- Email: [email protected]
Enhanced tumor necrosis factor-α production following endotoxin challenge in rats is an early event during magnesium deficiency
Biochimica et Biophysica Acta 1453 (1999) 35^40
Enhanced tumor necrosis factor-K production following endotoxin challenge in rats is an early event during magnesium de¢ciency Corinne Malpuech-Bruge©re, Wojciech Nowacki 1 , Edmond Rock, Elyett Gueux, Andrzej Mazur, Yves Rayssiguier * Centre de Recherches en Nutrition Humaine, Unite¨ Maladies Me¨taboliques et Micronutriments-INRA-Theix, 63122 Saint-Gene©s-Champanelle, France Received 4 June 1998; received in revised form 13 August 1998; accepted 24 August 1998
Abstract Magnesium (Mg) plays an essential role in fundamental cellular reactions and the importance of the immunoinflammatory processes in the pathology of Mg deficiency has been recently reconsidered. The purpose of the present study was to assess the effect of different stages of Mg deficiency on endotoxin response and tumor necrosis factor-K (TNFK) production. Weaning male Wistar rats were pair fed either a Mg-deficient or a control diet. At day 7, lipopolysaccharide (LPS) induced no lethal effects in control rats but resulted in 70% mortality in Mg-deficient rats within 3 h. The vulnerability of Mg-deficient rats to LPS was associated with higher TNFK plasma values. Mg-deficient animals that received magnesium supplementation before endotoxin challenge had significantly increased survival. At day 2, control and Mg-deficient rats were also subjected to endotoxin challenge with or without magnesium pre-treatment. A significant increase in TNFK plasma level was observed in Mg-deficient rats compared to rats fed the control diet. Mg-deficient rats that received magnesium replacement therapy before endotoxin challenge had significantly lower TNFK plasma values than those receiving saline before endotoxin. Thus, the results of this experiment suggest that the activated or primed state of immune cells is an early event occurring in Mg deficiency. ß 1999 Elsevier Science B.V. All rights reserved. Keywords: Magnesium; Endotoxin; Tumor necrosis factor-K; Rat
1. Introduction Magnesium (Mg), the most abundant intracellular divalent cation, plays an essential role in a wide range of fundamental cellular reactions, and many clinical signs, symptoms and disease states are attribAbbreviations: LPS, lipopolysaccharide; TNFK, tumor necrosis factor-K * Corresponding author. Fax: +33 (4) 73624238; E-mail: [email protected] 1 Present address: Veterinary Faculty, Academy of Agriculture, Norwida St 29, 5035 Wroclaw, Poland.
uted to altered Mg homeostasis [1]. In developed countries, the marginal Mg intake may induce a high prevalence of marginal Mg de¢ciency. Mg depletion may also occur because of dysregulations of control mechanisms of Mg metabolism [2]. Mg de¢ciency has been extensively studied in the rat where it is readily produced by dietary depletion. In 1932, Kruse et al. [3] reported peripheral vasodilatation with hyperemia of the ears in rats following Mg de¢ciency. The importance of the immuno-in£ammatory processes in the pathobiology of Mg de¢ciency has been recently reconsidered. The aim of several studies has been to characterize more precisely this
0925-4439 / 99 / $ ^ see front matter ß 1999 Elsevier Science B.V. All rights reserved. PII: S 0 9 2 5 - 4 4 3 9 ( 9 8 ) 0 0 0 8 1 - 7
BBADIS 61785 23-12-98
36
C. Malpuech-Bruge©re et al. / Biochimica et Biophysica Acta 1453 (1999) 35^40
in£ammatory response [4,5] which results in signi¢cant increase in free radical species [6^8] and subsequent tissue injury [9,10]. Moreover, Mg de¢ciency aggravates endotoxin shock [4,11], but the mechanism responsible for the detrimental e¡ect of Mg de¢ciency on immune cells remains unknown. The macrophage is a major e¡ector cell in the host response to endotoxin [12] and tumor necrosis factor-K (TNFK) has been implicated as an important mediator in the development of endotoxin toxicity [13^ 15]. Recent data from our group suggest that Mg de¢ciency results in macrophage activation [4]. Thus, it is possible that TNFK is involved in producing a state of shock following lipopolysaccharide (LPS) challenge. Other studies from our laboratory demonstrate that early modi¢cations in the immune system occur in Mg-de¢cient rats [16]. Therefore, the purpose of the present study was to assess the e¡ects of di¡erent stages of Mg de¢ciency on endotoxin response and TNFK production. 2. Materials and methods 2.1. Animals, diets and treatment Male weaning Wistar rats (I¡a-Credo, L'Arbresle, France) weighing about 60 þ 2 g (mean þ S.E.M.) were randomly divided into Mg-de¢cient and control groups. The institution's guide for the care and use of laboratory animals was used. The rats were housed in wire-bottomed cages in a temperature-controlled room (22³C) with a 12 h dark (20.00^08.00 h) and 12 h light period. They were pair fed with the appropriate diets for 8 days using an automatic feeding apparatus. Distilled water was provided ad libitum. The synthetic diets contained (g/kg): casein 200, sucrose 650, corn oil 50, alphacel 50, DL-methionine 3, choline bitartrate 2, modi¢ed AIN-76 mineral mix 35, AIN-76A vitamin mix 10 (ICN Biomedicals, Orsay, France). MgO was omitted from the AIN-mineral mix in Mg-de¢cient diet. The Mg concentrations of the diets, determined by £ame atomic absorption spectrometric analysis (Perkin Elmer 400, Norwalk, CT), were 29 mg/kg (de¢cient diet) and 943 mg/kg (control diet). Animals were anesthetized with sodium pentobarbital (40 mg/kg body weight, intraperitoneally). Blood was collected by exsanguination via
