Microbial Pathogenesis 1992 ; 13 : 37-47
Hydrazine sulfate protection against endotoxin lethality : analysis of effects on expression of hepatic cytokine genes and an acute-phase gene Swapan K . De, Richard Silverstein and Glen K . Andrews* Department of Biochemistry and Molecular Biology, University of Kansas Medical Center, 39th and Rainbow Blvd., Kansas City, KS 66103, U .S .A . (Received February 6, 1992 ; accepted in revised form April 6, 1992)
De, S . K . (Dept of Biochemistry and Molecular Biology, University of Kansas Medical Center, 39th and Rainbow Blvd ., Kansas City, KS 66103, U .S .A .), R . Silverstein and G . K . Andrews . Hydrazine sulfate protection against endotoxin lethality : analysis of effects on expression of hepatic cytokine genes and an acute-phase gene . Microbial Pathogenesis 1992 ; 13 : 37-47 . Hydrazine sulfate (HS) pretreatment protects mice against the lethal effects of bacterial endotoxin lipopolysaccharide (LPS) through mechanisms yet to be established . The liver was examined as a model organ to determine HS effects on (a) LPS activation of leukocyte (Kupffer cell) interleukin-1[J (IL-1 fl) and tumor necrosis factor-7 (TNF-a) genes and (b) subsequent cytokine-mediated induction of the acute-phase response as measured by hepatic metallothionein (MT) gene expression . The utility of this model was documented by in situ hybridization which showed that acute induction by LPS of the IL-1/f gene occurred in cells found in liver sinusoids, consistent with Kupffer cells, whereas induction of the MT gene occurred in hepatocytes . The cell specific expression of these genes was further verified by Northern blot hybridization to LPS-treated liver RNA which showed that the LPS-mediated increase in hepatic cytokine mRNA levels, unlike that of MT, was not prevented by Dgalactosamine (D-GaIN) treatment . Northern blot hybridization established that HS pretreatment did not block the acute induction of hepatic cytokine mRNAs (IL-1R and TNF-a) by LPS nor did it induce these cytokine mRNAs in the absence of LPS . Northern blot hybridization further established that HS did not prevent LPS-mediated activation of hepatocyte MT gene expression . Thus, HS does not prevent LPS from activating liver leukocytes . These results also suggest that HS pretreatment neither prevents the general release of cytokines from LPS activated leukocytes nor the general induction of acute-phase protein gene expression in hepatocytes . Therefore, it is unlikely that HS protection against LPS-mediated lethality entails a global down-regulation of LPS-mediated inflammatory events . Key words : hydrazine; lipopolysaccharide ; endotoxemia ; metallothionein ; tumor necrosis factor ; interleukin ; mouse liver .
Introduction Gram-negative bacterial LPS initiates a series of pathophysiological changes including impaired gas exchange of the lung and formation of lung edema,' activation of the plasmatic coagulation cascade and fibrinolysis, Z complement activation, :' formation of
*Author to whom correspondence should be addressed at : Department of Biochemistry and Molecular Biology, WHE 4018, University of Kansas Medical Center, 39th and Rainbow Blvd ., Kansas City, Kansas 66160-7421, U .S .A . 0882-4010/92/070037+11 508 .00/0
© 1992 Academic Press Limited
38
S . K. De et al.
oxygen free radicals from neutrophils and macrophages' and other changes 5-7 that may ultimately lead to death from multi-organ failure . 8-10 It has been shown that these responses are mediated by the LPS-induced synthesis and release of a variety of powerful mediators including cytokines 10• t 1 and that they can be prevented by antibody to LPS 12 or LPS receptor ." LPS activated macrophages are the major source of several cytokines, including IL-1 (a and /3), IL-6 and TNF-« . 14-16 TNF-a plays an important role in mediating responses to LPS . In several species, administration of TNF-a elicits 17,18 many of the pathophysiological effects of LPS, whereas passive immunity to TNF-a prevents LPS-induced organ failure and death . 19-21 IL-1 and TNF-a exhibit overlapping biological activities 15 and an IL-1 receptor antagonist 22,23 that inhibits some LPS-mediated effects 24 has recently been identified . Furthermore, passive immunity to IL-6 may also protect against LPS and TNF-a induced lethality in mice . 25 These findings, and other data, 26,27 indicate that multiple cytokines are involved in the pathophysiology of septic shock . An important physiological response to LPS is the induction of hepatic acute-phase protein synthesis . 2S The acute-phase response represents a protective mechanism designed to restore homeostasis after LPS challenge . Much attention has focused on the expression of acute-phase plasma protein genes which in primary cultures of rat and mouse hepatocytes are regulated by IL-6 and/or IL-1 and glucocorticoids . 29.30 However, intracellular acute-phase proteins have also been described . Among these is the cysteine-rich metal-binding protein metallothionein (MT) ." The mouse MT genes are induced in vivo in the liver by IL-6, IL-1 and TNF-a . 32 In rat primary hepatocyte cultures, IL-6 and low levels of zinc have been shown to synergize with glucocorticoids in the regulation of MT gene expression ." A great deal of interest exists in identifying agents potentially therapeutic for septic shock . One such agent may be hydrazine sulfate (HS) . Pretreatment with HS has been shown to protect mice against the lethal effects of LPS and TNF-« . 34 .35 Recently, isoniazid, which is metabolized to hydrazine, has also been reported to protect mice against LPS . 36 The mechanisms through which hydrazine exerts its protective effects against LPS and TNF-a have not been elucidated . It has been shown, however, that HS protection against LPS is pituitary-dependent, and may, in part, be glucocorticoidmediated . 34 s 5 Urbaschek et al. 36 reported that isoniazid does not effect the synthesis and release of TNF-a from human monocytes, but it has been reported that hydrazine can modulate certain in vitro activities of this cytokine . 37 The purpose of experiments reported herein was to explore several potential modes of action of HS in protecting against LPS-induced lethality . Specifically, the liver was used as a model system in which to ascertain the effects of pretreatment of mice with HS on LPS-mediated induction of cytokine gene expression and the subsequent cytokine- mediated induction of hepatic MT gene expression . Results LPS-mediated cell-type specific induction of cytokine mRNAs and MT-1 mRNA in the mouse liver The adult mouse liver was employed as a model system in which to examine the mechanisms of action of HS . In situ hybridization was used to localize IL-1 fl mRNA . This mRNA serves predominantly as a marker for activated macrophage (Kupffer cell) gene expression, since IL-1/3 represents 1-2% of the macrophage mRNA following LPS stimulation ." MT-I mRNA served as a marker for hepatocyte acute-phase gene expression . 31 s 2 Livers were collected 1 .5 h after an injection of a high (100,ug/mouse) but non-lethal dose of LPS . This time after LPS injection was analysed because near
HS effects on hepatic cytokine and acute-phase gene expression
39
maximal levels of IL-1/3 mRNA are detected in the liver at this time . MT-I mRNA levels have begun to increase after a 30 min to 1 h lag period after LPS injection . 32 Livers from control and LPS-treated mice were perfusion fixed, paraffin embedded and sectioned . Sections from control and LPS-treated liver were mounted onto the same slides . In situ hybridization to IL-1/3 mRNA detected only basal levels of autoradiographic signal in sections from control liver (Fig . 1A) . In contrast, IL-1/3 mRNA hybridization signals were increased substantially in sections taken from livers obtained 1 .5 h after an LPS injection . Hybridization with a sense strand IL-1/i probe, as a control for specificity of the hybridization, revealed only background levels of autoradiographic signals in both control and LPS-treated livers (data not shown) . In situ hybridization to IL-1# mRNA in LPS-treated liver occurred in lines of cells [arrow heads in Fig . 1 A(d)], and in a pattern consistent with a location in the sinusoidal spaces, as well as in groups of lymphoid cells (Fig . 1A) . In situ hybridization to MT-I mRNA detected low levels of autoradiographic signal throughout the hepatic parenchyma in control tissue (Fig . 113), whereas in LPS-treated liver the autoradiographic signal was elevated throughout the parenchyma, and was particularly high in hepatocytes near the central vein (Fig . 113) . These results establish that cell-specific expression IL-1(3
(A)
Control
LPS
Fig . 1 . In situ hybridization of IL-1fi mRNA and MT-I mRNA in mouse liver after an injection of LPS . Mice were injected with LPS (100 ag/mouse) and 1 .5 h after injection they were anesthetized and livers were paraformaldehyde fixed and paraffin embedded . Sections from several control and LPS-treated livers were mounted onto the same slide, and slides were hybridized with 35 S-labeled cRNA or sense RNA probes for lL-1/3 mRNA (A) and MT-I mRNA (B) . RNase A resistant hybrids were detected by autoradiography . Exposure times were 63 days and 9 days for IL-1 /f and MT-I, respectively . Slides were post-stained lightly with hematoxylin . Only results from the cRNA probes are shown . (a) and (c), Bright-field photomicrographs (100x magnification) ; (b) and (d), dark-field photomicrographs (100x magnification) of (a) and (c), respectively . Autoradiographic grains appear as white dots. (a) and (b), Control ; (c) and (d), LPS-treated liver. CV, central vein ; L, lymphatic tissue .
40
S. K.
De
et al.
