- Email: [email protected]
Chylomicrons can inhibit endotoxin activity in vitro
JOURNAL
OF SURGICAL
RESEARCH
51,413-416
Chylomicrons
(1991)
Can Inhibit Endotoxin Activity in vitro
ELDAN B. EICHBAUM, B.S., HOBART JOHN P. KANE, M.D., PH.D., AND Departments
W. HARRIS, M.D., M.P.H., JOSEPH H. RAPP, M.D.
of Surgery and Medicine, San Francisco Veterans Affairs Medical Center, San Francisco, California and Cardiouasculur Research Institute, University of California, San Francisco 94143 Submitted
for publication
Trauma, thermal injury, and nonlethal doses of endotoxin can promote the translocation of endotoxin across the mucosal barrier of the colon into the mesenteric lymphatics and systemic circulation. Bacterial endotoxemia induces changes in lipid metabolism, including an increase in circulating triglyceride-rich lipoproteins. Because cholesterol-rich lipoproteins can neutralize the toxic activity of endotoxin, both in vitro and in uiuo, we asked whether triglyceride-rich chylomicrons can inhibit endotoxin activity in vitro as measured by a chromogenic Limulus assay. We tested the effect of intact versus heat-denatured chylomicrons on the in vitro activity of increasing concentrations of Escherichia coli (065:BB) endotoxin. Intact chylomicrons inhibited up to la-fold the detection of as much as 1 rg of endotoxin/mg of chylomicron triglyceride, compared to denatured chylomicrons (P < 0.001). This study shows that chylomicrons are potent inhibitors of endotoxin activity in vitro. Because translocated endotoxin from the colon associates with gut-derived chylomicrons in the mesenteric lymphatics, this may represent a natural defensive mechanism against endotoxemia of enQ 1991 Academic Press. Inc. teric origin.
94121
August 13, 1990
ported via the mesenteric lymphatics, can neutralize endotoxin activity. The purpose of this study was to determine whether chylomicrons can inhibit endotoxin activity in vitro as measured by the Limulus assay. METHODS
Reagents and solutions. Reagent grade ortho-H,PO,, KBr, glacial acetic acid (Fisher Chemical Co., Fair Lawn, NJ) and NaOH (J. T. Baker Chemical Co., Phillipsburg, NJ), apyrogenic, preservative-free 0.9% NaCl (Kendall McGraw Labs, Inc., Irvine, CA), and H,O (Elkins-Sinn, Inc., Cherry Hill, NJ) were used as specified. Escherichia coli, strain 055:B5 endotoxin, was purchased (Difco Laboratories, Detroit, MI), reconstituted with sterile, apyrogenic water to a concentration of 1 rg/ml, and stored in 3-ml aliquots at -7O’C until its use. This preparation of endotoxin had a specific activity of -15 endotoxin units (EU)/ng (USP reference endotoxin). Depyrogenation. To avoidcontamination with exogenously derived endotoxin, all heat-stable materials used in the isolation, processing, and assay of chylomicrons to be used in this study, including test tubes, stoppers, beakers, pipettes, ultracentrifuge tubes, and O-rings, were rendered sterile and free of detectable endotoxin (~5-10 pg/ml) by a combination of steam autoclaving followed by dry heating at 180°C for a minimum of 4 hr [12, 131. All chylomicrons were isolated using depyrogenated stainless steel ultracentrifuge tubes (Beckman Instrument Co., Palo Alto, CA) with custom-crafted silicone O-rings. Plasma. Human postprandial venous blood was collected- from alcohol-cleansed antecubital fossae of five healthy volunteers into sterile plastic syringes containing endotoxin-free heparin (Sigma Chemical Co., St. Louis, MO), producing a final concentration of 10 IU/ml and immediately placed on ice. Plasma was expeditiously separated by centrifugation at 2000g for 20 min at 25°C and then stored in depyrogenated glass tubes at 4°C until further processing. Plasma samples were processed within 24 hr of their collection.
INTRODUCTION Trauma, thermal injury, and nonlethal doses of endotoxin can promote the translocation of viable bacteria and bacterial byproducts across the mucosal barrier of the colon [ 1, 21. Migrating via mesenteric lymphatics, these microorganisms appear initially to invade the mesenteric lymph nodes, then the spleen and liver, and ultimately the bloodstream, depending on the severity of the inciting injury [l-4]. The resultant bacterial endotoxemia can induce changes in lipid metabolism. The earliest and most significant of these changes is an increase in circulating triglyceride-rich lipoproteins [5-71. Because lipoproteins rich in cholesterol esters, specifically LDL and HDL, have been shown to neutralize the toxic activity of endotoxin both in vitro and in uiuo [B-11], we asked whether the triglyceride-rich chylomicrons, formed in the gut and like translocated endotoxin, are also trans413
0022~4804/91$1.50 Copyright 0 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.