the abdominal aorta and plasma was obtained by low-speed centrifugation (2000Ug) and stored at 380³C until analysis. LPS (Escherichia coli, 011: B4, Sigma, St Quentin Fallavier, France) was dissolved in sterile saline immediately before administration and rats received either LPS (1 mg/kg intraperitoneally) or normal saline solution. In experiments designed to assess the e¡ects of acute Mg supplementation on endotoxin response, rats received 50 mg/kg MgCl2 intraperitoneally 30 min before the animals were challenged with LPS. Animals were maintained on the experimental diet for either 2 or 7 days before experimentation. Day 7: rats were sacri¢ced for assessment of Mg status (eight animals for each dietary group), or subjected to endotoxin challenge. At day 7 the behavior of the animals was observed 180 min after endotoxin administration to assess the e¡ect of Mg status on mortality (24 animals for each dietary group) or sacri¢ced 90 min after administration of endotoxin or normal saline solution (24 animals for each dietary group). An additional Mg-de¢cient group was used to evaluate the e¤ciency of acute Mg therapy on response of Mg-de¢cient rats to endotoxin challenge (12 animals for each group). Day 2: rats were observed 180 min after endotoxin administration (12 animals for each dietary group) or sacri¢ced 90 min after endotoxin challenge with or without Mg pre-treatment (ten animals for each group) for blood plasma Mg and TNFK determination. 2.2. Analysis Leukocyte number was determined by a cell counter (Cobas, Ho¡mann La Roche). Plasma was determined by £ame atomic absorption spectrometry analysis (Perkin Elmer 420) after dilution in lanthanum chloride solution containing 1 g La/l. TNFK was measured in plasma samples by a cytotoxicity assay using L929 murine ¢broblasts [17]. Brie£y, after L929 cells were cultured in microtiter plates for 20 h, diluted plasma samples or standards were added in triplicate with actinomycin D (1 Wg/ml) (Sigma, St Quentin Fallavier, France). Three serial 2-fold dilutions (from 1/32 to 1/128) were tested for each plasma samples. After 24 h incubation, the cells were stained with 0.5% crystal violet in ethanolH2 O (1:5) and washed. A microtiter plate was used
BBADIS 61785 23-12-98
C. Malpuech-Bruge©re et al. / Biochimica et Biophysica Acta 1453 (1999) 35^40
37
for measurement. TNFK activity was expressed in pg/ml. The assay was calibrated using murine recombinant TNFK (National Institute for Biological Standards and Controls, NIBSC, 88/532, Hertfordshire, UK). 2.3. Statistical methods Results are expressed as means þ S.E.M. The statistical signi¢cance of di¡erences was assessed using Student's t-test, and ANOVA with the StatView 512, statistical software. When signi¢cant di¡erences between groups were identi¢ed (P 6 0.05), comparisons between individuals groups were made using Tukey's test. Survival data after LPS challenge were analyzed using chi-square analysis with the Instat statistical program. 3. Results The ¢rst set of experiments was performed on rats fed the experimental diet for 7 days. Clinical symptoms including hyperemia, slight growth retardation, leukocytosis, and decreased plasma Mg level indicated that the rats fed the Mg-de¢cient diet were indeed Mg-de¢cient (Table 1). LPS at 1 mg/kg induced no lethal e¡ect in control rats but induced the classical signs of shock and resulted in 70% mortality in Mg-de¢cient rats within 3 h (P 6 0.001, M2 ). The TNFK level remained beneath the limit of detection in control and Mg-de¢cient rats before LPS injection, but 90 min after endotoxin administration all animals displayed a rise in plasma TNFK levels. Importantly, higher plasma TNFK levels were found in Mg-de¢cient rats compared to control rats (about Table 1 Body weights, plasma magnesium concentrations and leukocytosis in control and Mg-de¢cient rats, after 7 days on experimental diets
Fig. 1. Mortality after LPS challenge (1 mg/kg) with or without magnesium pre-treatment (50 mg/kg MgCl2 ) in Mg-de¢cient rats (12 per group) after 7 days on experimental diet. Mortality rates di¡er signi¢cantly at P 6 0.001 (chi-square test).
10-fold) (Table 2). Thus, Mg de¢ciency enhances the animal's sensitivity to endotoxin. Mg-de¢cient animals that received Mg supplementation before endotoxin challenge had a signi¢cantly increased survival rate compared to those receiving saline (Fig. 1) (P 6 0.001 M2 ). The possibility that modi¢cation of immune response is an early event in Mg de¢ciency was assessed in rats following 2 days of Mg deprivation. Endotoxin administration did not result in mortality of any rat that received experimental diets for 2 days. After endotoxin challenge, higher plasma TNFK values were found in Mg-de¢cient rats than in control rats. Mg-de¢cient rats that received Mg supplementation before endotoxin challenge had signi¢cantly lower plasma TNFK values than animals that received saline before endotoxin. However, there were no signi¢cant di¡erences in plasma TNFK valTable 2 Plasma TNFK concentrations, before or after LPS challenge, in control and Mg-de¢cient rats, after 7 days on experimental diets Dietary group
Dietary group Control Body weight (g) 96 þ 2 Plasma Mg (mmol/l) 0.81 þ 0.04 White blood cellsU106 /ml 4.61 þ 0.87
Control Mg-de¢cient 91 þ 2 0.07 þ 0.01* 20.04 þ 2.29*
Values are means þ S.E.M., n = 8. *P 6 0.001: signi¢cant di¡erence between control and Mg-de¢cient rats.
Mg-de¢cient
Plasma concentrations (pg/ml) Without LPS With LPS
n.d. 341 þ 137
n.d. 3109 þ 616*
Values are means þ S.E.M., n = 8. *P 6 0.001: signi¢cant di¡erence between control and Mg-de¢cient rats. n.d., not detectable.
BBADIS 61785 23-12-98
38
C. Malpuech-Bruge©re et al. / Biochimica et Biophysica Acta 1453 (1999) 35^40
Table 3 Plasma TNFK concentrations after magnesium or saline injection to control and Mg-de¢cient rats, after 2 days on experimental diets Control Plasma Mg (mmol/l) Plasma TNFK (pg/ml)
Mg-de¢cient
ANOVA
Saline
Mg
Saline
Mg
0.85 þ 0.05a 1665 þ 463a
1.49 þ 0.05b 1821 þ 225a
0.53 þ 0.04c 4400 þ 798b
1.19 þ 0.06d 1712 þ 292a
D, T, DUT D, T, DUT
Values are means þ S.E.M., n = 8. Data were analyzed by two-way ANOVA to determine the e¡ect of diet (D), the e¡ect of Mg treatment (T) and interaction (DUT). Values with di¡erent superscripts are signi¢cantly di¡erent, P 6 0.05.
