h y l o m i c r o n s have a large capacity to inhibit e n d o t o x i n activity. E n d o t o x i n c o n c e n t r a t i o n s in t h e b l o o d o f patients with sepsis are r e p o r t e d to b e 80 to 250 p g / m l , as whereas t h e activity o f u p to 1000 p g o f e x o g e n o u s e n d o t o x i n c a n b e inhibited by 1 m g triglyceride o f chylomicrons in v i t r o ) ~ 2o However, t h e capacity o f t h e hepatocyte to clear the LPS-CM complexes may b e a critical factor in d e t e r m i n i n g the ultim a t e ability o f lipoproteins to r e m o v e LPS f r o m the body. T h e finding that LPS-CM complexes are cleared by a receptor-mediated process that is up-regulated by previous exposure to LPS is f u r t h e r evidence that the hypertriglyceridemia observed d u r i n g sepsis limits LPS toxicity.
REFERENCES 1. Flegel WA, Wolpl D, Mannel DN, Northoff H. hthibidon of endotoxin-induced activation of human monocytes by human lipoproteins. Infect Immun 1989;57:2237-45. 2. Cavaillon JM, Fitting C, Cawaillon NH, Warren H. Cytokine response by monocytes and macrophages to free and lipoprotein bound lipopolysaccharide. Infect Immun 1990;58:2375-82. 3. Harris HW, Grunteld C, Feingold KR, Read TE, Kane JP, Rapp JH. (3hylomicrons alter the fate of endotoxin, decreasing tumor necrosis factor release and preventing death. J Clin Invest 1993; 91:1028-34. 4. Read TE, Harris HW, Grunteld C, Feingold KR, KaneJP, Rapp JH. Cbylomicrons enhance endotoxin excretion in bile. Infect Immun 1993;61:3496-502. 5. Read TE, Grunfeld C, Feingold KR, Calhoun MC, Kumwenda Z, RappJH. Triglyceride-rich lipoproteins prevent septic death in rats. J Exp Med 1995;182:267-72. 6. Windler E, Greeve J, Daerr W, Greten H. Binding of rat chylomicrons and their remnants to the hepatic low-density-lipoprotein receptor and its role in remnant removal, BiochemJ 1988;252:55361. 7. Choi SY, Cooper AD. A comparison of the roles of the low density lipoprotein (LDL) receptor and the LDL receptor-related protein/alpha 2 macroglobulin receptor in chylomicron remnant removal in the mouse in vivo.J Biol Chem 1993;268:1580411. 8. Willnow TE, Goldstein SL, Orth K, Brown MS, HerzJ. Low density lipoprotein receptor-related protein and gp 330 bind similar ligands, including plasminogen activator-inhibitor complexes and lactoferrin an inhibitor of chylomicron remnant clearance. J Biol Cbem 1992;267:26172-80. 9. Yen FT, Mann CJ, Lydie MG, et al. Identification of a lipolysisstimulated receptor that is distinct from the LDL receptor and the LDL receptor-related protein. Biochemistry 1994;33:1172-80.
Ghermay et al.