MT-I (B)
Control
'LPS
Fig . 1 . Continued.
of the IL-1/3 and MT-I genes occurs in the liver, and are consistent with the LPSmediated induction of IL-1# mRNA in lymphoid cells and induction of MT-I mRNA in hepatocytes . The cell-specific expression of hepatic cytokine and MT genes was further documented by examining the effects of the hepatotoxic agent D-GaIN on LPS-mediated induction of mRNA levels (Fig . 2) . D-GaIN exerts hepatotoxic effects due to depletion of uridine intermediates leading to diminished RNA and protein synthesis .'"' These effects occur in some hepatoma cell lines, in primary cultures of hepatocytes and in vivo in the liver, but were not expected to occur in leukocytes . Furthermore, D-GaIN dramatically sensitizes mice to LPS-induced lethal ity . 42,43 Mice were injected with LPS plus or minus D-GaIN . In this experiment, a dosage of LPS (0 .1 jig) was employed that is more than 10-fold the lethal dose for mice in the D-GaIN model . 42,34 Total hepatic RNA was extracted at the indicated times after injection . Hepatic RNA was analysed by Northern blotting for TNF-a, IL-1a and MT-I mRNAs (Fig . 2) . Consistent with earlier studies, 32 it was found that an injection of LPS caused a rapid transient increase in the levels of hepatic IL-1# and TNF-a mRNAs which peaked by 1 h and declined by 4 h (Fig . 2) . LPS-mediated induction of these hepatic cytokine mRNAs was unaltered by D-GaIN . Induction of hepatic MT-I mRNA, after an LPS injection, persisted for several hours in control mice, but was largely prevented by D-GaIN (Fig . 2) . Thus, analysis of total RNA extracted from the liver provides information on cytokine gene expression in lymphoid cells, presumably Kupffer cells, as well as on acute-phase gene expression in the hepatocyte . Taken together, these data suggest that the liver is a suitable model system in which to examine the mechanisms of action of HS .
41
HS effects on hepatic cytokine and acute-phase gene expression +D-GaIN
TNF-a
Hours after LPS
MT
0
I
2
4
0
I
2
4 0
2 -
2 +
4 -
4 + D-GaIN
Fig . 2 . Northern blot detection of mouse hepatic IL-1/3, TNF-a and MT-I mRNAs after an LPS injection plus or minus o-GaIN . Total hepatic RNA was prepared at the indicated times after an i .p . injection of LPS alone (-) or in combination (+) with n-GaIN . RNA was fractionated by formaldehyde-agarose gel electrophoresis and Northern blotted to nitrocellulose filters . Filters were hybridized with 32 P-labeled cRNA probes for mouse IL-1 fl, TNF-a or MT-I . After hybridization the filters were washed under stringent conditions (68°C in 45 mm NaCl) and the hybrids were detected by autoradiography . MT-I hybrids were detected after 3 h, IL-1p hybrids after 18 hand TNF-a hybrids after 36 h of autoradiography . Only relevant portions of the blots are shown .
Analysis of the effects of HS on LPS-mediated induction of lL-1J3 and TNF-a genes in mouse liver A mechanism by which HS pretreatment may protect mice against the lethal effects of LPS 34,35 could involve the prevention or attenuation of LPS-mediated induction of cytokine gene expression . In order to test this possibility, total hepatic RNA was obtained from mice pretreated with HS and subsequently challenged with LPS . In these experiments, a dosage of LPS (200 µg) at the LD 50 was employed . 35 These RNA samples were compared with those from control animals that were not pretreated with HS prior to the LPS injection . RNA samples were analysed for TNF-a and IL-1 J3 mRNA by Northern blotting (Fig . 3) . Pretreatment with HS did not significantly influence either the timing or the levels of LPS-mediated induction of IL-1 J3 or TNF-a mRNAs in the liver . As reported previously, 32 hepatic TNF-a mRNA and IL-1Ji mRNA levels were each rapidly and transiently induced by LPS (Fig . 3) . Another conceivable mechanism by which HS may influence LPS lethality could involve establishment of a state of tolerance . Sublethal doses of LPS or cytokines can induce transient refractoriness to subsequent exposure to a lethal dose of these agents, a state termed tolerance . 44-48 However, an injection of HS, under the conditions described in Fig . 3, did not effect hepatic cytokine mRNA levels, which were detectable by Northern blotting only after a long period (7 days) of autoradiography (data not shown) . Furthermore, cytokine mRNA levels were not refractory to subsequent induction by LPS after HS pretreatment (Fig . 3) . Analysis of the effects of HS on LPS-mediated induction of hepatic acute-phase protein gene expression LPS mediates hepatic acute-phase protein synthesis" via induction of cytokines which in turn effect hepatocyte gene expression . 2930 Mouse MT genes are induced in vivo predominantly in the liver by IL-6 and IL-1, and to a lesser extent by TNF-a . 32 The possibility that HS may protect against LPS toxicity by influencing cytokine action, at this level, was explored by examining the effect of HS pretreatment on levels of hepatic
42
S . K . De et al.
TNF -a
HS Hours offer LPS
0
2
4
Fig . 3 . Effects of HS on LPS-mediated induction of IL-1# and TNF-a mRNAs in the liver. Mice were taken off feed and the next morning they were injected with HS (+) . These mice were then challenged with LPS (200 µg) . Control animals (-) were injected with vehicle (saline) . Hepatic total RNA was isolated at the indicated times after injection of LPS (0-4 h) and was analysed by Northern blot hybridization using TNF-a and IL-1 /l cRNA probes . Only relevant portions of the blots are shown .
MT mRNA before and after LPS challenge . Northern blot hybridization analysis of total hepatic RNA (Fig . 4) established that HS alone had no effect on MT mRNA levels, and HS pretreatment did not prevent LPS-mediated induction of this mRNA (Fig . 4) .
MT
HS
LPS
HS+LPS
Fig . 4 . Effect of HS on LPS-mediated induction of MT-1 mRNA levels in the liver . Experimental conditions were exactly as described in Fig . 3 . Livers were collected 4 h after LPS injection, and total RNA was extracted, fractionated by formaldehyde-agarose gel electrophoresis and analysed by Northern blot hybridization using an MT-I cRNA probe .
HS effects on hepatic cytokine and acute-phase gene expression
43
Discussion and conclusions This study was designed to examine several potential mechanisms by which HS protects against LPS-induced lethality in the mouse . The results establish that pretreatment with HS 5 h prior to LPS challenge does not influence constitutive or LPS-induced hepatic expression of cytokine genes (TNF-a and IL-1 fl) as measured by changes in mRNA levels . These cytokines have been associated with LPS-induced lethality . It was further noted that constitutive and LPS-induced expression of the mouse MT-I gene was unaffected by HS p retreatment . M T represents an example of the acute phase reactants that are regulated by cytokines and are thought to play a protective role in response to LPS . An explanation for HS protection against LPS that now appears unlikely based on the present data is the down-regulation of TNF-a or IL-1 gene expression, as defined by rapid induction of cytokine mRNA levels within the LPS-stimulated Kupffer cell . In related studies, it was also noted that LPS-induced cytokine gene expression in peritoneal macrophages is not attenuated by HS pretreatment (data not shown) . These findings are particularly noteworthy inasmuch as TNF-a has been shown to be an essential mediator of LPS lethality both in normal mice and in mice sensitized with DGaIN . 19.49 The data establish that HS does not antagonize the LPS-induced activation of macrophage gene expression, but cannot exclude the possibility that HS exerts some effects on the synthesis and/or release of cytokines from the macrophage . An argument against the possibility of a global inhibitory effect of HS on cytokine release from the macrophage is the finding that the induction of MT mRNA in hepatocytes is not prevented by HS pretreatment . Although several exogenously administered recombinant cytokines [TNF-a, IL-1 (a and J3) and IL-6] can induce mouse MT gene expression,32 IL-6 has been implicated as playing a pivotal role in hepatocyte acutephase MT gene expression . 32 • a 3 However, the possibility that HS may exert some effect on the synthesis and/or release of TNF-a from the macrophage cannot be excluded . Significant post-transcriptional regulation of TNF-a gene expression has been documented . 50 '51 However, Urbaschek et a/. 36 have shown that isoniazid does not influence the synthesis or release of TNF-a by human monocytes in vitro, but effects of HS were not reported, nor were the in vivo effects of pretreatment with isoniazid on subsequent LPS-induced TNF-a levels . Results presented herein also do not exclude the possibility that HS attenuates other aspects of the LPS-induced activation of lymphoid cells, such as the production of toxic oxygen radicals . It has been suggested that HS may protect against LPS-induced lethality via de novo hepatocyte protein synthesis . 34 .35 For example, pre-induction of the hepatic acute-phase proteins could be envisioned to provide protection from subsequent exposure to LPS . 28 This suggestion for a mechanism of HS action was based, in part, on the ability of HS pretreatment to essentially reverse the LPS-sensitizing effect of o-GaIN, a general inhibitor of hepatocyte RNA synthesis . Moreover, the presence of D-GaIN at the start of the HS pretreatment period completely abrogates HS protection against LPS . 34 A recent report suggests that induction of the acute-phase response by carbon tetrachloride protects mice from D-GaIN sensitization to LPS and TNF-a . 52 While our present findings do not support a generalized effect of HS on hepatocyte acute-phase gene expression, since no effect on MT mRNA levels was noted, the data do not exclude the possibility that HS induces other hepatocyte mRNAs and proteins specifically . Such a possibility might still conceivably contribute to HS protection against LPS lethality in normal mice and in mice sensitized with D-GaIN . The present study also provides confirmatory evidence of D-GaIN inhibition of hepatocyte RNA synthesis, but not inhibition of RNA synthesis in the neighboring
44
S . K . De et al.