ues between control animals that received Mg or saline before endotoxin challenge (Table 3). Mg-de¢cient rats had decreased plasma Mg values compared to controls. Furthermore, a signi¢cant increase in plasma Mg values was noted in animals that received Mg compared to baseline values (Table 3). 4. Discussion Mg plays an important role in the maintenance of host regulatory mechanisms in in£ammation. It is well known that dietary Mg de¢ciency in the rat gives rise, after a few days, to a characteristic peripheral vasodilatation. Leukocytosis, enlarged spleen and exhaustive degranulation of mast cells have also been reported [3,18]. More recent studies indicate that Mg de¢ciency produces an elevation in plasma in£ammatory cytokines [5,19,20] and nitric oxide [21]. The increased circulating interleukin-6 in Mg-de¢cient rats also induces the synthesis of acute phase reactants [4]. Macrophages, found in greater number in the peritoneal cavity of Mg-de¢cient rats, appear to be primed for superoxide production following phorbol myristate acetate stimulation [4], i.e. these cells have enhanced response to stimulation. Together, these results indicate that in Mg-de¢cient rats the phagocytic cells are metabolically stimulated. There is increasing evidence of the importance of a cytokine cascade in the induction and control of the in£ammatory reaction. It has been reported that the increased production of free radicals in Mg de¢ciency is caused by elevated concentrations of TNFK, which in turn was increased by early induction of substance P [5]. Increased TNFK concentrations, however, were only detected after several weeks of Mg de¢ciency [19,22]. In the present study,
after 1 week on Mg-de¢cient diet we were unable to detect TNFK in the plasma of Mg-de¢cient rats despite the high sensitivity of the bioassay used. After LPS administration, TNFK is generated in su¤cient amounts to reach detectable levels in the blood of control and Mg-de¢cient rats. The major ¢nding of this study is that the vulnerability of Mg-de¢cient rats to LPS is associated with higher production of TNFK and that this alteration occurs as early as 48 h following the ¢rst Mg depleted meal. The fact that after 2 days on magnesium-de¢cient diet the higher production of TNFK following LPS challenge is not followed by mortality should be underlined, even though after 7 days of Mg deprivation a higher mortality is observed after LPS challenge. The mechanisms underlying septic shock have been the subject of intensive investigations [23]. LPS induces a spectrum of pathophysiological activities, many of which are strikingly similar to those seen in patients with Gram-negative infection. Bacterial LPS reacts with plasma proteins and with speci¢c cellular receptors initiating a response by the host defence system to the invading Gramnegative bacteria. However, in some situations, the response to LPS may become detrimental to the host. An intravenous injection of a su¤cient amount of LPS into an animal produces a severe clinical syndrome known as endotoxin shock [24]. The generalized vasodilatation produces hypotension and the stimulated phagocytic cells release toxic reactive oxygen intermediates and proteolytic enzymes resulting in severe tissue damage. Many of these e¡ects are mediated by endogenous cytokines. Although cytokines are required for the in£ammatory process in response to infection or injury, their overproduction can lead to local and systemic pathology [25]. TNFK is thought to be one of the important mediators in
BBADIS 61785 23-12-98
C. Malpuech-Bruge©re et al. / Biochimica et Biophysica Acta 1453 (1999) 35^40
septic shock [13^15]. And it has been shown that TNFK can induce pathophysiological changes characteristic of septic shock. Furthermore, a treatment with TNFK antibodies has been shown to protect against shocks induced by endotoxins [14]. The present results clearly indicate that Mg-de¢cient rats are hypersensitive to LPS as evidenced by the level of lethality following 7 days of Mg deprivation. The underlying mechanisms of the vulnerability of Mg-de¢cient rats to LPS are unknown. Feeding the animals with the Mg-de¢cient diet resulted in a rapid decline in plasma Mg while the cellular concentration of Mg falls only slightly in most tissues from these animals [26]. Thus, the e¡ect of Mg de¢ciency on the immune response is probably more related to the reduction in the extracellular Mg concentration. In agreement with this ¢nding is the result that high mortality in Mg-de¢cient rats is reduced by parenteral Mg administration before LPS challenge. The alteration of the e¡ects of Mg de¢ciency by Mg supplementation also suggests that the bene¢cial e¡ect of Mg is due to increased plasma concentrations. Based on this concept one potential mechanism for the e¡ect of Mg may be through its partial antagonism to calcium [27]. Previous studies have established a role for calcium in the activation of immune cells [28] and the results raise the possibility that extracellular Mg exerts an inhibitory e¡ect on Ca in£ux or Ca mobilization in immune cells. Several studies have provided convincing evidence for the role of TNFK as a mediator of septic shock. For this reason, therapeutic interventions that decrease the level of cytokines have come to be of considerable interest for controlling the sepsis syndrome [29]. Our data suggest that plasma Mg levels may be an important regulatory element in immune response since supplemental Mg is able to reduce the secretion of TNFK. Hypomagnesemia is a common electrolyte disorder in hospitalized patients [30]. Sepsis and its associated syndromes represent a potentially devastating systemic in£ammatory response estimated to occur in 1% of hospitalized patients. Mortality approaches 60^90% in those who develop septic shock [31]. TNFK is released during the early stages of septic shock. Circulating TNFK levels might correlate with outcome in severe sepsis [32^35]. Moreover, recent re-evaluation of the clinical de¢nition of sepsis has emphasized the importance of the host response
39
to the insult rather than the virulence of an infecting organism. The exact relevance of the current study to the potential bene¢t of Mg supplementation in preventing septic shock is not entirely clear. However, the present experiment indicates that a relatively activated or primed state of LPS-induced TNFK production is an early event occurring within the ¢rst 2 days of Mg deprivation. Acknowledgements We are grateful to Dr. J. Dornand (INSERM U 65, Montpellier) for kindly providing the method and cell lines for cytokine assays.
References [1] M.E. Shils, Magnesium, in: M.E. Shils, J.A. Olson, M. Shike (Eds.), Modern Nutrition in Health and Disease, vol. 1, Lea and Febiger, Malvern, 1994, ch. 8, pp. 164^184. [2] J. Durlach, in: J. Durlach (Ed.), Magnesium in Clinical Practice, John Libbey and Co., London, 1988. [3] H.D. Kruse, E.R. Orent, E.V. McCollum, Studies on magnesium de¢ciency in animals. I. Symptomatology resulting from magnesium deprivation, J. Biol. Chem. 96 (1932) 519^539. [4] Y. Rayssiguier, C. Malpuech, W. Nowacki, E. Rock, E. Gueux, A. Mazur, In£ammatory response in magnesium de¢ciency, in: R. Smetana (Ed.), Advances in Magnesium Research. 1. Magnesium in Cardiology, John Libbey and Co., London, 1995, pp. 415^416. [5] W.B. Weglicki, T.M. Phillips, Pathobiology of magnesium de¢ciency: a cytokine/neurogenic in£ammation hypothesis, Am. J. Physiol. 263 (1992) R734^R737. [6] T. Gu«nther, J. Vormann, V. Ho«llriegl, Lipid peroxidation and magnesium de¢ciency, in: M.J. Halpern (Ed.), Current Research in Magnesium, John Libbey and Co., London, 1996, pp. 189^193. [7] Y. Rayssiguier, E. Gueux, L. Bussie©re, J. Durlach, A. Mazur, Dietary magnesium a¡ects susceptibility of lipoproteins and tissues to peroxidation in rats, J. Am. Coll. Nutr. 12 (1993) 133^137. [8] E. Rock, C. Astier, C. Lab, X. Vignon, E. Gueux, C. Motta, Y. Rayssiguier, Dietary magnesium de¢ciency in rats enhances free radical production in skeletal muscle, J. Nutr. 125 (1995) 1205^1210. [9] C. Astier, E. Rock, C. Lab, E. Gueux, A. Mazur, Y. Rayssiguier, Functional alterations in sarcoplasmic reticulum membranes of magnesium-de¢cient rat skeletal muscle as consequences of free radical mediated process, Free Radic. Biol. Med. 20 (1996) 667^674.