414
JOURNAL
OF SURGICAL
RESEARCH:
Chylomicrons were isolated from normal human postprandial plasma by ultracentrifugation at plasma density for 30 min, 12”C, at 20,000 rpm using a 40.3 rotor [ 141. The lipoproteins were diluted using endotoxin-free 50 n&f phosphate-buffered saline (PBS, pH = 7.4) as needed to reach the desired triglyceride concentration of 1.0 mg/ml. For all experiments described the chylomicrons were used within 7 days of their isolation. The triglyceride and cholesterol content of the chylomicron preparations was determined using standard enzymatic assays (Sigma, and Wako Chemicals USA, Inc., Dallas, TX). Chromogenic limulus assay. To assess the ability of chylomicrons to inhibit endotoxin activity in vitro we used a chromogenic modification of the Limulus assay [ 151. Chylomicrons suspended in endotoxin-free PBS were incubated at 37°C for 3 hr in a shaking water bath with increasing concentrations of endotoxin ranging from 0 to l,OOO,OOOpg endotoxin/mg of chylomicron triglyceride. Then aliquots were assayed for detectable endotoxin activity. One set of aliquots (control samples) were diluted 1:9 (vol:vol) with endotoxin-free PBS, followed by heating at 75°C for 5 min to denature the chylomicrons. A second set of aliquots (test samples) were diluted but not heated thereby retaining intact chylomicrons. The diluted samples (0.1 ml) were added to 0.1 ml Limulus lysate (Sigma), reconstituted to 20-fold less than the manufacturer’s recommended concentration with endotoxin-free H,O, and incubated at 37°C for 60 min in a shaking water bath. Thereafter 0.4 ml of 0.5 n&f chromogenic substrate S-2222 (Helena Laboratories, Beaumont, TX), reconstituted using endotoxinfree H,O, was added, and the mixture was incubated at 37°C for an additional 30 min. The reaction was then quenched by the addition of 0.4 ml 60% acetic acid. Absorption of the final mixture was measured at 405 nm using a Cary Model 219 spectrophotometer zeroed against distilled water. The mean value from each set of diluted chylomicron samples was adjusted by subtracting the optical density of a blank sample. The blank consisted of diluted chylomicrons with no endotoxin, incubated as previously described with lysate and chromogenic substrate, and quenched by acetic acid. Statistical analysis. Each sample was assayed in triplicate and comparisons between groups were assessed by paired t test. Statistical significance was assigned a P =s 0.05.
RESULTS
The chromogenic Limulus assay was linear over ,a range of 5-100 pg endotoxin/ml saline, producing a maximal optical density reading of approximately 2.0 (Fig. 1). The chylomicron isolation procedure yielded samples with a mean triglyceride concentration of 8.5 f 2.8 mg/ml (*SD).
VOL.
51, NO. 5, NOVEMBER
1991
2 Y I 0.25 + 0.01X,
r
q
0.99
1
“0
2’0
4’0
6’0
[Endotoxin]
FIG. 1. Standard curve for the mogenic Limulus assay. The data determinations. A simple equation the data points: Y = 0.25 + 0.01X,
6’0
li0
(pg/ml)
detection of endotoxin by a chroare the mean + SD of the three is generated which best describes r = 0.99. OD, optical density.
At all concentrations tested the addition of endotoxin to unheated samples resulted in a 4- to 12-fold reduction in detectable endotoxin activity as compared to heated samples (Fig. 2). The heated samples showed endotoxin activity over the entire range of added endotoxin (500 pg to 1 pg endotoxin/mg chylomicron TG). In contrast, unheated samples consistently yielded lower levels of endotoxin activity over the same range of endotoxin concentrations (P < 0.001). DISCUSSION
Early studies examining the effect of serum on endotoxin were directed at understanding the mechanism by which serum reduced the toxicity of endotoxin [16, 171. From these studies Skarnes suggested that endotoxin might interact with p- and a,-lipoproteins with a resultant modification of endotoxin’s physical and immunochemical properties [ 181. Subsequently, Ulevitch and Johnston demonstrated that the incubation of endotoxin with serum in vitro reduced its buoyant density and subsequent toxicity [8], a process termed detoxifiation. It has been proposed that following incubation with serum, endotoxin is first disaggregated and then binds to HDL to form a stable lipoprotein-endotoxin complex [9]. Whereas these studies of the interaction between lipoproteins and endotoxin have primarily focused on HDL, recent studies have demonstrated that chylomicrons, as well as VLDL, B-VLDL, and LDL, can also bind endotoxin [ 11, 191. Data from our laboratory suggest that chylomicrons and VLDL can protect mice against endotoxin-induced death [19]. While the mecha-
EICHBAUM
ET AL.:
CHYLOMICRONS
1.5
q HEATED 0
UNHEATED
1.0 E g E 8 0.5 l
0.0
T
-
2.7
0 Log
[Endotoxin]
3 (pg/mg
4
5
Chylomicron
6 TG)
FIG. 2. Effect of chylomicrons on the detectable Limulus assay activity of increasing endotoxin concentrations. Chylomicron samples (N = 5, 1 mg TG/ml) suspended in endotoxin-free PBS were incubated for 3 hr at 37°C in a shaking water bath with increasing concentrations of E. coli (055:B5) endotoxin (500 pg - 1 rg/mg chylomicron TG). Then aliquots were assayed for detectable endotoxin activity (see Methods). The data are the mean & SD of the three determinations and are presented as a semilog plot. The SD is omitted when it is smaller than the data point. TG, triglyceride; PBS, phosphate-buffered saline; OD, optical density; heated, denatured chylomicrons (control); unheated, intact chylomicrons; * P < 0.001.