Kupffer cell . In that regard, we have shown that, in LPS-challenged mice, D-GaIN abrogates MT mRNA synthesis, but does not affect LPS-induced expression of the TNF-a, IL-1a and IL-1ll genes . Moreover, we have demonstrated that, in the liver, MT mRNA is localized to the hepatocyte, whereas cytokine mRNA (IL-1 S) is localized to the Kupffer cell . In conclusion, it is proposed that HS protection against LPS lethality occurs subsequent to LPS-induced activation of macrophage gene transcription, and that it does not result from a generalized interference with cytokine actions on the liver or from a generalized induction of hepatocyte protein synthesis .
Materials and methods Chemicals. LPS was isolated from E. co/i 011 : B4 by a modification of the phenol-water method as described by Morrison and Leive53 and was provided by Dr David C . Morrison (University of Kansas Medical Center, Kansas City, KS) . LPS was dissolved by sonication in phosphate-buffered saline (PBS) . HS (Sigma Chemical Co ., St Louis, MO) was prepared fresh before use in pyrogen-free normal saline (8 mg HS/ml) and the solution was brought to neutral pH using 10 N NaOH . D-GaIN (Sigma Chemical Co .) was prepared fresh by dissolving in PBS at a concentration of 45 mg/ml . PBS was prepared fresh with pyrogen-free normal saline . Animals. CF, female mice (6-8 weeks ; 20-25 g) were obtained from Charles River Breeding Laboratories, Inc . (Wilmington, MA) and were maintained in the Animal Resources Facility with food and water ad libitum and 12 h light and dark cycles . In experiments involving HS, mice were taken off feed at 2200 h and injected intraperitoneally (i .p .) with 2 mg HS in 250 pl (80 mg HS per kg body weight), or with vehicle alone (pyrogen free saline) at 0800 h the next morning . LPS was injected at a dosage of 100 to 200 pg/mouse, as noted, 5 h after HS injection (1300 h) . In the experiments involving D-GaIN, mice were injected i .p. with 18 mg D-GaIN along with LPS (0 .1 pg) in 400 pl PBS, according to the procedure of Galanos et al. 12 These conditions of HS pretreatment have previously been shown to protect normal and D-GaINsensitized mice against the lethal effects of an acute exposure to LPS . 34.35 Liver samples were removed at the indicated times after injection, quick-frozen in liquid freon, and stored at -70°C until processed for RNA extraction . In each experiment for each time point shown, approximately equal amounts (0 .3 g) of the livers from three mice were pooled and flash frozen for extraction of RNA . Isolation of total hepatic RNA . Hepatic RNA was extracted using the sodium dodecylsulfate (SDS)-phenol-chloroform procedure described in detail by Andrews et al." Frozen samples were homogenized in 0 .5% SDS, 25 mm EDTA, 75 mm NaCl, pH 8 .0 (SDS-buffer), extracted with phenol, and then with phenol/chloroform :isoamyl alcohol (24 :1, v/v) . RNA was precipitated from the aqueous phase with 3 M ammonium acetate, and reprecipitated twice more to remove DNA . The RNA pellet was redissolved in water and precipitated with ethanol . Hybridization probes. Mouse cDNA clones for MT-I (R . D . Palmiter, University of Washington, Seattle, WA), IL-1/i (The DuPont Merck Pharmaceutical Co ., Glenolden, PA) and TNF-a (Genentech Inc ., San Francisco, CA) were inserted into the Sp6 or pGEM vectors (Promega Biotech, Madison, WI) and used as templates for the synthesis of 32 P- or 35S-labeled cRNA probes as described by Melton et al ." Both sense strand and antisense strand RNA probes were synthesized for use in 'the in situ hybridization experiments . Probes had specific activities of about 2 x 10 9 dpm/pg . Northern blot hybridization . Northern blot hybridization was performed as described previously . 32,3s,4o Briefly, RNA was denatured 5 min at 60°C in a solution of 1 x 3- (4-morpholino) propane sulfonic acid (MOPS) buffer (20 mm MOPS, 5 mm sodium acetate, 1 mm EDTA, pH 7 .0) containing 50% formamide and 2 .2 M formaldehyde. Denatured RNA (2 Yg for MT-1 blots and 6 pg for the cytokine blots) was fractionated by formaldehyde-1 .5% agarose gel electrophoresis ." Gels were soaked for 40 min in 10 mm sodium phosphate, pH 7 .0, and transferred to nitrocellulose in the presence of 20xSSC (3 M NaCl and 0 .3 M sodium citrate, pH 7 .4) . Following transfer, the filters were baked in a vacuum oven at 75°C for 5 h . Northern
HS effects on hepatic cytokine and acute-phase gene expression
45
blots were pre-hybridized, hybridized, and washed as described in detail by Andrews et al . 14 In all experiments, duplicate gels were stained with acridine orange to ensure integrity of the RNA sample and to confirm that equal amounts of RNA had been loaded onto each lane . In situ hybridization. The details of these techniques have been described previously . 32,39,40,57 Mice were anesthetized with Avertin, and perfusion-fixed livers were removed, cut into small pieces and further fixed with 4% paraformaldehyde in PBS for 2 h . Livers were embedded in paraffin blocks and serially sectioned at 7 µm . Sections from four separate control and four separate treated livers were placed on the same poly-l-lysine coated slides . Multiple slides were hybridized with 35 S-labeled MT-I and IL-1/3 cRNA (antisense) probes . Control slides were hybridized with 35S-labeled MT-I and IL-1 /3 sense strand RNA probes . After hybridization, slides were washed and treated with RNase A to reduce the background and ensure specificity of the remaining hybrids as described ,32,39,40 Hybrids were detected by autoradiography using Kodak NTB-2 liquid emulsion (Eastman Kodak, Rochester, NY) . Autoradiographic exposure times for MT-I and IL-1$ hybrids were 9 and 63 days, respectively . Slides were post-stained lightly in hematoxylin .
This study was supported, in part, by grants (ES04725 to GKA and A130500 to RS) from the NIH . SKD is a Postdoctoral Fellow supported by the Wesley Foundation Scholars Program .
References 1 . Lefer AM . Significance of lipid mediators in shock states . Circ Shock 1989; 27 : 3-12 . 2 . West RL, Elovitz MJ, Hardaway RM . Blood coagulation factor activity in experimental endotoxemia . Ann Surg 1966 ; 163 : 567-72 . 3 . Snedegard G, Lui L, Hugli TE . Endotoxin induced shock in the rat . A role of C5a . Am J Pathol 1989 ; 135:489-97 . 4 . Flohe L, Giertz H . Endotoxins, arachidonic acid, and superoxide formation . In: Urbaschek B, ed . Perspectives on bacterial pathogenesis and host defense . Chicago : University of Chicago Press, 1988 ; 123-31 . 5 . Doebber TW, Wu MS, Robbins JC, Choy BM et al. Platelet-activating factor (PAF) involvement in endotoxin-induced hypotension . Studies with PAF receptor antagonist kadsurenone . Biochem Biophys Res Commun 1985; 127 : 799-808 . 6 . Fritz H, Jochum M, Duswald KH et al. Granulocyte proteinases as mediators of unspecific proteolysis in inflammation : a review . In : Tschesche H, ed . Proteinases in inflammation and tumor invasion . Berlin : Walter de Gruyter, 1986 : 1-23 . 7 . Carlos TM, Harlan JM . Membrane proteins involved in phagocyte adherence to endothelium . Immunol Rev 1990 ; 114: 5-28 . 8 . Kreger BE, Craven DE, Carling PC et a/. Gram-negative bacteremia . III . Reassessment of etiology, epidemiology and ecology in 612 patients. Am J Med 1980 ; 68 : 332-7 . 9 . Wolff SM, Bennett JV . Gram-negative rod bacteremia (editorial) . N Engl J Med 1974; 291 : 733-4 . 10 . Morrison DC, Ryan JL. Endotoxins and disease mechanisms . Ann Rev Med 1987 ; 38 : 417-32 . 11 . Beutler B, Cerami A . Cachectin : more than a tumor necrosis factor . N Engl J Med 1987 ; 316 : 379-85 . 12 . Ziegler EJ, Fisher CL Jr, Sprung CL et al. Treatment of gram-negative bacteremia and septic shock with HA-1 A human monoclonal antibody against endotoxin . N Engl J Med 1991 ; 324 : 429-36 . 13 . Morrison DC, Silverstein R, Bright SW et al . Monoclonal antibody to mouse lipopolysaccharide receptor protects mice against the lethal effects of endotoxin . J Infect Dis 1990 ; 162 : 1063-8 . 14 . Arai K-i, Lee F, Miyajima A et a/. Cytokines : coordinators of immune and inflammatory processes . Ann Rev Biochem 1990; 59 : 783-836 . 15 . Akira S, Hirano T, Taga T et al. Biological multifunctional cytokines : IL-6 and related molecules (IL-1 and TNF) . FASEB J 1990 ; 4 : 2860-7 . 16 . Whicher JT, Evans SW. Cytokines and disease . Clin Chem 1990; 36 :1269-81 . 17 . Bauss F, Droge W, Mannel DN . Tumor necrosis factor mediates endotoxic effects in mice . Infect Immun 1987 ; 55 :1622-5 . 18 . Tracey KJ, Beutler B, Lowry SF et al. Shock and tissue injury induced by recombinant human cachectin . Science 1986 ; 234 : 470-4 . 19 . Beutler B, Milsark IW, Cerami A . Passive immunization against cachectin/tumor necrosis factor protects mice from lethal effects of endotoxin . Science 1985; 229 : 869-971 . 20 . Tracey KJ, Fong Y, Hesse DG et al. Anti-cachectin/TNF monoclonal antibodies prevent septic shock during lethal bacteremia . Nature 1987 ; 330 : 662-4 . 21 . Sherry BA, Gelin J, Fong Y et al. Anticachectin/tumor necrosis factor-a antibodies attenuate development of cachexia in tumor models . FASEB J 1989; 3 : 1956-62 .