393
10. Herz J, Qui s, Oesterle A, Desilva HV, Shaft S, Havel RJ. Initial hepatic removal of chylomicron remnants is unaffected but endocytosis is delayed in mice lacking the low density lipoprotein receptor. Proc Nail Acad Sci 1995;92:4611-5. 11. LevinJ, Tomasulo PA, Oser RS. Detection ofendotoxin in human blood and demonstration of an inhibitor. J Lab Clin Med 1970; 75:903-11. 12. Uleveitch RJ. The preparation and characterization of radioiodinated bacterial lipopolysaccharide. Immunochemistry 1978;15: 157-64. 13. Belcher JD, Hamilton RL, Brady SE, et al. Isolation and characterization of three endosomal fractions from the liver of estradiol-treated rats. Proc Nail Acad Sci 1987;84:6785-9. 14. Havel RJ. Chylomicron remnants: hepatic receptors and metabolism. Curr Opin Lipidol 1995;6:312-6. 15. Mokuno H, Brady S, Kotite L, HerzJ, Havel R. Effect of 39 K-da receptor-associated protein on the hepatic uptake and endocytosis of chylomicron remnants and low density lipoproteins in rats. J Biol Chem 1994;269:13238-43. 16.Jackson RL, Busch SJ, Cardin AD. Glycosaminoglycans: molecular properties, protein interactions, and role in physiological processes. Physiol Rev 1991;71:481-539. 17. Shaft S, Brady SE, Bensadoun S, Havel RJ. Role of hepatic lipase in the uptake of chylomicron remnants in rat liver. J Lipid Res 1994;35:709-19. 18. Obayashi T. Addition of perchloric acid to blood samples for colorimetric limulus test using chromogenic substrate: comparison with conventional procedures and clinical applications. J Lab Clin Med 1984;104:321-30. 19. Harris HW, Eichbaum EB, KaneJP, RappJH. Detection ofendotoxin in triglyceride-rich lipoproteins in vitro. J Lab Clin Med 1991;118:186-93. 20. Eichbaum EB, Harris HW, KaneJP, RappJH. Chylomicrons can inhibit endotoxin activity in vitro. J Surg Res 1991;51:413-6.
DISCUSSION Dr. Timothy R. Billiar (Pittsburgh, Pa.). Can you give us a little more information on how you determined the content of the endosome? Are you sure that the endotoxin is really in there, or has the label somehow been transferred to some other lipid moiety associated with the chylomicron? Second, in the past Dr. Rapp's group has shown that chylomicrons protect animals from sepsis and reduce cytokine release. Could this be a mechanism of endotoxin tolerance? Could you c o m m e n t o n the possibility of the increased endotoxin clearance as a mechanism of endotoxin tolerance? Finally, do you have any more information on the importance of the LDL receptor in this process? We know that LDL receptors can act as scavenger receptors for endotoxin in a n u m b e r of cell types. Do you know whether the LDL receptor is really involved? Dr. Ghermay. I shall start by answering your last question first. In terms of the LDL receptor, it is one of the receptors that has been identified as probably the physiologically important receptor among many receptors that are involved in the uptake of chylomicron remnants. We are in the process of trying to determine which receptor is determinant in the uptake of the LPS-CM complex. At this time I can say that it does not appear to be the LDL receptor in our early, preliminary studies. To answer the question of whether the radiolabel sticks to another lipid moiety in terms of going in, we radiolabeled endotoxin and then b o u n d that to the chylomicrons. The ability of radiolabel to j u m p from one moiety to another moiety is a
394
Ghermay et al.
potential. That is something that we should probably investigate; however, that does n o t appear to be the case with several previous experiments we have done in determining this. Dr. David L. Duma (Minneapolis, Minn.). First, could you describe the effect of this agent on mortality? Does this agent in a treatment mode prevent endotoxic mortality in this model? Second, have you attempted to use this agent to prevent mortality in a bacteremic or an infection-type model with or without antibiotics? Dr. Ghermay. We have looked at chylomicron infusions in both cecal ligafion and puncture with significant decrease in mortality in this model with infusions up to 30 minutes after the cecal ligafion and puncture in serial infusions. In addition, we have looked at it with lethal endotoxin injections, and we have shown that there is again significant survival benefit to this in the rat model. H u m a n studies are yet to be done by us, and antibiotics have not b e e n initiated in rats.
Surgery August 1996 Dr. Paul E. Bankey (Dallas, Texas). Do you have any information on whether Kupffer cells or liver macrophages are involved in this clearance of the lipid particles and endotoxins? We have been experimenting with liposomes, and there seem to be more liposomes taken up by the liver macrophages rather than the hepatocytes. In sepsis it seems that most of the hypertriglyceridemia is due to very low density lipoproteins. Have you looked at very low density lipoprotein particles as having a similar effect? Dr. Ghermay. In terms of very low density lipoprotein and LDL, both have been studied a n d b o t h have been shown to have similar survival benefits, both by our group and other groups. The question about the Kupffer cells was one that occurred to us during the design of the experiment. Gadolinium will selectively inhibit the function of Kupffer cells. Although not presented, we did do this a n d there was no significant change to the endosomal data with gadolinium-treated livers.