nism responsible for the protection is unclear, the interaction between endotoxin and triglyceride-rich lipoproteins was dependent on the dose of lipoprotein lipid administered. This study suggests that ultracentrifugally isolated chylomicrons are potent inhibitors of endotoxin activity in vitro as measured by a chromogenic Limulus assay. Intact chylomicrons significantly inhibited the detection of 1 pg of E. coli endotoxin/mg of lipoprotein triglyceride. In contrast, control samples revealed the presence of pg quantities of endotoxin. Both control and test samples were diluted with PBS but the control samples were then heated to denature the chylomicrons. Apparently, once the lipoprotein particles were denatured their ability to inhibit endotoxin activity was lost, suggesting that the inhibitory property of chylomicrons may be dependent on an intact lipoprotein structure. The thermal treatment would be expected to denature at least some of the apolipoproteins on the chylomicron complex. Large quantities of endotoxin can be present in chylomicron preparations yet go undetected by the standard Limulus assay. This suggests that chylomicrons have a significant capacity to inhibit endotoxin activity. We have shown that unheated chylomicron samples to which 1 pg of endotoxin/mg of chylomicron TG had been added yielded a mean OD reading of 0.2719 f 0.013 (MD). Since this level of endotoxin activity corre-
INHIBIT
ENDOTOXIN
IN
VITRO
415
sponds to an endotoxin concentration of ~10 pg/ml (Fig. 1) the average chylomicron preparation (8.5 mg TG/ml) could contain over 8 /lg of endotoxin/ml and still yield a negative Limulus assay. Of course, ultracentrifugally isolated chylomicrons contain small quantities of whole plasma that coisolate with these large triglyceride-rich particles. Therefore, plasma proteins present in our chylomicron samples could also be contributing to the inhibition of endotoxin activity. While it is known that lipoprotein-free plasma is required for lipoproteins to bind endotoxin effectively [E&11,19], we have recently shown that lo-fold dilutions of plasma, as performed in this study, significantly reduce the capacity of lipoprotein-free plasma to inhibit endotoxin activity [20]. Lipoprotein-free plasma alone does not demonstrate nearly the degree of endotoxin inhibition shown in this study. The other plasma lipoproteins, especially the cholesterol-rich LDL and HDL, are also present in the chylomicron samples. However, the total cholesterol content of our chylomicron samples was low enough (to.2 mg/ml) to render improbable any significant contribution by these lipoproteins to the inhibition of endotoxin activity observed in this study. Demonstrating the ability of chylomicrons to inhibit the detection of endotoxin by the Limulus assay in vitro does not directly address the biological activity of chylomicron-bound endotoxin in uivo. There is, however, strong evidence to suggest that endotoxin’s activity in vitro, as measured by the Limulus assay, is a reflection of its activity in vivo. The Limulus assay is a bioassay. It measures the bioavailability of the lipid A region of the endotoxin macromolecule by catalyzing the activation of an enzymatic cascade [21-231. Lipid A is certainly the most biologically significant region of the molecule as it is responsible for most if not all of endotoxin’s pathogenic activities [24]. In addition, we have recently shown that triglyceride-rich lipoproteins can protect against endotoxin lethality in uiuo [19]. In summary, it appears as though chylomicrons can be potent inhibitors of endotoxin activity in vitro. Since translocated endotoxin from the colon travels along with gut-derived chylomicrons via mesenteric lymphatics the in vitro interaction between the two may provide insight into an endogenous host defensive mechanism against endotoxemia. ACKNOWLEDGMENTS The technical advice of Robert I. Roth and Jack Levin, and the able technical assistance of Judy Tweedie are greatly appreciated. This work was supported in part by NIH Grants HL41470, HL07737, and grants from the ARCS Foundation, the Veterans Administration, and the Pacific Vascular Research Foundation.
REFERENCES 1.
Ravin, H. A., Rowley, D., Jenkins, C., and Fine, J. On the absorption of bacterial endotoxin from the gastro-intestinal tract of the normal and shocked animal. J. Exp. Med. 112: 783, 1960.
416
JOURNAL
OF SURGICAL
RESEARCH:
VOL.
2.