46
S . K . De et al.
22 . Hannum CH, Wilcox CJ, Arend WP et al. Interleukin-1 receptor antagonist activity of a human interleukin-1 inhibitor . Nature 1990 ; 343 : 336-40. 23 . Eisenberg SP, Evans RJ, Arend WP et al. Primary structure and functional expression from complementary DNA of a human interleukin-1 receptor antagonist . Nature 1990 ; 343 : 341-6 . 24 . Henricson BE, Neta R, Vogel SN . An interleukin-1 receptor antagonist blocks lipopolysaccharideinduced colony-stimulating factor production and early endotoxin tolerance . Infect Immun 1991 ; 59 : 1188-91 . 25 . Starnes FH, Pearce MK, Tewari A et al. Anti IL-6 monoclonal antibodies protect against lethal Escherichia coli infection and lethal tumor necrosis factor-a challenge in mice . J Immunol 1990; 145 : 4185-91 . 26 . Zuckerman SH, Evans GF, Butler LD . Endotoxin tolerance : independent regulation of interleukin-1 and tumor necrosis factor expression . Infect Immun 191 ; 59 : 2774-80 . 27 . Franks AK, Kujawa KI, Yaffe LJ . Experimental elimination of tumor necrosis factor in low-dose endotoxin models has variable effects on survival . Infect Immun 1991 ; 59 : 2609-14. 28 . Koj A . In : Gordon AH, Koj A, eds . The acute phase response to injury and infection . Amsterdam : Elsevier Science Publishers BV. 1985 ; 139-232. 29 . Prowse KR, Baumann H . Interleukin-1 and interleukin-6 stimulate acute-phase protein production in primary mouse hepatocytes. J Leukocyte Biol 1989 ; 45 : 55-61 . 30 . Andus T, Geiger T, Hirano T et al. Regulation of synthesis and secretion of major rat acute-phase proteins by recombinant human interleukin-6 (BSF-2/IL-6) in hepatocyte primary cultures . Eur J Biochem 1988 ; 173 : 287-93 . 31 . Andrews GK. Regulation of metallothionein gene expression . Prog Food Nutr Sci 1990 ; 14 : 193258 . 32 . De SK, McMaster M, Andrews GK . Endotoxin induction of murine metallothionein genes . J Biol Chem 1990; 265 : 15 267-74 . 33 . Schroeder JJ, Cousins RJ . Interleukin-6 regulates metallothionein gene expression and zinc metabolism in hepatocyte monolayer cultures . Proc Natl Acad Sci USA 1990; 87 : 3137-41 . 34 . Silverstein R, Turley BR, Christoffersen et al. Hydrazine sulfate protects D-galactosamine-sensitized mice against endotoxin and tumor necrosis factor/cachectin lethality : evidence of a role for the pituitary . J Exp Med 1991 ; 173 : 357-65 . 35 . Silverstein R, Christoffersen CA, Morrison DC . Modulation of endotoxin lethality in mice by hydrazine sulfate . Infect Immun 1989 ; 57 : 2072-8. 36 . Urbaschek R, Mannel DN, Urbanczik R . Isoniazid protects mice from endotoxin lethality without influencing tumor necrosis factor synthesis and release. Antimicrob Agents Chemother 1991 ; 35 :16668. 37 . Hughes TK, Cadet P, Lamed CS . Modulation of tumor necrosis factor activities by a potential anticachexia compound, hydrazine sulfate . Int J Immunopharmacol 1989; 11 : 501-7 . 38 . Dinarello CA . The biology of interleukin-1 and comparison to tumor necrosis factor . Immunol Lett 1987 ;227-32 . 39 . De SK, McMaster MT, Day SK et al. Cell-specific metallothionein gene expression in mouse decidua and placentae . Development 1989; 107 : 611-21 . 40. De SK, Enders GC, Andrews GK . High levels of metallothionein mRNAs in male germ cells of the adult mouse. Mol Endocrinol 1991 ; 5 : 628-36 . 41 . Decker K, Keppler D . Galactosamine hepatitis : key role of the nucleotide deficiency period in the pathogenesis of cell injury and cell death . Rev Physiol Biochem Pharmacol 1974; 71 : 78-106 . 42 . Galanos C, Freudenberg MA, Reutter W . Galactosamine-induced sensitization to the lethal effects of endotoxin . Proc Natl Acad Sci USA 1979; 76 : 5939-43 . 43 . Freudenberg MA, Keppler D, Galanos C . Requirement for lipopolysaccharide-responsive macrophages in galactosamine-induced sensitization to endotoxin . Infect Immun 1986 ; 891-895 . 44 . Galanos C, Freudenberg MA, Coumbos A et al Induction of lethality and tolerance by endotoxin are mediated by macrophages through tumor necrosis factor . In : Bonavida B, Gifford GE, Kirchner H, Old LJ, eds . International conference on tumor necrosis factor/cachectin and related cytokines . Heidelberg : S . Karger, 1988 ; 114-27 . 45 . Freudenberg MA, Galanos C . Induction of tolerance to lipopolysaccharide (LPS)-D-galactosamine lethality by pretreatment with LPS is mediated by macrophages . Infect Immun 1988 ; 56: 1352-7 . 46 . Sanchez-Cantu L, Rode HN, Christou NV . Endotoxin tolerance is associated with reduced secretion of tumor necrosis factor . Arch Surg 1989; 124 : 1432-6 . 47 . Van der Meer JWM, Helle M, Aarden A . Comparison of the effect of recombinant interleukin-6 and recombinant interleukin-1 on nonspecific resistance to infection . Eur J Immunol 1989 ; 19 : 413-6 . 48 . Fraker DL, Stovroff MC, Merino MJ et al. Tolerance to tumor necrosis factor in rats and the relationship to endotoxin tolerance and toxicity . J Exp Med 1988 ; 168 : 95-105 . 49 . Galanos C, Freudenberg MA . Tumor necrosis factor mediates endotoxin shock : the protective effects of antibodies and cortisone . In : Bonavida B, Grainger G, eds . Tumor necrosis factor : structure, mechanism of action, role in disease and therapy . Basel : S . Karger, 1990 ; 187-93. 50 . Beutler B, Krochin N, Milsark IW et al. Control of cachectin (tumor necrosis factor) synthesis : mechanisms of endotoxin resistance . Science 1986 ; 232 : 977-80 . 51 . Zuckerman SH, Evans GF, Snyder YM et al. Endotoxin-macrophage interaction : post-translation al regulation of tumor necrosis factor expression . J Immunol 1989 ; 143 : 1223-7 .
HS effects on hepatic cytokine and acute-phase gene expression
47
52 . Alcorn JM, Fierer J, Chojkier M . The acute-phase response protects mice from D-galactosamine sensitization to endotoxin and tumor necrosis factor-a . Hepatology 1992; 15 : 122-9 . 53 . Morrison DC, Leive L. Fractions of lipopolysaccharide from Escherichia coli 0111 : B4 prepared by two extraction procedures . J Biol Chem 1975 ; 250: 2911-19 . 54 . Andrews GK, Lehman LD, Huet YM et al. Metallothionein gene regulation in the preimplantation rabbit blastocyst. Development 1987 ; 100: 463-5 . 55 . Melton DA, Krieg PA, Rebagliati MR et al. Efficient in vitro synthesis of biologically active RNA and RNA hybridization probes . from plasmid containing a bacteriophage SP6 promoter . Nucl Acids Res 1984 ;12 :7035-57 . 56 . Lehrach H, Diamond D, Wozney JM et al RNA molecular weight determinations by gel electrophoresis under denaturing conditions, a critical reexamination . Biochemistry 1977 ; 16 : 4743-51 . 57 . Huet-Hudson YM, Chakraborty CC, De SK et al. Estrogen regulates synthesis of EGF in mouse uterine epithelial cells . Mol Endocrinol 1990 ; 4: 510-23 .