BBADIS 61785 23-12-98
40
C. Malpuech-Bruge©re et al. / Biochimica et Biophysica Acta 1453 (1999) 35^40
[10] A.M. Freedman, A.H. Atrakchi, M.M. Cassidy, R.A. Muesing, W.B. Weglicki, Magnesium de¢ciency-induced cardiomyopathy: protection by vitamin E, Biochem. Biophys. Res. Commun. 170 (1990) 1102^1106. [11] M. Salem, N. Kasinski, R. Munoz, B. Chernow, Progressive magnesium de¢ciency increases mortality from endotoxin challenge: protective e¡ects of acute magnesium replacement therapy, Crit. Care Med. 23 (1995) 108^118. [12] M.A. Freudenberg, D. Keppler, C. Galanos, Requirement for lipopolysaccharide-responsive macrophages in galactosamine-induced sensibilization to endotoxin, Infect. Immun. 51 (1986) 891^895. [13] B. Beutler, A. Cerami, Cachectin/tumor necrosis factor: an endogenous mediator of shock and in£ammation, Annu. Rev. Biochem. 57 (1988) 505^516. [14] B. Beutler, I.W. Milsark, A. Cerami, Passive immunization against cachectin/tumor necrosis factor (TNF) protects mice from the lethal e¡ect of endotoxin, Science 229 (1985) 869^ 871. [15] M.A. Freudenberg, C. Galanos, Tumor necrosis factor alpha mediates lethal activity of killed Gram-negative and Grampositive bacteria in D-galactosamine-treated mice, Infect. Immun. 59 (1991) 2110^2115. [16] C. Malpuech-Bruge©re, J. Kuryszko, W. Nowacki, E. Rock, Y. Rayssiguier, A. Mazur, Early morphological and immunological alterations in the spleen during magnesium de¢ciency in the rat, Magnes. Res. (1998) in press. [17] L. Givalois, J. Dornand, M. Mekaouche, M.D. Solier, A.F. Bristow, G. Ixart, P. Siaud, I. Assenmacher, G. Barbanel, Temporal cascade of plasma level surges in ACTH, corticosterone, and cytokines in endotoxin-challenged rats, Am. J. Physiol. 276 (1994) R164^R170. [18] P. Bois, A. Gascon, A. Beaulnes, Histamine-liberating e¡ect of magnesium de¢ciency in the rat, Nature 197 (1963) 501^ 502. [19] T. Gu«nther, K. Schu«mann, J. Vormann, Tumor necrosis factor-K, prostanoids and immunoglobulins in magnesium de¢ciency, Magnes. Bull. 17 (1995) 109^114. [20] J.A.M. Maier, C. Malpuech-Bruge©re, E. Rock, Y. Rayssiguier, A. Mazur, Serum from magnesium-de¢cient rats a¡ects vascular endothelial cells in culture: role of hyperlipemia and in£ammation, J. Nutr. Biochem. 9 (1998) 17^22. [21] E. Rock, C. Astier, C. Lab, C. Malpuech, W. Nowacki, E. Gueux, A. Mazur, Y. Rayssiguier, Magnesium de¢ciency in rats induces a rise in plasma nitric oxide level, Magnes. Res. 8 (1995) 237^242. [22] W.B. Weglicki, I.T. Mak, J.H. Kramer, B.F. Dickens, M.M. Cassidy, R.E. Sta¡ord, T.M. Phillips, Role of free radicals
[23] [24]
[25]
[26] [27] [28]
[29] [30]
[31]
[32]
[33]
[34]
[35]
and substance P in magnesium de¢ciency, Cardiovasc. Res. 31 (1996) 677^682. J.S. Downey, J. Han, Cellular activation mechanisms in septic shock, Front. Biosci. 3 (1998) D468^D476. S. Bahrami, H. Redl, G. Schlag, Models of endotoxemia in rodents, in: G. Schlag, H. Redl (Eds.), Pathophysiology of Shock, Sepsis, and Organ Failure, Springer-Verlag, Berlin, 1993, pp. 1019^1030. H. Redl, G. Schlag, S. Bahrami, J. Davies, A. Waage, M. Ceska, W.A. Buurman, G. Adolf, The cytokine network in trauma and sepsis. I. TNF and IL-8, in: G. Schlag, H. Redl (Eds.), Pathophysiology of Shock, Sepsis, and Organ Failure, Springer-Verlag, Berlin, 1993, pp. 468^490. L. Martindale, F.W. Heaton, Magnesium de¢ciency in the adult rat, Biochem. J. 92 (1994) 119^126. L.T. Iseri, J.H. French, Magnesium: nature's physiologic calcium blocker, Am. Heart J. 108 (1984) 188^193. R.S. Hotchkiss, W.N. Bowling, I.E. Karl, D.F. Osborne, M.W. Flye, Calcium antagonists inhibit oxidative burst and nitrite formation in lipopolysaccharide-stimulated rat peritoneal macrophages, Shock 8 (1997) 170^178. G. Schlag, H. Redl (Eds.), Pathophysiology of Shock, Sepsis, and Organ Failure, Springer-Verlag, Berlin, 1993. R.K. Rude, Magnesium homeostasis, in: M. Bilezikian (Ed.), Principles of Bone Biology, Academic Press, London, 1996, pp. 277^293. R.C. Bone, C.J. Fisher, T.P. Clemmer, C.J. Slotman, C.A. Metz, R.A. Balk, Sepsis syndrome: a valid clinical entity, Crit. Care Med. 171 (1989) 389^393. T. Calandra, J. Baumgartner G.E. Grau, M. Wu, P. Lambert, J. Schellekens, J. Verhoef, M.P. Glauser, Swiss-Dutch J5. Immunoglobulin Study Group, Prognostic values of tumor necrosis factor/cachectin, interleukin-1, interferon-K, and interferon-Q in the serum of patients with septic shock, J. Infect. Dis. 161 (1990) 982^987. J.M. Debets, H.R. Kampmeijer, P.M.H. van der Linden, W.A. Buurman, C.J. van der Linden, Plasma tumor necrosis factor and mortality in critically ill septic patients, Crit. Care Med. 17 (1989) 489^494. A. Waage, A. Halstensen, T. Espevik, An association between tumor necrosis factor and serum and fatal outcome in patients with meningococcal disease, Lancet 14 (1987) 355^357. A. Waage, P. Brandtzaeg, A. Halstelsen, P. Kierulf, T. Espevik, The complex pattern of cytokines in serum from patients with meningococcal septic shock, J. Exp. Med. 169 (1989) 333^338.