Deitch, E. A., Winterton, J., and Berg, R. Effect of starvation, malnutrition, and trauma on the gastrointestinal tract flora and bacterial translocation. Arch. Surg. 122: 1019, 1987.
13.
3.
Deitch, E. A., Winterton, J., Li, M., and Berg, R. The gut as a portal of entry for bacteremia. Ann. Surg. 205: 681, 1987.
14.
4.
Deitch, E. A., Berg, R., and Specian, R. Endotoxin promotes the translocation of bacteria from the gut. Arch. Surg. 122: 185, 1987.
15.
Sammalkorpi, K., Vahonen, V., Kerttula, Y., Nikkila, E., and Taskinen, M. Changes in serum lipoprotein pattern induced by acute infections. Metabolism 37: 859, 1988.
16.
Sakaguchi, S. Metabolic disorders of serum lipoproteins in endotoxin-poisoned mice: The role of high density lipoprotein (HDL) and triglyceride-rich lipoproteins. Microbial. Immunol. 26: 1017, 1982.
17.
5.
6.
7.
Lanza-Jacoby, S., Lansey, S. C., Cleary, M. P., and Rosato, F. E. Alterations in lipogenic enzymes and lipoprotein lipase activity during gram-negative sepsis in the rat. Arch. Surg. 11'7: 144, 1982.
8.
Ulevitch, R. J., and Johnston, A. R. The modification of the biophysical and endotoxin properties of bacterial lipopolysaccharide by serum. J. Clin. Invest. 62: 1313, 1978.
9.
Ulevitch, R. J., Johnston, A. R., and Weinstein, D. B. New function for high density lipoproteins. Their participation in intravascular reactions of bacterial lipopolysaccharides. J. Clin. Invest. 64: 1516,197s.
10.
Ulevitch, R. J., Johnston, A. R., and Weinstein, D. B. New function for high density lipoproteins. Isolation and characterization of a bacterial lipopolysaccharide-high density lipoprotein complex formed in rabbit plasma. J. Clin. Inuest. 67: 827, 1981.
11.
Van Lenten, B. J., Fogelman, A. M., Haberland, M. E., and Edwards, P. A. The role of lipoproteins and receptor-mediated endocytosis in the transport of bacterial lipopolysaccharide. Proc. Natl. Acad. Sci. USA 83: 2704, 1986.
12.
Levin, J., Tomasulo, P. A., and Oser, R. S. Detection of endotoxin in human blood and demonstration of an inhibitor. J. Lab. Clin. Med. 75: 903, 1970.
18.
19.
20.
21.
51, NO. 5, NOVEMBER
1991
Jacob, A. I., Goldberg, P. K., Bloom, N., Degenshein, G. A., and Kozinn, P. J. Endotoxin and bacteria in portal blood. Gastroentero&y 72: 1268,1977. Havel, R. J., Eder, H. A., and Bragdon, J. H. The distribution and chemical composition of ultracentrifugally separated lipoproteins in human serum. J. Clin. Invest. 34: 1345, 1955. Harris, R. I., Stone, P. C. W., and Stuart, J. An improved chromogenic substrate endotoxin assay for clinical use. J. Clin. Path&. 36: 1145, 1983. Rosen, F. S., Skarnes, R. C., Landy, M., and Shear, M. J. Inactivation of endotoxin by a humoral component. III. Role of divalent cation and a dialyzable component. J. hp. Med. 108: 701, 1958. Skarnes, R. C., Rosen, F. S., Shear, M. J., and Landy, M. Inactivation of endotoxin by a humoral component. II. Interaction of endotoxin with serum and plasma. J. Exp. Med. 108: 685,1958. Skarnes, R. C. The inactivation of endotoxin after interaction with certain proteins of normal serum. Ann. NY Acad. Sci. 133: 644, 1966. Harris, H. W., Grunfeld, C., Feingold, K. R., and Rapp, J. H. Human very low density lipoproteins and chylomicrons can protect against endotoxin-induced death in mice. J. Clin. Inoest. 86: 6S6,1996. Harris, H. W., Eichbaum, E. B., Kane, J. P., and Rapp, J. H. Detection of endotoxin in triglyceride-rich lipoproteins in vitro. J. Lab. Clin. Med., in press. Levin, J., and Bang, F. B. The role of endotoxin in the extracellular coagulation of Limulus blood. BuU. Johns Hopkins Hosp.
115:265,1964. 22. 23.
24.
Levin, J., and Bang, F. B. A description of cellular coagulation in the Limulus. BuU. Johns Hopkins Hosp. 115: 337, 1964. Young, N. S., Levin, J., and Prendergast, R. A. An inverterbrate coagulation system activated by endotoxin: Evidence for enzymatic mediation. J. Clin. Inuest. 51: 1790, 1972. Luderitz, O., Galanos, C., Lehmann, V., Nurminen, M., Rietschel, E. T., Rosenfelder, G., Simon, M., and Westphal, 0. Lipid A: Chemical structure and biological activity. J. Infect. Dis. 128(Suppl): s17, 1973.