Hydrazine sulfate protection against endotoxin lethality : analysis of effects on expression of hepatic cytokine genes and an acute-phase gene Swapan K . De, Richard Silverstein and Glen K . Andrews* Department of Biochemistry and Molecular Biology, University of Kansas Medical Center, 39th and Rainbow Blvd., Kansas City, KS 66103, U .S .A . (Received February 6, 1992 ; accepted in revised form April 6, 1992)
De, S . K . (Dept of Biochemistry and Molecular Biology, University of Kansas Medical Center, 39th and Rainbow Blvd ., Kansas City, KS 66103, U .S .A .), R . Silverstein and G . K . Andrews . Hydrazine sulfate protection against endotoxin lethality : analysis of effects on expression of hepatic cytokine genes and an acute-phase gene . Microbial Pathogenesis 1992 ; 13 : 37-47 . Hydrazine sulfate (HS) pretreatment protects mice against the lethal effects of bacterial endotoxin lipopolysaccharide (LPS) through mechanisms yet to be established . The liver was examined as a model organ to determine HS effects on (a) LPS activation of leukocyte (Kupffer cell) interleukin-1[J (IL-1 fl) and tumor necrosis factor-7 (TNF-a) genes and (b) subsequent cytokine-mediated induction of the acute-phase response as measured by hepatic metallothionein (MT) gene expression . The utility of this model was documented by in situ hybridization which showed that acute induction by LPS of the IL-1/f gene occurred in cells found in liver sinusoids, consistent with Kupffer cells, whereas induction of the MT gene occurred in hepatocytes . The cell specific expression of these genes was further verified by Northern blot hybridization to LPS-treated liver RNA which showed that the LPS-mediated increase in hepatic cytokine mRNA levels, unlike that of MT, was not prevented by Dgalactosamine (D-GaIN) treatment . Northern blot hybridization established that HS pretreatment did not block the acute induction of hepatic cytokine mRNAs (IL-1R and TNF-a) by LPS nor did it induce these cytokine mRNAs in the absence of LPS . Northern blot hybridization further established that HS did not prevent LPS-mediated activation of hepatocyte MT gene expression . Thus, HS does not prevent LPS from activating liver leukocytes . These results also suggest that HS pretreatment neither prevents the general release of cytokines from LPS activated leukocytes nor the general induction of acute-phase protein gene expression in hepatocytes . Therefore, it is unlikely that HS protection against LPS-mediated lethality entails a global down-regulation of LPS-mediated inflammatory events . Key words : hydrazine; lipopolysaccharide ; endotoxemia ; metallothionein ; tumor necrosis factor ; interleukin ; mouse liver .
Introduction Gram-negative bacterial LPS initiates a series of pathophysiological changes including impaired gas exchange of the lung and formation of lung edema,' activation of the plasmatic coagulation cascade and fibrinolysis, Z complement activation, :' formation of
*Author to whom correspondence should be addressed at : Department of Biochemistry and Molecular Biology, WHE 4018, University of Kansas Medical Center, 39th and Rainbow Blvd ., Kansas City, Kansas 66160-7421, U .S .A . 0882-4010/92/070037+11 508 .00/0
© 1992 Academic Press Limited
38
S . K. De et al.
oxygen free radicals from neutrophils and macrophages' and other changes 5-7 that may ultimately lead to death from multi-organ failure . 8-10 It has been shown that these responses are mediated by the LPS-induced synthesis and release of a variety of powerful mediators including cytokines 10• t 1 and that they can be prevented by antibody to LPS 12 or LPS receptor ." LPS activated macrophages are the major source of several cytokines, including IL-1 (a and /3), IL-6 and TNF-« . 14-16 TNF-a plays an important role in mediating responses to LPS . In several species, administration of TNF-a elicits 17,18 many of the pathophysiological effects of LPS, whereas passive immunity to TNF-a prevents LPS-induced organ failure and death . 19-21 IL-1 and TNF-a exhibit overlapping biological activities 15 and an IL-1 receptor antagonist 22,23 that inhibits some LPS-mediated effects 24 has recently been identified . Furthermore, passive immunity to IL-6 may also protect against LPS and TNF-a induced lethality in mice . 25 These findings, and other data, 26,27 indicate that multiple cytokines are involved in the pathophysiology of septic shock . An important physiological response to LPS is the induction of hepatic acute-phase protein synthesis . 2S The acute-phase response represents a protective mechanism designed to restore homeostasis after LPS challenge . Much attention has focused on the expression of acute-phase plasma protein genes which in primary cultures of rat and mouse hepatocytes are regulated by IL-6 and/or IL-1 and glucocorticoids . 29.30 However, intracellular acute-phase proteins have also been described . Among these is the cysteine-rich metal-binding protein metallothionein (MT) ." The mouse MT genes are induced in vivo in the liver by IL-6, IL-1 and TNF-a . 32 In rat primary hepatocyte cultures, IL-6 and low levels of zinc have been shown to synergize with glucocorticoids in the regulation of MT gene expression ." A great deal of interest exists in identifying agents potentially therapeutic for septic shock . One such agent may be hydrazine sulfate (HS) . Pretreatment with HS has been shown to protect mice against the lethal effects of LPS and TNF-« . 34 .35 Recently, isoniazid, which is metabolized to hydrazine, has also been reported to protect mice against LPS . 36 The mechanisms through which hydrazine exerts its protective effects against LPS and TNF-a have not been elucidated . It has been shown, however, that HS protection against LPS is pituitary-dependent, and may, in part, be glucocorticoidmediated . 34 s 5 Urbaschek et al. 36 reported that isoniazid does not effect the synthesis and release of TNF-a from human monocytes, but it has been reported that hydrazine can modulate certain in vitro activities of this cytokine . 37 The purpose of experiments reported herein was to explore several potential modes of action of HS in protecting against LPS-induced lethality . Specifically, the liver was used as a model system in which to ascertain the effects of pretreatment of mice with HS on LPS-mediated induction of cytokine gene expression and the subsequent cytokine- mediated induction of hepatic MT gene expression . Results LPS-mediated cell-type specific induction of cytokine mRNAs and MT-1 mRNA in the mouse liver The adult mouse liver was employed as a model system in which to examine the mechanisms of action of HS . In situ hybridization was used to localize IL-1 fl mRNA . This mRNA serves predominantly as a marker for activated macrophage (Kupffer cell) gene expression, since IL-1/3 represents 1-2% of the macrophage mRNA following LPS stimulation ." MT-I mRNA served as a marker for hepatocyte acute-phase gene expression . 31 s 2 Livers were collected 1 .5 h after an injection of a high (100,ug/mouse) but non-lethal dose of LPS . This time after LPS injection was analysed because near
HS effects on hepatic cytokine and acute-phase gene expression
39
maximal levels of IL-1/3 mRNA are detected in the liver at this time . MT-I mRNA levels have begun to increase after a 30 min to 1 h lag period after LPS injection . 32 Livers from control and LPS-treated mice were perfusion fixed, paraffin embedded and sectioned . Sections from control and LPS-treated liver were mounted onto the same slides . In situ hybridization to IL-1/3 mRNA detected only basal levels of autoradiographic signal in sections from control liver (Fig . 1A) . In contrast, IL-1/3 mRNA hybridization signals were increased substantially in sections taken from livers obtained 1 .5 h after an LPS injection . Hybridization with a sense strand IL-1/i probe, as a control for specificity of the hybridization, revealed only background levels of autoradiographic signals in both control and LPS-treated livers (data not shown) . In situ hybridization to IL-1# mRNA in LPS-treated liver occurred in lines of cells [arrow heads in Fig . 1 A(d)], and in a pattern consistent with a location in the sinusoidal spaces, as well as in groups of lymphoid cells (Fig . 1A) . In situ hybridization to MT-I mRNA detected low levels of autoradiographic signal throughout the hepatic parenchyma in control tissue (Fig . 113), whereas in LPS-treated liver the autoradiographic signal was elevated throughout the parenchyma, and was particularly high in hepatocytes near the central vein (Fig . 113) . These results establish that cell-specific expression IL-1(3
(A)
Control
LPS
Fig . 1 . In situ hybridization of IL-1fi mRNA and MT-I mRNA in mouse liver after an injection of LPS . Mice were injected with LPS (100 ag/mouse) and 1 .5 h after injection they were anesthetized and livers were paraformaldehyde fixed and paraffin embedded . Sections from several control and LPS-treated livers were mounted onto the same slide, and slides were hybridized with 35 S-labeled cRNA or sense RNA probes for lL-1/3 mRNA (A) and MT-I mRNA (B) . RNase A resistant hybrids were detected by autoradiography . Exposure times were 63 days and 9 days for IL-1 /f and MT-I, respectively . Slides were post-stained lightly with hematoxylin . Only results from the cRNA probes are shown . (a) and (c), Bright-field photomicrographs (100x magnification) ; (b) and (d), dark-field photomicrographs (100x magnification) of (a) and (c), respectively . Autoradiographic grains appear as white dots. (a) and (b), Control ; (c) and (d), LPS-treated liver. CV, central vein ; L, lymphatic tissue .
40
S. K.
De
et al.