BBADIS 61785 23-12-98
Enhanced tumor necrosis factor-K production following endotoxin challenge in rats is an early event during magnesium de¢ciency Corinne Malpuech-Bruge©re, Wojciech Nowacki 1 , Edmond Rock, Elyett Gueux, Andrzej Mazur, Yves Rayssiguier * Centre de Recherches en Nutrition Humaine, Unite¨ Maladies Me¨taboliques et Micronutriments-INRA-Theix, 63122 Saint-Gene©s-Champanelle, France Received 4 June 1998; received in revised form 13 August 1998; accepted 24 August 1998
Abstract Magnesium (Mg) plays an essential role in fundamental cellular reactions and the importance of the immunoinflammatory processes in the pathology of Mg deficiency has been recently reconsidered. The purpose of the present study was to assess the effect of different stages of Mg deficiency on endotoxin response and tumor necrosis factor-K (TNFK) production. Weaning male Wistar rats were pair fed either a Mg-deficient or a control diet. At day 7, lipopolysaccharide (LPS) induced no lethal effects in control rats but resulted in 70% mortality in Mg-deficient rats within 3 h. The vulnerability of Mg-deficient rats to LPS was associated with higher TNFK plasma values. Mg-deficient animals that received magnesium supplementation before endotoxin challenge had significantly increased survival. At day 2, control and Mg-deficient rats were also subjected to endotoxin challenge with or without magnesium pre-treatment. A significant increase in TNFK plasma level was observed in Mg-deficient rats compared to rats fed the control diet. Mg-deficient rats that received magnesium replacement therapy before endotoxin challenge had significantly lower TNFK plasma values than those receiving saline before endotoxin. Thus, the results of this experiment suggest that the activated or primed state of immune cells is an early event occurring in Mg deficiency. ß 1999 Elsevier Science B.V. All rights reserved. Keywords: Magnesium; Endotoxin; Tumor necrosis factor-K; Rat
1. Introduction Magnesium (Mg), the most abundant intracellular divalent cation, plays an essential role in a wide range of fundamental cellular reactions, and many clinical signs, symptoms and disease states are attribAbbreviations: LPS, lipopolysaccharide; TNFK, tumor necrosis factor-K * Corresponding author. Fax: +33 (4) 73624238; E-mail: [email protected] 1 Present address: Veterinary Faculty, Academy of Agriculture, Norwida St 29, 5035 Wroclaw, Poland.
uted to altered Mg homeostasis [1]. In developed countries, the marginal Mg intake may induce a high prevalence of marginal Mg de¢ciency. Mg depletion may also occur because of dysregulations of control mechanisms of Mg metabolism [2]. Mg de¢ciency has been extensively studied in the rat where it is readily produced by dietary depletion. In 1932, Kruse et al. [3] reported peripheral vasodilatation with hyperemia of the ears in rats following Mg de¢ciency. The importance of the immuno-in£ammatory processes in the pathobiology of Mg de¢ciency has been recently reconsidered. The aim of several studies has been to characterize more precisely this
0925-4439 / 99 / $ ^ see front matter ß 1999 Elsevier Science B.V. All rights reserved. PII: S 0 9 2 5 - 4 4 3 9 ( 9 8 ) 0 0 0 8 1 - 7
BBADIS 61785 23-12-98
36
C. Malpuech-Bruge©re et al. / Biochimica et Biophysica Acta 1453 (1999) 35^40
in£ammatory response [4,5] which results in signi¢cant increase in free radical species [6^8] and subsequent tissue injury [9,10]. Moreover, Mg de¢ciency aggravates endotoxin shock [4,11], but the mechanism responsible for the detrimental e¡ect of Mg de¢ciency on immune cells remains unknown. The macrophage is a major e¡ector cell in the host response to endotoxin [12] and tumor necrosis factor-K (TNFK) has been implicated as an important mediator in the development of endotoxin toxicity [13^ 15]. Recent data from our group suggest that Mg de¢ciency results in macrophage activation [4]. Thus, it is possible that TNFK is involved in producing a state of shock following lipopolysaccharide (LPS) challenge. Other studies from our laboratory demonstrate that early modi¢cations in the immune system occur in Mg-de¢cient rats [16]. Therefore, the purpose of the present study was to assess the e¡ects of di¡erent stages of Mg de¢ciency on endotoxin response and TNFK production. 2. Materials and methods 2.1. Animals, diets and treatment Male weaning Wistar rats (I¡a-Credo, L'Arbresle, France) weighing about 60 þ 2 g (mean þ S.E.M.) were randomly divided into Mg-de¢cient and control groups. The institution's guide for the care and use of laboratory animals was used. The rats were housed in wire-bottomed cages in a temperature-controlled room (22³C) with a 12 h dark (20.00^08.00 h) and 12 h light period. They were pair fed with the appropriate diets for 8 days using an automatic feeding apparatus. Distilled water was provided ad libitum. The synthetic diets contained (g/kg): casein 200, sucrose 650, corn oil 50, alphacel 50, DL-methionine 3, choline bitartrate 2, modi¢ed AIN-76 mineral mix 35, AIN-76A vitamin mix 10 (ICN Biomedicals, Orsay, France). MgO was omitted from the AIN-mineral mix in Mg-de¢cient diet. The Mg concentrations of the diets, determined by £ame atomic absorption spectrometric analysis (Perkin Elmer 400, Norwalk, CT), were 29 mg/kg (de¢cient diet) and 943 mg/kg (control diet). Animals were anesthetized with sodium pentobarbital (40 mg/kg body weight, intraperitoneally). Blood was collected by exsanguination via