OF SURGICAL
RESEARCH
51,413-416
Chylomicrons
(1991)
Can Inhibit Endotoxin Activity in vitro
ELDAN B. EICHBAUM, B.S., HOBART JOHN P. KANE, M.D., PH.D., AND Departments
W. HARRIS, M.D., M.P.H., JOSEPH H. RAPP, M.D.
of Surgery and Medicine, San Francisco Veterans Affairs Medical Center, San Francisco, California and Cardiouasculur Research Institute, University of California, San Francisco 94143 Submitted
for publication
Trauma, thermal injury, and nonlethal doses of endotoxin can promote the translocation of endotoxin across the mucosal barrier of the colon into the mesenteric lymphatics and systemic circulation. Bacterial endotoxemia induces changes in lipid metabolism, including an increase in circulating triglyceride-rich lipoproteins. Because cholesterol-rich lipoproteins can neutralize the toxic activity of endotoxin, both in vitro and in uiuo, we asked whether triglyceride-rich chylomicrons can inhibit endotoxin activity in vitro as measured by a chromogenic Limulus assay. We tested the effect of intact versus heat-denatured chylomicrons on the in vitro activity of increasing concentrations of Escherichia coli (065:BB) endotoxin. Intact chylomicrons inhibited up to la-fold the detection of as much as 1 rg of endotoxin/mg of chylomicron triglyceride, compared to denatured chylomicrons (P < 0.001). This study shows that chylomicrons are potent inhibitors of endotoxin activity in vitro. Because translocated endotoxin from the colon associates with gut-derived chylomicrons in the mesenteric lymphatics, this may represent a natural defensive mechanism against endotoxemia of enQ 1991 Academic Press. Inc. teric origin.
94121
August 13, 1990
ported via the mesenteric lymphatics, can neutralize endotoxin activity. The purpose of this study was to determine whether chylomicrons can inhibit endotoxin activity in vitro as measured by the Limulus assay. METHODS
Reagents and solutions. Reagent grade ortho-H,PO,, KBr, glacial acetic acid (Fisher Chemical Co., Fair Lawn, NJ) and NaOH (J. T. Baker Chemical Co., Phillipsburg, NJ), apyrogenic, preservative-free 0.9% NaCl (Kendall McGraw Labs, Inc., Irvine, CA), and H,O (Elkins-Sinn, Inc., Cherry Hill, NJ) were used as specified. Escherichia coli, strain 055:B5 endotoxin, was purchased (Difco Laboratories, Detroit, MI), reconstituted with sterile, apyrogenic water to a concentration of 1 rg/ml, and stored in 3-ml aliquots at -7O’C until its use. This preparation of endotoxin had a specific activity of -15 endotoxin units (EU)/ng (USP reference endotoxin). Depyrogenation. To avoidcontamination with exogenously derived endotoxin, all heat-stable materials used in the isolation, processing, and assay of chylomicrons to be used in this study, including test tubes, stoppers, beakers, pipettes, ultracentrifuge tubes, and O-rings, were rendered sterile and free of detectable endotoxin (~5-10 pg/ml) by a combination of steam autoclaving followed by dry heating at 180°C for a minimum of 4 hr [12, 131. All chylomicrons were isolated using depyrogenated stainless steel ultracentrifuge tubes (Beckman Instrument Co., Palo Alto, CA) with custom-crafted silicone O-rings. Plasma. Human postprandial venous blood was collected- from alcohol-cleansed antecubital fossae of five healthy volunteers into sterile plastic syringes containing endotoxin-free heparin (Sigma Chemical Co., St. Louis, MO), producing a final concentration of 10 IU/ml and immediately placed on ice. Plasma was expeditiously separated by centrifugation at 2000g for 20 min at 25°C and then stored in depyrogenated glass tubes at 4°C until further processing. Plasma samples were processed within 24 hr of their collection.
INTRODUCTION Trauma, thermal injury, and nonlethal doses of endotoxin can promote the translocation of viable bacteria and bacterial byproducts across the mucosal barrier of the colon [ 1, 21. Migrating via mesenteric lymphatics, these microorganisms appear initially to invade the mesenteric lymph nodes, then the spleen and liver, and ultimately the bloodstream, depending on the severity of the inciting injury [l-4]. The resultant bacterial endotoxemia can induce changes in lipid metabolism. The earliest and most significant of these changes is an increase in circulating triglyceride-rich lipoproteins [5-71. Because lipoproteins rich in cholesterol esters, specifically LDL and HDL, have been shown to neutralize the toxic activity of endotoxin both in vitro and in uiuo [B-11], we asked whether the triglyceride-rich chylomicrons, formed in the gut and like translocated endotoxin, are also trans413
0022~4804/91$1.50 Copyright 0 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.