MT-I (B)
Control
'LPS
Fig . 1 . Continued.
of the IL-1/3 and MT-I genes occurs in the liver, and are consistent with the LPSmediated induction of IL-1# mRNA in lymphoid cells and induction of MT-I mRNA in hepatocytes . The cell-specific expression of hepatic cytokine and MT genes was further documented by examining the effects of the hepatotoxic agent D-GaIN on LPS-mediated induction of mRNA levels (Fig . 2) . D-GaIN exerts hepatotoxic effects due to depletion of uridine intermediates leading to diminished RNA and protein synthesis .'"' These effects occur in some hepatoma cell lines, in primary cultures of hepatocytes and in vivo in the liver, but were not expected to occur in leukocytes . Furthermore, D-GaIN dramatically sensitizes mice to LPS-induced lethal ity . 42,43 Mice were injected with LPS plus or minus D-GaIN . In this experiment, a dosage of LPS (0 .1 jig) was employed that is more than 10-fold the lethal dose for mice in the D-GaIN model . 42,34 Total hepatic RNA was extracted at the indicated times after injection . Hepatic RNA was analysed by Northern blotting for TNF-a, IL-1a and MT-I mRNAs (Fig . 2) . Consistent with earlier studies, 32 it was found that an injection of LPS caused a rapid transient increase in the levels of hepatic IL-1# and TNF-a mRNAs which peaked by 1 h and declined by 4 h (Fig . 2) . LPS-mediated induction of these hepatic cytokine mRNAs was unaltered by D-GaIN . Induction of hepatic MT-I mRNA, after an LPS injection, persisted for several hours in control mice, but was largely prevented by D-GaIN (Fig . 2) . Thus, analysis of total RNA extracted from the liver provides information on cytokine gene expression in lymphoid cells, presumably Kupffer cells, as well as on acute-phase gene expression in the hepatocyte . Taken together, these data suggest that the liver is a suitable model system in which to examine the mechanisms of action of HS .
41
HS effects on hepatic cytokine and acute-phase gene expression +D-GaIN
TNF-a
Hours after LPS
MT
0
I
2
4
0
I
2
4 0
2 -
2 +
4 -
4 + D-GaIN
Fig . 2 . Northern blot detection of mouse hepatic IL-1/3, TNF-a and MT-I mRNAs after an LPS injection plus or minus o-GaIN . Total hepatic RNA was prepared at the indicated times after an i .p . injection of LPS alone (-) or in combination (+) with n-GaIN . RNA was fractionated by formaldehyde-agarose gel electrophoresis and Northern blotted to nitrocellulose filters . Filters were hybridized with 32 P-labeled cRNA probes for mouse IL-1 fl, TNF-a or MT-I . After hybridization the filters were washed under stringent conditions (68°C in 45 mm NaCl) and the hybrids were detected by autoradiography . MT-I hybrids were detected after 3 h, IL-1p hybrids after 18 hand TNF-a hybrids after 36 h of autoradiography . Only relevant portions of the blots are shown .
Analysis of the effects of HS on LPS-mediated induction of lL-1J3 and TNF-a genes in mouse liver A mechanism by which HS pretreatment may protect mice against the lethal effects of LPS 34,35 could involve the prevention or attenuation of LPS-mediated induction of cytokine gene expression . In order to test this possibility, total hepatic RNA was obtained from mice pretreated with HS and subsequently challenged with LPS . In these experiments, a dosage of LPS (200 µg) at the LD 50 was employed . 35 These RNA samples were compared with those from control animals that were not pretreated with HS prior to the LPS injection . RNA samples were analysed for TNF-a and IL-1 J3 mRNA by Northern blotting (Fig . 3) . Pretreatment with HS did not significantly influence either the timing or the levels of LPS-mediated induction of IL-1 J3 or TNF-a mRNAs in the liver . As reported previously, 32 hepatic TNF-a mRNA and IL-1Ji mRNA levels were each rapidly and transiently induced by LPS (Fig . 3) . Another conceivable mechanism by which HS may influence LPS lethality could involve establishment of a state of tolerance . Sublethal doses of LPS or cytokines can induce transient refractoriness to subsequent exposure to a lethal dose of these agents, a state termed tolerance . 44-48 However, an injection of HS, under the conditions described in Fig . 3, did not effect hepatic cytokine mRNA levels, which were detectable by Northern blotting only after a long period (7 days) of autoradiography (data not shown) . Furthermore, cytokine mRNA levels were not refractory to subsequent induction by LPS after HS pretreatment (Fig . 3) . Analysis of the effects of HS on LPS-mediated induction of hepatic acute-phase protein gene expression LPS mediates hepatic acute-phase protein synthesis" via induction of cytokines which in turn effect hepatocyte gene expression . 2930 Mouse MT genes are induced in vivo predominantly in the liver by IL-6 and IL-1, and to a lesser extent by TNF-a . 32 The possibility that HS may protect against LPS toxicity by influencing cytokine action, at this level, was explored by examining the effect of HS pretreatment on levels of hepatic
42
S . K . De et al.
TNF -a
HS Hours offer LPS
0
2
4
Fig . 3 . Effects of HS on LPS-mediated induction of IL-1# and TNF-a mRNAs in the liver. Mice were taken off feed and the next morning they were injected with HS (+) . These mice were then challenged with LPS (200 µg) . Control animals (-) were injected with vehicle (saline) . Hepatic total RNA was isolated at the indicated times after injection of LPS (0-4 h) and was analysed by Northern blot hybridization using TNF-a and IL-1 /l cRNA probes . Only relevant portions of the blots are shown .
MT mRNA before and after LPS challenge . Northern blot hybridization analysis of total hepatic RNA (Fig . 4) established that HS alone had no effect on MT mRNA levels, and HS pretreatment did not prevent LPS-mediated induction of this mRNA (Fig . 4) .
MT
HS
LPS
HS+LPS
Fig . 4 . Effect of HS on LPS-mediated induction of MT-1 mRNA levels in the liver . Experimental conditions were exactly as described in Fig . 3 . Livers were collected 4 h after LPS injection, and total RNA was extracted, fractionated by formaldehyde-agarose gel electrophoresis and analysed by Northern blot hybridization using an MT-I cRNA probe .
HS effects on hepatic cytokine and acute-phase gene expression
43
Discussion and conclusions This study was designed to examine several potential mechanisms by which HS protects against LPS-induced lethality in the mouse . The results establish that pretreatment with HS 5 h prior to LPS challenge does not influence constitutive or LPS-induced hepatic expression of cytokine genes (TNF-a and IL-1 fl) as measured by changes in mRNA levels . These cytokines have been associated with LPS-induced lethality . It was further noted that constitutive and LPS-induced expression of the mouse MT-I gene was unaffected by HS p retreatment . M T represents an example of the acute phase reactants that are regulated by cytokines and are thought to play a protective role in response to LPS . An explanation for HS protection against LPS that now appears unlikely based on the present data is the down-regulation of TNF-a or IL-1 gene expression, as defined by rapid induction of cytokine mRNA levels within the LPS-stimulated Kupffer cell . In related studies, it was also noted that LPS-induced cytokine gene expression in peritoneal macrophages is not attenuated by HS pretreatment (data not shown) . These findings are particularly noteworthy inasmuch as TNF-a has been shown to be an essential mediator of LPS lethality both in normal mice and in mice sensitized with DGaIN . 19.49 The data establish that HS does not antagonize the LPS-induced activation of macrophage gene expression, but cannot exclude the possibility that HS exerts some effects on the synthesis and/or release of cytokines from the macrophage . An argument against the possibility of a global inhibitory effect of HS on cytokine release from the macrophage is the finding that the induction of MT mRNA in hepatocytes is not prevented by HS pretreatment . Although several exogenously administered recombinant cytokines [TNF-a, IL-1 (a and J3) and IL-6] can induce mouse MT gene expression,32 IL-6 has been implicated as playing a pivotal role in hepatocyte acutephase MT gene expression . 32 • a 3 However, the possibility that HS may exert some effect on the synthesis and/or release of TNF-a from the macrophage cannot be excluded . Significant post-transcriptional regulation of TNF-a gene expression has been documented . 50 '51 However, Urbaschek et a/. 36 have shown that isoniazid does not influence the synthesis or release of TNF-a by human monocytes in vitro, but effects of HS were not reported, nor were the in vivo effects of pretreatment with isoniazid on subsequent LPS-induced TNF-a levels . Results presented herein also do not exclude the possibility that HS attenuates other aspects of the LPS-induced activation of lymphoid cells, such as the production of toxic oxygen radicals . It has been suggested that HS may protect against LPS-induced lethality via de novo hepatocyte protein synthesis . 34 .35 For example, pre-induction of the hepatic acute-phase proteins could be envisioned to provide protection from subsequent exposure to LPS . 28 This suggestion for a mechanism of HS action was based, in part, on the ability of HS pretreatment to essentially reverse the LPS-sensitizing effect of o-GaIN, a general inhibitor of hepatocyte RNA synthesis . Moreover, the presence of D-GaIN at the start of the HS pretreatment period completely abrogates HS protection against LPS . 34 A recent report suggests that induction of the acute-phase response by carbon tetrachloride protects mice from D-GaIN sensitization to LPS and TNF-a . 52 While our present findings do not support a generalized effect of HS on hepatocyte acute-phase gene expression, since no effect on MT mRNA levels was noted, the data do not exclude the possibility that HS induces other hepatocyte mRNAs and proteins specifically . Such a possibility might still conceivably contribute to HS protection against LPS lethality in normal mice and in mice sensitized with D-GaIN . The present study also provides confirmatory evidence of D-GaIN inhibition of hepatocyte RNA synthesis, but not inhibition of RNA synthesis in the neighboring
44
S . K . De et al.