the abdominal aorta and plasma was obtained by low-speed centrifugation (2000Ug) and stored at 380³C until analysis. LPS (Escherichia coli, 011: B4, Sigma, St Quentin Fallavier, France) was dissolved in sterile saline immediately before administration and rats received either LPS (1 mg/kg intraperitoneally) or normal saline solution. In experiments designed to assess the e¡ects of acute Mg supplementation on endotoxin response, rats received 50 mg/kg MgCl2 intraperitoneally 30 min before the animals were challenged with LPS. Animals were maintained on the experimental diet for either 2 or 7 days before experimentation. Day 7: rats were sacri¢ced for assessment of Mg status (eight animals for each dietary group), or subjected to endotoxin challenge. At day 7 the behavior of the animals was observed 180 min after endotoxin administration to assess the e¡ect of Mg status on mortality (24 animals for each dietary group) or sacri¢ced 90 min after administration of endotoxin or normal saline solution (24 animals for each dietary group). An additional Mg-de¢cient group was used to evaluate the e¤ciency of acute Mg therapy on response of Mg-de¢cient rats to endotoxin challenge (12 animals for each group). Day 2: rats were observed 180 min after endotoxin administration (12 animals for each dietary group) or sacri¢ced 90 min after endotoxin challenge with or without Mg pre-treatment (ten animals for each group) for blood plasma Mg and TNFK determination. 2.2. Analysis Leukocyte number was determined by a cell counter (Cobas, Ho¡mann La Roche). Plasma was determined by £ame atomic absorption spectrometry analysis (Perkin Elmer 420) after dilution in lanthanum chloride solution containing 1 g La/l. TNFK was measured in plasma samples by a cytotoxicity assay using L929 murine ¢broblasts [17]. Brie£y, after L929 cells were cultured in microtiter plates for 20 h, diluted plasma samples or standards were added in triplicate with actinomycin D (1 Wg/ml) (Sigma, St Quentin Fallavier, France). Three serial 2-fold dilutions (from 1/32 to 1/128) were tested for each plasma samples. After 24 h incubation, the cells were stained with 0.5% crystal violet in ethanolH2 O (1:5) and washed. A microtiter plate was used
BBADIS 61785 23-12-98
C. Malpuech-Bruge©re et al. / Biochimica et Biophysica Acta 1453 (1999) 35^40
37
for measurement. TNFK activity was expressed in pg/ml. The assay was calibrated using murine recombinant TNFK (National Institute for Biological Standards and Controls, NIBSC, 88/532, Hertfordshire, UK). 2.3. Statistical methods Results are expressed as means þ S.E.M. The statistical signi¢cance of di¡erences was assessed using Student's t-test, and ANOVA with the StatView 512, statistical software. When signi¢cant di¡erences between groups were identi¢ed (P 6 0.05), comparisons between individuals groups were made using Tukey's test. Survival data after LPS challenge were analyzed using chi-square analysis with the Instat statistical program. 3. Results The ¢rst set of experiments was performed on rats fed the experimental diet for 7 days. Clinical symptoms including hyperemia, slight growth retardation, leukocytosis, and decreased plasma Mg level indicated that the rats fed the Mg-de¢cient diet were indeed Mg-de¢cient (Table 1). LPS at 1 mg/kg induced no lethal e¡ect in control rats but induced the classical signs of shock and resulted in 70% mortality in Mg-de¢cient rats within 3 h (P 6 0.001, M2 ). The TNFK level remained beneath the limit of detection in control and Mg-de¢cient rats before LPS injection, but 90 min after endotoxin administration all animals displayed a rise in plasma TNFK levels. Importantly, higher plasma TNFK levels were found in Mg-de¢cient rats compared to control rats (about Table 1 Body weights, plasma magnesium concentrations and leukocytosis in control and Mg-de¢cient rats, after 7 days on experimental diets
Fig. 1. Mortality after LPS challenge (1 mg/kg) with or without magnesium pre-treatment (50 mg/kg MgCl2 ) in Mg-de¢cient rats (12 per group) after 7 days on experimental diet. Mortality rates di¡er signi¢cantly at P 6 0.001 (chi-square test).
10-fold) (Table 2). Thus, Mg de¢ciency enhances the animal's sensitivity to endotoxin. Mg-de¢cient animals that received Mg supplementation before endotoxin challenge had a signi¢cantly increased survival rate compared to those receiving saline (Fig. 1) (P 6 0.001 M2 ). The possibility that modi¢cation of immune response is an early event in Mg de¢ciency was assessed in rats following 2 days of Mg deprivation. Endotoxin administration did not result in mortality of any rat that received experimental diets for 2 days. After endotoxin challenge, higher plasma TNFK values were found in Mg-de¢cient rats than in control rats. Mg-de¢cient rats that received Mg supplementation before endotoxin challenge had signi¢cantly lower plasma TNFK values than animals that received saline before endotoxin. However, there were no signi¢cant di¡erences in plasma TNFK valTable 2 Plasma TNFK concentrations, before or after LPS challenge, in control and Mg-de¢cient rats, after 7 days on experimental diets Dietary group
Dietary group Control Body weight (g) 96 þ 2 Plasma Mg (mmol/l) 0.81 þ 0.04 White blood cellsU106 /ml 4.61 þ 0.87
Control Mg-de¢cient 91 þ 2 0.07 þ 0.01* 20.04 þ 2.29*
Values are means þ S.E.M., n = 8. *P 6 0.001: signi¢cant di¡erence between control and Mg-de¢cient rats.
Mg-de¢cient
Plasma concentrations (pg/ml) Without LPS With LPS
n.d. 341 þ 137
n.d. 3109 þ 616*
Values are means þ S.E.M., n = 8. *P 6 0.001: signi¢cant di¡erence between control and Mg-de¢cient rats. n.d., not detectable.
BBADIS 61785 23-12-98
38
C. Malpuech-Bruge©re et al. / Biochimica et Biophysica Acta 1453 (1999) 35^40
Table 3 Plasma TNFK concentrations after magnesium or saline injection to control and Mg-de¢cient rats, after 2 days on experimental diets Control Plasma Mg (mmol/l) Plasma TNFK (pg/ml)
Mg-de¢cient
ANOVA
Saline
Mg
Saline
Mg
0.85 þ 0.05a 1665 þ 463a
1.49 þ 0.05b 1821 þ 225a
0.53 þ 0.04c 4400 þ 798b
1.19 þ 0.06d 1712 þ 292a
D, T, DUT D, T, DUT
Values are means þ S.E.M., n = 8. Data were analyzed by two-way ANOVA to determine the e¡ect of diet (D), the e¡ect of Mg treatment (T) and interaction (DUT). Values with di¡erent superscripts are signi¢cantly di¡erent, P 6 0.05.