414
JOURNAL
OF SURGICAL
RESEARCH:
Chylomicrons were isolated from normal human postprandial plasma by ultracentrifugation at plasma density for 30 min, 12”C, at 20,000 rpm using a 40.3 rotor [ 141. The lipoproteins were diluted using endotoxin-free 50 n&f phosphate-buffered saline (PBS, pH = 7.4) as needed to reach the desired triglyceride concentration of 1.0 mg/ml. For all experiments described the chylomicrons were used within 7 days of their isolation. The triglyceride and cholesterol content of the chylomicron preparations was determined using standard enzymatic assays (Sigma, and Wako Chemicals USA, Inc., Dallas, TX). Chromogenic limulus assay. To assess the ability of chylomicrons to inhibit endotoxin activity in vitro we used a chromogenic modification of the Limulus assay [ 151. Chylomicrons suspended in endotoxin-free PBS were incubated at 37°C for 3 hr in a shaking water bath with increasing concentrations of endotoxin ranging from 0 to l,OOO,OOOpg endotoxin/mg of chylomicron triglyceride. Then aliquots were assayed for detectable endotoxin activity. One set of aliquots (control samples) were diluted 1:9 (vol:vol) with endotoxin-free PBS, followed by heating at 75°C for 5 min to denature the chylomicrons. A second set of aliquots (test samples) were diluted but not heated thereby retaining intact chylomicrons. The diluted samples (0.1 ml) were added to 0.1 ml Limulus lysate (Sigma), reconstituted to 20-fold less than the manufacturer’s recommended concentration with endotoxin-free H,O, and incubated at 37°C for 60 min in a shaking water bath. Thereafter 0.4 ml of 0.5 n&f chromogenic substrate S-2222 (Helena Laboratories, Beaumont, TX), reconstituted using endotoxinfree H,O, was added, and the mixture was incubated at 37°C for an additional 30 min. The reaction was then quenched by the addition of 0.4 ml 60% acetic acid. Absorption of the final mixture was measured at 405 nm using a Cary Model 219 spectrophotometer zeroed against distilled water. The mean value from each set of diluted chylomicron samples was adjusted by subtracting the optical density of a blank sample. The blank consisted of diluted chylomicrons with no endotoxin, incubated as previously described with lysate and chromogenic substrate, and quenched by acetic acid. Statistical analysis. Each sample was assayed in triplicate and comparisons between groups were assessed by paired t test. Statistical significance was assigned a P =s 0.05.
RESULTS
The chromogenic Limulus assay was linear over ,a range of 5-100 pg endotoxin/ml saline, producing a maximal optical density reading of approximately 2.0 (Fig. 1). The chylomicron isolation procedure yielded samples with a mean triglyceride concentration of 8.5 f 2.8 mg/ml (*SD).
VOL.
51, NO. 5, NOVEMBER
1991
2 Y I 0.25 + 0.01X,
r
q
0.99
1
“0
2’0
4’0
6’0
[Endotoxin]
FIG. 1. Standard curve for the mogenic Limulus assay. The data determinations. A simple equation the data points: Y = 0.25 + 0.01X,
6’0
li0
(pg/ml)
detection of endotoxin by a chroare the mean + SD of the three is generated which best describes r = 0.99. OD, optical density.
At all concentrations tested the addition of endotoxin to unheated samples resulted in a 4- to 12-fold reduction in detectable endotoxin activity as compared to heated samples (Fig. 2). The heated samples showed endotoxin activity over the entire range of added endotoxin (500 pg to 1 pg endotoxin/mg chylomicron TG). In contrast, unheated samples consistently yielded lower levels of endotoxin activity over the same range of endotoxin concentrations (P < 0.001). DISCUSSION
Early studies examining the effect of serum on endotoxin were directed at understanding the mechanism by which serum reduced the toxicity of endotoxin [16, 171. From these studies Skarnes suggested that endotoxin might interact with p- and a,-lipoproteins with a resultant modification of endotoxin’s physical and immunochemical properties [ 181. Subsequently, Ulevitch and Johnston demonstrated that the incubation of endotoxin with serum in vitro reduced its buoyant density and subsequent toxicity [8], a process termed detoxifiation. It has been proposed that following incubation with serum, endotoxin is first disaggregated and then binds to HDL to form a stable lipoprotein-endotoxin complex [9]. Whereas these studies of the interaction between lipoproteins and endotoxin have primarily focused on HDL, recent studies have demonstrated that chylomicrons, as well as VLDL, B-VLDL, and LDL, can also bind endotoxin [ 11, 191. Data from our laboratory suggest that chylomicrons and VLDL can protect mice against endotoxin-induced death [19]. While the mecha-
EICHBAUM
ET AL.:
CHYLOMICRONS
1.5
q HEATED 0
UNHEATED
1.0 E g E 8 0.5 l
0.0
T
-
2.7
0 Log
[Endotoxin]
3 (pg/mg
4
5
Chylomicron
6 TG)
FIG. 2. Effect of chylomicrons on the detectable Limulus assay activity of increasing endotoxin concentrations. Chylomicron samples (N = 5, 1 mg TG/ml) suspended in endotoxin-free PBS were incubated for 3 hr at 37°C in a shaking water bath with increasing concentrations of E. coli (055:B5) endotoxin (500 pg - 1 rg/mg chylomicron TG). Then aliquots were assayed for detectable endotoxin activity (see Methods). The data are the mean & SD of the three determinations and are presented as a semilog plot. The SD is omitted when it is smaller than the data point. TG, triglyceride; PBS, phosphate-buffered saline; OD, optical density; heated, denatured chylomicrons (control); unheated, intact chylomicrons; * P < 0.001.