Kupffer cell . In that regard, we have shown that, in LPS-challenged mice, D-GaIN abrogates MT mRNA synthesis, but does not affect LPS-induced expression of the TNF-a, IL-1a and IL-1ll genes . Moreover, we have demonstrated that, in the liver, MT mRNA is localized to the hepatocyte, whereas cytokine mRNA (IL-1 S) is localized to the Kupffer cell . In conclusion, it is proposed that HS protection against LPS lethality occurs subsequent to LPS-induced activation of macrophage gene transcription, and that it does not result from a generalized interference with cytokine actions on the liver or from a generalized induction of hepatocyte protein synthesis .
Materials and methods Chemicals. LPS was isolated from E. co/i 011 : B4 by a modification of the phenol-water method as described by Morrison and Leive53 and was provided by Dr David C . Morrison (University of Kansas Medical Center, Kansas City, KS) . LPS was dissolved by sonication in phosphate-buffered saline (PBS) . HS (Sigma Chemical Co ., St Louis, MO) was prepared fresh before use in pyrogen-free normal saline (8 mg HS/ml) and the solution was brought to neutral pH using 10 N NaOH . D-GaIN (Sigma Chemical Co .) was prepared fresh by dissolving in PBS at a concentration of 45 mg/ml . PBS was prepared fresh with pyrogen-free normal saline . Animals. CF, female mice (6-8 weeks ; 20-25 g) were obtained from Charles River Breeding Laboratories, Inc . (Wilmington, MA) and were maintained in the Animal Resources Facility with food and water ad libitum and 12 h light and dark cycles . In experiments involving HS, mice were taken off feed at 2200 h and injected intraperitoneally (i .p .) with 2 mg HS in 250 pl (80 mg HS per kg body weight), or with vehicle alone (pyrogen free saline) at 0800 h the next morning . LPS was injected at a dosage of 100 to 200 pg/mouse, as noted, 5 h after HS injection (1300 h) . In the experiments involving D-GaIN, mice were injected i .p. with 18 mg D-GaIN along with LPS (0 .1 pg) in 400 pl PBS, according to the procedure of Galanos et al. 12 These conditions of HS pretreatment have previously been shown to protect normal and D-GaINsensitized mice against the lethal effects of an acute exposure to LPS . 34.35 Liver samples were removed at the indicated times after injection, quick-frozen in liquid freon, and stored at -70°C until processed for RNA extraction . In each experiment for each time point shown, approximately equal amounts (0 .3 g) of the livers from three mice were pooled and flash frozen for extraction of RNA . Isolation of total hepatic RNA . Hepatic RNA was extracted using the sodium dodecylsulfate (SDS)-phenol-chloroform procedure described in detail by Andrews et al." Frozen samples were homogenized in 0 .5% SDS, 25 mm EDTA, 75 mm NaCl, pH 8 .0 (SDS-buffer), extracted with phenol, and then with phenol/chloroform :isoamyl alcohol (24 :1, v/v) . RNA was precipitated from the aqueous phase with 3 M ammonium acetate, and reprecipitated twice more to remove DNA . The RNA pellet was redissolved in water and precipitated with ethanol . Hybridization probes. Mouse cDNA clones for MT-I (R . D . Palmiter, University of Washington, Seattle, WA), IL-1/i (The DuPont Merck Pharmaceutical Co ., Glenolden, PA) and TNF-a (Genentech Inc ., San Francisco, CA) were inserted into the Sp6 or pGEM vectors (Promega Biotech, Madison, WI) and used as templates for the synthesis of 32 P- or 35S-labeled cRNA probes as described by Melton et al ." Both sense strand and antisense strand RNA probes were synthesized for use in 'the in situ hybridization experiments . Probes had specific activities of about 2 x 10 9 dpm/pg . Northern blot hybridization . Northern blot hybridization was performed as described previously . 32,3s,4o Briefly, RNA was denatured 5 min at 60°C in a solution of 1 x 3- (4-morpholino) propane sulfonic acid (MOPS) buffer (20 mm MOPS, 5 mm sodium acetate, 1 mm EDTA, pH 7 .0) containing 50% formamide and 2 .2 M formaldehyde. Denatured RNA (2 Yg for MT-1 blots and 6 pg for the cytokine blots) was fractionated by formaldehyde-1 .5% agarose gel electrophoresis ." Gels were soaked for 40 min in 10 mm sodium phosphate, pH 7 .0, and transferred to nitrocellulose in the presence of 20xSSC (3 M NaCl and 0 .3 M sodium citrate, pH 7 .4) . Following transfer, the filters were baked in a vacuum oven at 75°C for 5 h . Northern
HS effects on hepatic cytokine and acute-phase gene expression
45
blots were pre-hybridized, hybridized, and washed as described in detail by Andrews et al . 14 In all experiments, duplicate gels were stained with acridine orange to ensure integrity of the RNA sample and to confirm that equal amounts of RNA had been loaded onto each lane . In situ hybridization. The details of these techniques have been described previously . 32,39,40,57 Mice were anesthetized with Avertin, and perfusion-fixed livers were removed, cut into small pieces and further fixed with 4% paraformaldehyde in PBS for 2 h . Livers were embedded in paraffin blocks and serially sectioned at 7 µm . Sections from four separate control and four separate treated livers were placed on the same poly-l-lysine coated slides . Multiple slides were hybridized with 35 S-labeled MT-I and IL-1/3 cRNA (antisense) probes . Control slides were hybridized with 35S-labeled MT-I and IL-1 /3 sense strand RNA probes . After hybridization, slides were washed and treated with RNase A to reduce the background and ensure specificity of the remaining hybrids as described ,32,39,40 Hybrids were detected by autoradiography using Kodak NTB-2 liquid emulsion (Eastman Kodak, Rochester, NY) . Autoradiographic exposure times for MT-I and IL-1$ hybrids were 9 and 63 days, respectively . Slides were post-stained lightly in hematoxylin .
This study was supported, in part, by grants (ES04725 to GKA and A130500 to RS) from the NIH . SKD is a Postdoctoral Fellow supported by the Wesley Foundation Scholars Program .
References 1 . Lefer AM . Significance of lipid mediators in shock states . Circ Shock 1989; 27 : 3-12 . 2 . West RL, Elovitz MJ, Hardaway RM . Blood coagulation factor activity in experimental endotoxemia . Ann Surg 1966 ; 163 : 567-72 . 3 . Snedegard G, Lui L, Hugli TE . Endotoxin induced shock in the rat . A role of C5a . Am J Pathol 1989 ; 135:489-97 . 4 . Flohe L, Giertz H . Endotoxins, arachidonic acid, and superoxide formation . In: Urbaschek B, ed . Perspectives on bacterial pathogenesis and host defense . Chicago : University of Chicago Press, 1988 ; 123-31 . 5 . Doebber TW, Wu MS, Robbins JC, Choy BM et al. Platelet-activating factor (PAF) involvement in endotoxin-induced hypotension . Studies with PAF receptor antagonist kadsurenone . Biochem Biophys Res Commun 1985; 127 : 799-808 . 6 . Fritz H, Jochum M, Duswald KH et al. Granulocyte proteinases as mediators of unspecific proteolysis in inflammation : a review . In : Tschesche H, ed . Proteinases in inflammation and tumor invasion . Berlin : Walter de Gruyter, 1986 : 1-23 . 7 . Carlos TM, Harlan JM . Membrane proteins involved in phagocyte adherence to endothelium . Immunol Rev 1990 ; 114: 5-28 . 8 . Kreger BE, Craven DE, Carling PC et a/. Gram-negative bacteremia . III . Reassessment of etiology, epidemiology and ecology in 612 patients. Am J Med 1980 ; 68 : 332-7 . 9 . Wolff SM, Bennett JV . Gram-negative rod bacteremia (editorial) . N Engl J Med 1974; 291 : 733-4 . 10 . Morrison DC, Ryan JL. Endotoxins and disease mechanisms . Ann Rev Med 1987 ; 38 : 417-32 . 11 . Beutler B, Cerami A . Cachectin : more than a tumor necrosis factor . N Engl J Med 1987 ; 316 : 379-85 . 12 . Ziegler EJ, Fisher CL Jr, Sprung CL et al. Treatment of gram-negative bacteremia and septic shock with HA-1 A human monoclonal antibody against endotoxin . N Engl J Med 1991 ; 324 : 429-36 . 13 . Morrison DC, Silverstein R, Bright SW et al . Monoclonal antibody to mouse lipopolysaccharide receptor protects mice against the lethal effects of endotoxin . J Infect Dis 1990 ; 162 : 1063-8 . 14 . Arai K-i, Lee F, Miyajima A et a/. Cytokines : coordinators of immune and inflammatory processes . Ann Rev Biochem 1990; 59 : 783-836 . 15 . Akira S, Hirano T, Taga T et al. Biological multifunctional cytokines : IL-6 and related molecules (IL-1 and TNF) . FASEB J 1990 ; 4 : 2860-7 . 16 . Whicher JT, Evans SW. Cytokines and disease . Clin Chem 1990; 36 :1269-81 . 17 . Bauss F, Droge W, Mannel DN . Tumor necrosis factor mediates endotoxic effects in mice . Infect Immun 1987 ; 55 :1622-5 . 18 . Tracey KJ, Beutler B, Lowry SF et al. Shock and tissue injury induced by recombinant human cachectin . Science 1986 ; 234 : 470-4 . 19 . Beutler B, Milsark IW, Cerami A . Passive immunization against cachectin/tumor necrosis factor protects mice from lethal effects of endotoxin . Science 1985; 229 : 869-971 . 20 . Tracey KJ, Fong Y, Hesse DG et al. Anti-cachectin/TNF monoclonal antibodies prevent septic shock during lethal bacteremia . Nature 1987 ; 330 : 662-4 . 21 . Sherry BA, Gelin J, Fong Y et al. Anticachectin/tumor necrosis factor-a antibodies attenuate development of cachexia in tumor models . FASEB J 1989; 3 : 1956-62 .