ues between control animals that received Mg or saline before endotoxin challenge (Table 3). Mg-de¢cient rats had decreased plasma Mg values compared to controls. Furthermore, a signi¢cant increase in plasma Mg values was noted in animals that received Mg compared to baseline values (Table 3). 4. Discussion Mg plays an important role in the maintenance of host regulatory mechanisms in in£ammation. It is well known that dietary Mg de¢ciency in the rat gives rise, after a few days, to a characteristic peripheral vasodilatation. Leukocytosis, enlarged spleen and exhaustive degranulation of mast cells have also been reported [3,18]. More recent studies indicate that Mg de¢ciency produces an elevation in plasma in£ammatory cytokines [5,19,20] and nitric oxide [21]. The increased circulating interleukin-6 in Mg-de¢cient rats also induces the synthesis of acute phase reactants [4]. Macrophages, found in greater number in the peritoneal cavity of Mg-de¢cient rats, appear to be primed for superoxide production following phorbol myristate acetate stimulation [4], i.e. these cells have enhanced response to stimulation. Together, these results indicate that in Mg-de¢cient rats the phagocytic cells are metabolically stimulated. There is increasing evidence of the importance of a cytokine cascade in the induction and control of the in£ammatory reaction. It has been reported that the increased production of free radicals in Mg de¢ciency is caused by elevated concentrations of TNFK, which in turn was increased by early induction of substance P [5]. Increased TNFK concentrations, however, were only detected after several weeks of Mg de¢ciency [19,22]. In the present study,
after 1 week on Mg-de¢cient diet we were unable to detect TNFK in the plasma of Mg-de¢cient rats despite the high sensitivity of the bioassay used. After LPS administration, TNFK is generated in su¤cient amounts to reach detectable levels in the blood of control and Mg-de¢cient rats. The major ¢nding of this study is that the vulnerability of Mg-de¢cient rats to LPS is associated with higher production of TNFK and that this alteration occurs as early as 48 h following the ¢rst Mg depleted meal. The fact that after 2 days on magnesium-de¢cient diet the higher production of TNFK following LPS challenge is not followed by mortality should be underlined, even though after 7 days of Mg deprivation a higher mortality is observed after LPS challenge. The mechanisms underlying septic shock have been the subject of intensive investigations [23]. LPS induces a spectrum of pathophysiological activities, many of which are strikingly similar to those seen in patients with Gram-negative infection. Bacterial LPS reacts with plasma proteins and with speci¢c cellular receptors initiating a response by the host defence system to the invading Gramnegative bacteria. However, in some situations, the response to LPS may become detrimental to the host. An intravenous injection of a su¤cient amount of LPS into an animal produces a severe clinical syndrome known as endotoxin shock [24]. The generalized vasodilatation produces hypotension and the stimulated phagocytic cells release toxic reactive oxygen intermediates and proteolytic enzymes resulting in severe tissue damage. Many of these e¡ects are mediated by endogenous cytokines. Although cytokines are required for the in£ammatory process in response to infection or injury, their overproduction can lead to local and systemic pathology [25]. TNFK is thought to be one of the important mediators in
BBADIS 61785 23-12-98
C. Malpuech-Bruge©re et al. / Biochimica et Biophysica Acta 1453 (1999) 35^40
septic shock [13^15]. And it has been shown that TNFK can induce pathophysiological changes characteristic of septic shock. Furthermore, a treatment with TNFK antibodies has been shown to protect against shocks induced by endotoxins [14]. The present results clearly indicate that Mg-de¢cient rats are hypersensitive to LPS as evidenced by the level of lethality following 7 days of Mg deprivation. The underlying mechanisms of the vulnerability of Mg-de¢cient rats to LPS are unknown. Feeding the animals with the Mg-de¢cient diet resulted in a rapid decline in plasma Mg while the cellular concentration of Mg falls only slightly in most tissues from these animals [26]. Thus, the e¡ect of Mg de¢ciency on the immune response is probably more related to the reduction in the extracellular Mg concentration. In agreement with this ¢nding is the result that high mortality in Mg-de¢cient rats is reduced by parenteral Mg administration before LPS challenge. The alteration of the e¡ects of Mg de¢ciency by Mg supplementation also suggests that the bene¢cial e¡ect of Mg is due to increased plasma concentrations. Based on this concept one potential mechanism for the e¡ect of Mg may be through its partial antagonism to calcium [27]. Previous studies have established a role for calcium in the activation of immune cells [28] and the results raise the possibility that extracellular Mg exerts an inhibitory e¡ect on Ca in£ux or Ca mobilization in immune cells. Several studies have provided convincing evidence for the role of TNFK as a mediator of septic shock. For this reason, therapeutic interventions that decrease the level of cytokines have come to be of considerable interest for controlling the sepsis syndrome [29]. Our data suggest that plasma Mg levels may be an important regulatory element in immune response since supplemental Mg is able to reduce the secretion of TNFK. Hypomagnesemia is a common electrolyte disorder in hospitalized patients [30]. Sepsis and its associated syndromes represent a potentially devastating systemic in£ammatory response estimated to occur in 1% of hospitalized patients. Mortality approaches 60^90% in those who develop septic shock [31]. TNFK is released during the early stages of septic shock. Circulating TNFK levels might correlate with outcome in severe sepsis [32^35]. Moreover, recent re-evaluation of the clinical de¢nition of sepsis has emphasized the importance of the host response
39
to the insult rather than the virulence of an infecting organism. The exact relevance of the current study to the potential bene¢t of Mg supplementation in preventing septic shock is not entirely clear. However, the present experiment indicates that a relatively activated or primed state of LPS-induced TNFK production is an early event occurring within the ¢rst 2 days of Mg deprivation. Acknowledgements We are grateful to Dr. J. Dornand (INSERM U 65, Montpellier) for kindly providing the method and cell lines for cytokine assays.
References [1] M.E. Shils, Magnesium, in: M.E. Shils, J.A. Olson, M. Shike (Eds.), Modern Nutrition in Health and Disease, vol. 1, Lea and Febiger, Malvern, 1994, ch. 8, pp. 164^184. [2] J. Durlach, in: J. Durlach (Ed.), Magnesium in Clinical Practice, John Libbey and Co., London, 1988. [3] H.D. Kruse, E.R. Orent, E.V. McCollum, Studies on magnesium de¢ciency in animals. I. Symptomatology resulting from magnesium deprivation, J. Biol. Chem. 96 (1932) 519^539. [4] Y. Rayssiguier, C. Malpuech, W. Nowacki, E. Rock, E. Gueux, A. Mazur, In£ammatory response in magnesium de¢ciency, in: R. Smetana (Ed.), Advances in Magnesium Research. 1. Magnesium in Cardiology, John Libbey and Co., London, 1995, pp. 415^416. [5] W.B. Weglicki, T.M. Phillips, Pathobiology of magnesium de¢ciency: a cytokine/neurogenic in£ammation hypothesis, Am. J. Physiol. 263 (1992) R734^R737. [6] T. Gu«nther, J. Vormann, V. Ho«llriegl, Lipid peroxidation and magnesium de¢ciency, in: M.J. Halpern (Ed.), Current Research in Magnesium, John Libbey and Co., London, 1996, pp. 189^193. [7] Y. Rayssiguier, E. Gueux, L. Bussie©re, J. Durlach, A. Mazur, Dietary magnesium a¡ects susceptibility of lipoproteins and tissues to peroxidation in rats, J. Am. Coll. Nutr. 12 (1993) 133^137. [8] E. Rock, C. Astier, C. Lab, X. Vignon, E. Gueux, C. Motta, Y. Rayssiguier, Dietary magnesium de¢ciency in rats enhances free radical production in skeletal muscle, J. Nutr. 125 (1995) 1205^1210. [9] C. Astier, E. Rock, C. Lab, E. Gueux, A. Mazur, Y. Rayssiguier, Functional alterations in sarcoplasmic reticulum membranes of magnesium-de¢cient rat skeletal muscle as consequences of free radical mediated process, Free Radic. Biol. Med. 20 (1996) 667^674.