nism responsible for the protection is unclear, the interaction between endotoxin and triglyceride-rich lipoproteins was dependent on the dose of lipoprotein lipid administered. This study suggests that ultracentrifugally isolated chylomicrons are potent inhibitors of endotoxin activity in vitro as measured by a chromogenic Limulus assay. Intact chylomicrons significantly inhibited the detection of 1 pg of E. coli endotoxin/mg of lipoprotein triglyceride. In contrast, control samples revealed the presence of pg quantities of endotoxin. Both control and test samples were diluted with PBS but the control samples were then heated to denature the chylomicrons. Apparently, once the lipoprotein particles were denatured their ability to inhibit endotoxin activity was lost, suggesting that the inhibitory property of chylomicrons may be dependent on an intact lipoprotein structure. The thermal treatment would be expected to denature at least some of the apolipoproteins on the chylomicron complex. Large quantities of endotoxin can be present in chylomicron preparations yet go undetected by the standard Limulus assay. This suggests that chylomicrons have a significant capacity to inhibit endotoxin activity. We have shown that unheated chylomicron samples to which 1 pg of endotoxin/mg of chylomicron TG had been added yielded a mean OD reading of 0.2719 f 0.013 (MD). Since this level of endotoxin activity corre-
INHIBIT
ENDOTOXIN
IN
VITRO
415
sponds to an endotoxin concentration of ~10 pg/ml (Fig. 1) the average chylomicron preparation (8.5 mg TG/ml) could contain over 8 /lg of endotoxin/ml and still yield a negative Limulus assay. Of course, ultracentrifugally isolated chylomicrons contain small quantities of whole plasma that coisolate with these large triglyceride-rich particles. Therefore, plasma proteins present in our chylomicron samples could also be contributing to the inhibition of endotoxin activity. While it is known that lipoprotein-free plasma is required for lipoproteins to bind endotoxin effectively [E&11,19], we have recently shown that lo-fold dilutions of plasma, as performed in this study, significantly reduce the capacity of lipoprotein-free plasma to inhibit endotoxin activity [20]. Lipoprotein-free plasma alone does not demonstrate nearly the degree of endotoxin inhibition shown in this study. The other plasma lipoproteins, especially the cholesterol-rich LDL and HDL, are also present in the chylomicron samples. However, the total cholesterol content of our chylomicron samples was low enough (to.2 mg/ml) to render improbable any significant contribution by these lipoproteins to the inhibition of endotoxin activity observed in this study. Demonstrating the ability of chylomicrons to inhibit the detection of endotoxin by the Limulus assay in vitro does not directly address the biological activity of chylomicron-bound endotoxin in uivo. There is, however, strong evidence to suggest that endotoxin’s activity in vitro, as measured by the Limulus assay, is a reflection of its activity in vivo. The Limulus assay is a bioassay. It measures the bioavailability of the lipid A region of the endotoxin macromolecule by catalyzing the activation of an enzymatic cascade [21-231. Lipid A is certainly the most biologically significant region of the molecule as it is responsible for most if not all of endotoxin’s pathogenic activities [24]. In addition, we have recently shown that triglyceride-rich lipoproteins can protect against endotoxin lethality in uiuo [19]. In summary, it appears as though chylomicrons can be potent inhibitors of endotoxin activity in vitro. Since translocated endotoxin from the colon travels along with gut-derived chylomicrons via mesenteric lymphatics the in vitro interaction between the two may provide insight into an endogenous host defensive mechanism against endotoxemia. ACKNOWLEDGMENTS The technical advice of Robert I. Roth and Jack Levin, and the able technical assistance of Judy Tweedie are greatly appreciated. This work was supported in part by NIH Grants HL41470, HL07737, and grants from the ARCS Foundation, the Veterans Administration, and the Pacific Vascular Research Foundation.
REFERENCES 1.
Ravin, H. A., Rowley, D., Jenkins, C., and Fine, J. On the absorption of bacterial endotoxin from the gastro-intestinal tract of the normal and shocked animal. J. Exp. Med. 112: 783, 1960.
416
JOURNAL
OF SURGICAL
RESEARCH:
VOL.
2.
Deitch, E. A., Winterton, J., and Berg, R. Effect of starvation, malnutrition, and trauma on the gastrointestinal tract flora and bacterial translocation. Arch. Surg. 122: 1019, 1987.