46
S . K . De et al.
22 . Hannum CH, Wilcox CJ, Arend WP et al. Interleukin-1 receptor antagonist activity of a human interleukin-1 inhibitor . Nature 1990 ; 343 : 336-40. 23 . Eisenberg SP, Evans RJ, Arend WP et al. Primary structure and functional expression from complementary DNA of a human interleukin-1 receptor antagonist . Nature 1990 ; 343 : 341-6 . 24 . Henricson BE, Neta R, Vogel SN . An interleukin-1 receptor antagonist blocks lipopolysaccharideinduced colony-stimulating factor production and early endotoxin tolerance . Infect Immun 1991 ; 59 : 1188-91 . 25 . Starnes FH, Pearce MK, Tewari A et al. Anti IL-6 monoclonal antibodies protect against lethal Escherichia coli infection and lethal tumor necrosis factor-a challenge in mice . J Immunol 1990; 145 : 4185-91 . 26 . Zuckerman SH, Evans GF, Butler LD . Endotoxin tolerance : independent regulation of interleukin-1 and tumor necrosis factor expression . Infect Immun 191 ; 59 : 2774-80 . 27 . Franks AK, Kujawa KI, Yaffe LJ . Experimental elimination of tumor necrosis factor in low-dose endotoxin models has variable effects on survival . Infect Immun 1991 ; 59 : 2609-14. 28 . Koj A . In : Gordon AH, Koj A, eds . The acute phase response to injury and infection . Amsterdam : Elsevier Science Publishers BV. 1985 ; 139-232. 29 . Prowse KR, Baumann H . Interleukin-1 and interleukin-6 stimulate acute-phase protein production in primary mouse hepatocytes. J Leukocyte Biol 1989 ; 45 : 55-61 . 30 . Andus T, Geiger T, Hirano T et al. Regulation of synthesis and secretion of major rat acute-phase proteins by recombinant human interleukin-6 (BSF-2/IL-6) in hepatocyte primary cultures . Eur J Biochem 1988 ; 173 : 287-93 . 31 . Andrews GK. Regulation of metallothionein gene expression . Prog Food Nutr Sci 1990 ; 14 : 193258 . 32 . De SK, McMaster M, Andrews GK . Endotoxin induction of murine metallothionein genes . J Biol Chem 1990; 265 : 15 267-74 . 33 . Schroeder JJ, Cousins RJ . Interleukin-6 regulates metallothionein gene expression and zinc metabolism in hepatocyte monolayer cultures . Proc Natl Acad Sci USA 1990; 87 : 3137-41 . 34 . Silverstein R, Turley BR, Christoffersen et al. Hydrazine sulfate protects D-galactosamine-sensitized mice against endotoxin and tumor necrosis factor/cachectin lethality : evidence of a role for the pituitary . J Exp Med 1991 ; 173 : 357-65 . 35 . Silverstein R, Christoffersen CA, Morrison DC . Modulation of endotoxin lethality in mice by hydrazine sulfate . Infect Immun 1989 ; 57 : 2072-8. 36 . Urbaschek R, Mannel DN, Urbanczik R . Isoniazid protects mice from endotoxin lethality without influencing tumor necrosis factor synthesis and release. Antimicrob Agents Chemother 1991 ; 35 :16668. 37 . Hughes TK, Cadet P, Lamed CS . Modulation of tumor necrosis factor activities by a potential anticachexia compound, hydrazine sulfate . Int J Immunopharmacol 1989; 11 : 501-7 . 38 . Dinarello CA . The biology of interleukin-1 and comparison to tumor necrosis factor . Immunol Lett 1987 ;227-32 . 39 . De SK, McMaster MT, Day SK et al. Cell-specific metallothionein gene expression in mouse decidua and placentae . Development 1989; 107 : 611-21 . 40. De SK, Enders GC, Andrews GK . High levels of metallothionein mRNAs in male germ cells of the adult mouse. Mol Endocrinol 1991 ; 5 : 628-36 . 41 . Decker K, Keppler D . Galactosamine hepatitis : key role of the nucleotide deficiency period in the pathogenesis of cell injury and cell death . Rev Physiol Biochem Pharmacol 1974; 71 : 78-106 . 42 . Galanos C, Freudenberg MA, Reutter W . Galactosamine-induced sensitization to the lethal effects of endotoxin . Proc Natl Acad Sci USA 1979; 76 : 5939-43 . 43 . Freudenberg MA, Keppler D, Galanos C . Requirement for lipopolysaccharide-responsive macrophages in galactosamine-induced sensitization to endotoxin . Infect Immun 1986 ; 891-895 . 44 . Galanos C, Freudenberg MA, Coumbos A et al Induction of lethality and tolerance by endotoxin are mediated by macrophages through tumor necrosis factor . In : Bonavida B, Gifford GE, Kirchner H, Old LJ, eds . International conference on tumor necrosis factor/cachectin and related cytokines . Heidelberg : S . Karger, 1988 ; 114-27 . 45 . Freudenberg MA, Galanos C . Induction of tolerance to lipopolysaccharide (LPS)-D-galactosamine lethality by pretreatment with LPS is mediated by macrophages . Infect Immun 1988 ; 56: 1352-7 . 46 . Sanchez-Cantu L, Rode HN, Christou NV . Endotoxin tolerance is associated with reduced secretion of tumor necrosis factor . Arch Surg 1989; 124 : 1432-6 . 47 . Van der Meer JWM, Helle M, Aarden A . Comparison of the effect of recombinant interleukin-6 and recombinant interleukin-1 on nonspecific resistance to infection . Eur J Immunol 1989 ; 19 : 413-6 . 48 . Fraker DL, Stovroff MC, Merino MJ et al. Tolerance to tumor necrosis factor in rats and the relationship to endotoxin tolerance and toxicity . J Exp Med 1988 ; 168 : 95-105 . 49 . Galanos C, Freudenberg MA . Tumor necrosis factor mediates endotoxin shock : the protective effects of antibodies and cortisone . In : Bonavida B, Grainger G, eds . Tumor necrosis factor : structure, mechanism of action, role in disease and therapy . Basel : S . Karger, 1990 ; 187-93. 50 . Beutler B, Krochin N, Milsark IW et al. Control of cachectin (tumor necrosis factor) synthesis : mechanisms of endotoxin resistance . Science 1986 ; 232 : 977-80 . 51 . Zuckerman SH, Evans GF, Snyder YM et al. Endotoxin-macrophage interaction : post-translation al regulation of tumor necrosis factor expression . J Immunol 1989 ; 143 : 1223-7 .
HS effects on hepatic cytokine and acute-phase gene expression
47
52 . Alcorn JM, Fierer J, Chojkier M . The acute-phase response protects mice from D-galactosamine sensitization to endotoxin and tumor necrosis factor-a . Hepatology 1992; 15 : 122-9 . 53 . Morrison DC, Leive L. Fractions of lipopolysaccharide from Escherichia coli 0111 : B4 prepared by two extraction procedures . J Biol Chem 1975 ; 250: 2911-19 . 54 . Andrews GK, Lehman LD, Huet YM et al. Metallothionein gene regulation in the preimplantation rabbit blastocyst. Development 1987 ; 100: 463-5 . 55 . Melton DA, Krieg PA, Rebagliati MR et al. Efficient in vitro synthesis of biologically active RNA and RNA hybridization probes . from plasmid containing a bacteriophage SP6 promoter . Nucl Acids Res 1984 ;12 :7035-57 . 56 . Lehrach H, Diamond D, Wozney JM et al RNA molecular weight determinations by gel electrophoresis under denaturing conditions, a critical reexamination . Biochemistry 1977 ; 16 : 4743-51 . 57 . Huet-Hudson YM, Chakraborty CC, De SK et al. Estrogen regulates synthesis of EGF in mouse uterine epithelial cells . Mol Endocrinol 1990 ; 4: 510-23 .