BBADIS 61785 23-12-98
40
C. Malpuech-Bruge©re et al. / Biochimica et Biophysica Acta 1453 (1999) 35^40
[10] A.M. Freedman, A.H. Atrakchi, M.M. Cassidy, R.A. Muesing, W.B. Weglicki, Magnesium de¢ciency-induced cardiomyopathy: protection by vitamin E, Biochem. Biophys. Res. Commun. 170 (1990) 1102^1106. [11] M. Salem, N. Kasinski, R. Munoz, B. Chernow, Progressive magnesium de¢ciency increases mortality from endotoxin challenge: protective e¡ects of acute magnesium replacement therapy, Crit. Care Med. 23 (1995) 108^118. [12] M.A. Freudenberg, D. Keppler, C. Galanos, Requirement for lipopolysaccharide-responsive macrophages in galactosamine-induced sensibilization to endotoxin, Infect. Immun. 51 (1986) 891^895. [13] B. Beutler, A. Cerami, Cachectin/tumor necrosis factor: an endogenous mediator of shock and in£ammation, Annu. Rev. Biochem. 57 (1988) 505^516. [14] B. Beutler, I.W. Milsark, A. Cerami, Passive immunization against cachectin/tumor necrosis factor (TNF) protects mice from the lethal e¡ect of endotoxin, Science 229 (1985) 869^ 871. [15] M.A. Freudenberg, C. Galanos, Tumor necrosis factor alpha mediates lethal activity of killed Gram-negative and Grampositive bacteria in D-galactosamine-treated mice, Infect. Immun. 59 (1991) 2110^2115. [16] C. Malpuech-Bruge©re, J. Kuryszko, W. Nowacki, E. Rock, Y. Rayssiguier, A. Mazur, Early morphological and immunological alterations in the spleen during magnesium de¢ciency in the rat, Magnes. Res. (1998) in press. [17] L. Givalois, J. Dornand, M. Mekaouche, M.D. Solier, A.F. Bristow, G. Ixart, P. Siaud, I. Assenmacher, G. Barbanel, Temporal cascade of plasma level surges in ACTH, corticosterone, and cytokines in endotoxin-challenged rats, Am. J. Physiol. 276 (1994) R164^R170. [18] P. Bois, A. Gascon, A. Beaulnes, Histamine-liberating e¡ect of magnesium de¢ciency in the rat, Nature 197 (1963) 501^ 502. [19] T. Gu«nther, K. Schu«mann, J. Vormann, Tumor necrosis factor-K, prostanoids and immunoglobulins in magnesium de¢ciency, Magnes. Bull. 17 (1995) 109^114. [20] J.A.M. Maier, C. Malpuech-Bruge©re, E. Rock, Y. Rayssiguier, A. Mazur, Serum from magnesium-de¢cient rats a¡ects vascular endothelial cells in culture: role of hyperlipemia and in£ammation, J. Nutr. Biochem. 9 (1998) 17^22. [21] E. Rock, C. Astier, C. Lab, C. Malpuech, W. Nowacki, E. Gueux, A. Mazur, Y. Rayssiguier, Magnesium de¢ciency in rats induces a rise in plasma nitric oxide level, Magnes. Res. 8 (1995) 237^242. [22] W.B. Weglicki, I.T. Mak, J.H. Kramer, B.F. Dickens, M.M. Cassidy, R.E. Sta¡ord, T.M. Phillips, Role of free radicals
[23] [24]
[25]
[26] [27] [28]
[29] [30]
[31]
[32]
[33]
[34]
[35]
and substance P in magnesium de¢ciency, Cardiovasc. Res. 31 (1996) 677^682. J.S. Downey, J. Han, Cellular activation mechanisms in septic shock, Front. Biosci. 3 (1998) D468^D476. S. Bahrami, H. Redl, G. Schlag, Models of endotoxemia in rodents, in: G. Schlag, H. Redl (Eds.), Pathophysiology of Shock, Sepsis, and Organ Failure, Springer-Verlag, Berlin, 1993, pp. 1019^1030. H. Redl, G. Schlag, S. Bahrami, J. Davies, A. Waage, M. Ceska, W.A. Buurman, G. Adolf, The cytokine network in trauma and sepsis. I. TNF and IL-8, in: G. Schlag, H. Redl (Eds.), Pathophysiology of Shock, Sepsis, and Organ Failure, Springer-Verlag, Berlin, 1993, pp. 468^490. L. Martindale, F.W. Heaton, Magnesium de¢ciency in the adult rat, Biochem. J. 92 (1994) 119^126. L.T. Iseri, J.H. French, Magnesium: nature's physiologic calcium blocker, Am. Heart J. 108 (1984) 188^193. R.S. Hotchkiss, W.N. Bowling, I.E. Karl, D.F. Osborne, M.W. Flye, Calcium antagonists inhibit oxidative burst and nitrite formation in lipopolysaccharide-stimulated rat peritoneal macrophages, Shock 8 (1997) 170^178. G. Schlag, H. Redl (Eds.), Pathophysiology of Shock, Sepsis, and Organ Failure, Springer-Verlag, Berlin, 1993. R.K. Rude, Magnesium homeostasis, in: M. Bilezikian (Ed.), Principles of Bone Biology, Academic Press, London, 1996, pp. 277^293. R.C. Bone, C.J. Fisher, T.P. Clemmer, C.J. Slotman, C.A. Metz, R.A. Balk, Sepsis syndrome: a valid clinical entity, Crit. Care Med. 171 (1989) 389^393. T. Calandra, J. Baumgartner G.E. Grau, M. Wu, P. Lambert, J. Schellekens, J. Verhoef, M.P. Glauser, Swiss-Dutch J5. Immunoglobulin Study Group, Prognostic values of tumor necrosis factor/cachectin, interleukin-1, interferon-K, and interferon-Q in the serum of patients with septic shock, J. Infect. Dis. 161 (1990) 982^987. J.M. Debets, H.R. Kampmeijer, P.M.H. van der Linden, W.A. Buurman, C.J. van der Linden, Plasma tumor necrosis factor and mortality in critically ill septic patients, Crit. Care Med. 17 (1989) 489^494. A. Waage, A. Halstensen, T. Espevik, An association between tumor necrosis factor and serum and fatal outcome in patients with meningococcal disease, Lancet 14 (1987) 355^357. A. Waage, P. Brandtzaeg, A. Halstelsen, P. Kierulf, T. Espevik, The complex pattern of cytokines in serum from patients with meningococcal septic shock, J. Exp. Med. 169 (1989) 333^338.
BBADIS 61785 23-12-98