13.
3.
Deitch, E. A., Winterton, J., Li, M., and Berg, R. The gut as a portal of entry for bacteremia. Ann. Surg. 205: 681, 1987.
14.
4.
Deitch, E. A., Berg, R., and Specian, R. Endotoxin promotes the translocation of bacteria from the gut. Arch. Surg. 122: 185, 1987.
15.
Sammalkorpi, K., Vahonen, V., Kerttula, Y., Nikkila, E., and Taskinen, M. Changes in serum lipoprotein pattern induced by acute infections. Metabolism 37: 859, 1988.
16.
Sakaguchi, S. Metabolic disorders of serum lipoproteins in endotoxin-poisoned mice: The role of high density lipoprotein (HDL) and triglyceride-rich lipoproteins. Microbial. Immunol. 26: 1017, 1982.
17.
5.
6.
7.
Lanza-Jacoby, S., Lansey, S. C., Cleary, M. P., and Rosato, F. E. Alterations in lipogenic enzymes and lipoprotein lipase activity during gram-negative sepsis in the rat. Arch. Surg. 11'7: 144, 1982.
8.
Ulevitch, R. J., and Johnston, A. R. The modification of the biophysical and endotoxin properties of bacterial lipopolysaccharide by serum. J. Clin. Invest. 62: 1313, 1978.
9.
Ulevitch, R. J., Johnston, A. R., and Weinstein, D. B. New function for high density lipoproteins. Their participation in intravascular reactions of bacterial lipopolysaccharides. J. Clin. Invest. 64: 1516,197s.
10.
Ulevitch, R. J., Johnston, A. R., and Weinstein, D. B. New function for high density lipoproteins. Isolation and characterization of a bacterial lipopolysaccharide-high density lipoprotein complex formed in rabbit plasma. J. Clin. Inuest. 67: 827, 1981.
11.
Van Lenten, B. J., Fogelman, A. M., Haberland, M. E., and Edwards, P. A. The role of lipoproteins and receptor-mediated endocytosis in the transport of bacterial lipopolysaccharide. Proc. Natl. Acad. Sci. USA 83: 2704, 1986.
12.
Levin, J., Tomasulo, P. A., and Oser, R. S. Detection of endotoxin in human blood and demonstration of an inhibitor. J. Lab. Clin. Med. 75: 903, 1970.
18.
19.
20.
21.
51, NO. 5, NOVEMBER
1991
Jacob, A. I., Goldberg, P. K., Bloom, N., Degenshein, G. A., and Kozinn, P. J. Endotoxin and bacteria in portal blood. Gastroentero&y 72: 1268,1977. Havel, R. J., Eder, H. A., and Bragdon, J. H. The distribution and chemical composition of ultracentrifugally separated lipoproteins in human serum. J. Clin. Invest. 34: 1345, 1955. Harris, R. I., Stone, P. C. W., and Stuart, J. An improved chromogenic substrate endotoxin assay for clinical use. J. Clin. Path&. 36: 1145, 1983. Rosen, F. S., Skarnes, R. C., Landy, M., and Shear, M. J. Inactivation of endotoxin by a humoral component. III. Role of divalent cation and a dialyzable component. J. hp. Med. 108: 701, 1958. Skarnes, R. C., Rosen, F. S., Shear, M. J., and Landy, M. Inactivation of endotoxin by a humoral component. II. Interaction of endotoxin with serum and plasma. J. Exp. Med. 108: 685,1958. Skarnes, R. C. The inactivation of endotoxin after interaction with certain proteins of normal serum. Ann. NY Acad. Sci. 133: 644, 1966. Harris, H. W., Grunfeld, C., Feingold, K. R., and Rapp, J. H. Human very low density lipoproteins and chylomicrons can protect against endotoxin-induced death in mice. J. Clin. Inoest. 86: 6S6,1996. Harris, H. W., Eichbaum, E. B., Kane, J. P., and Rapp, J. H. Detection of endotoxin in triglyceride-rich lipoproteins in vitro. J. Lab. Clin. Med., in press. Levin, J., and Bang, F. B. The role of endotoxin in the extracellular coagulation of Limulus blood. BuU. Johns Hopkins Hosp.
115:265,1964. 22. 23.
24.
Levin, J., and Bang, F. B. A description of cellular coagulation in the Limulus. BuU. Johns Hopkins Hosp. 115: 337, 1964. Young, N. S., Levin, J., and Prendergast, R. A. An inverterbrate coagulation system activated by endotoxin: Evidence for enzymatic mediation. J. Clin. Inuest. 51: 1790, 1972. Luderitz, O., Galanos, C., Lehmann, V., Nurminen, M., Rietschel, E. T., Rosenfelder, G., Simon, M., and Westphal, 0. Lipid A: Chemical structure and biological activity. J. Infect. Dis. 128(Suppl): s17, 1973.