FEMS MicrobiologyImmunology89 (1992) 317-322 © 1992 Federation of European Microbiological Societies 0920-8534/92/$05.00 Published by Elsevier

317

FEMSIM 00220

Interleukin-6 is a better marker of lethality than tumor necrosis factor in endotoxin treated mice N i a m h M. Kelly a n d A l a n S. Cross Department of Bacterial Diseases, Walter Reed Army Institute of Research, Washington D.C., USA

Received 27 January 1992 Revision received 10 April 1992 Accepted 14 April 1992 Key words: Tumor necrosis factor; Interleukin-6; Endotoxemia

1. S U M M A R Y We established a mouse model to differentiate between a lethal and non-lethal presentation of endotoxic shock. The model involved injecting different amounts of E s c h e r i c h i a coli LPS into C 3 H / H e N mice which had been 'primed' with BCG. We found that the mice receiving non-lethal and lethal doses of LPS could not be differentiated in terms of their physical symptoms for the first 8 h post-injection. Tumour necrosis factor (TNF) was detected at concentrations 2-9-fold greater in mice receiving lethal doses of LPS when compared with non-lethally injected mice. However, given that (i) the successful detection of this differential was dependent on the time of sampling and (ii) that T N F was only detected in the first 3 - 4 h post LPS challenge, we suggest that T N F may not be very useful as a prognostic

Correspondence to: N.M. Kelly, Present address: Division of

Medical Microbiology, University of British Columbia, 2733 Heather St., Vancouver, British Columbia V5Z 1M9, Canada.

marker in endotoxic shock. In contrast, circulating IL-6 appeared to mirror the symptoms of the endotoxic mice. The relative disappearance of IL-6 after 10 h in the non-lethally injected mice corresponded with their symptomatic recovery, while IL-6 continued to circulate up to the time of death in the lethally injected mice. Furthermore, there appeared to be a good correlation between the levels of injected LPS and the levels of IL-6 induced into the circulation. Our results suggest that IL-6, rather than TNF, may serve as a prognostic marker for endotoxic shock.

2. I N T R O D U C T I O N The involvement of tumour necrosis factor (TNF) as a mediator of the lethal effects of endotoxic shock, induced by the systemic release of lipopolysaccharide (LPS) from Gram-negative bacteria, has been well established [1,2]. The presumption is that this represents the overzealous behaviour of a normally beneficial role for T N F in the hosts' immune response [1,2]. With this in

318

mind we wished to establish a model which would differentiate between a non-lethal and lethal induction of T N F in response to bacterial LPS. The model we established involves the injection of non-lethal and lethal doses of LPS into mice which have been primed for T N F production by a prior injection of BCG. Using this model we asked if the difference between a non-lethal and lethal induction of T N F reflects (i) differences in the kinetics of circulating TNF, a n d / o r (ii) a difference in the levels of circulating TNF. We also asked similar questions concerning interleukin-6 (IL-6).

3. M A T E R I A L S A N D M E T H O D S 3.1. E. coli strain Bort and isolated LPS E. coli, 018: KI: H7, strain Bort, originally isolated from a neonatal case of meningitis, [3] and therefore considered to be of pathogenic significance, was used in these studies. The organism was grown to log phase in Trypticase soya broth (TSB, BBL Microbiology Systems, Cockeysville, MD) for animal injections. The LPS was isolated using the hot phenol water extraction method [4]. LPS solutions were made up in saline, autoclaved, and sonicated immediately prior to use.

3.2. Injections into mice Female C 3 H / H e N mice were used throughout. The mice had attained 9-12 weeks of age by the time of experimentation. Injections were given intraperitoneally in a final volume of 100 ~xl. A standard BCG vaccination, obtained from the Swiss Serum and Vaccine Institute (Berne, Switzerland) was given at a 1:10 dilution 8 - 1 0 days previous to the LPS injection. Blood was collected from anaesthesised mice by cardiac puncture. The blood was allowed to clot and the serum collected by centrifugation. 3.3. TNF assay T N F activity was assessed on the basis of its cytotoxicity for L929 cells after 48 h incubation, as previously described [5]. The serum samples were first heat-inactivated at 56°C for 30 min and

then diluted three fold in series in 96-well microtitre plates. The L929 cells were laid down at 105 cells per well. After 48 h incubation, at 37°C in 6% CO 2, cell death was determined colorimetrically using the viable dye MTT. The standard curve consisted of five-told dilutions, from 400 n g - 6 n g / m l , of r T N F (Genzyme, Boston, MA) made up in normal mouse serum and heat-inactivated. The T N F in the test samples was computed on a Macintosh computer using the Excel program, and expressed in units. One unit of T N F was the amount required to kill 50% of the L cells. 3.4. IL-6 assay IL-6 activity was assessed on the basis of its mitogenic activity for the IL-6 dependent murine B9 cell line [6]. The B9 cells were laid down at a concentration of 4 - 5 × 10 ~ in 96-well microtitre plates in R P M I 1640 (Hazelton Biologics Inc., St. Lenexa, KS) supplemented with heat-inactivated FCS to 10% (Sigma Chemical Company, St. Louis, MO), 60 /xM 2-mercaptoethanol (Sigma), 60 U / m l penicillin and 60 # g / m l streptomycin (Sigma). The test serum samples were diluted ten-fold in series in R P M I 1640 supplemented with 5% FCS, and 2-mercaptoethanol, penicillin and streptomycin as above. The standard curve consisted of two-fold dilutions of murine rlL-6 which had been kindly provided to us by Dr. Richard Nordan at the N.I.H. Incubation was for 4 days at 37°C in 6% CO 2. Cell growth was measured by pulsing with 0.5 /.t Ci 3H thymidine per well (specific activity = 6.7 C i / m M ) for 4 h. The radioactively labelled cells were harvested (1295 Cell Harvester, Pharmacia LKB Biotechnology Inc., Piscataway, N J) over glass filters (Pharmacia) and sealed into plastic bags (sample bag for use with 1295 Betaplate, Pharmacia) in the presence of scintillation fluid (Beta Plate Scintillation Fluid, Pharmacia). Radioactive determinations were performed on the 1295 Betaplate Liquid Scintillation counter (Pharmacia). The IL-6 in the test sample was computed on a Macintosh computer using the Excel program and expressed in units. One unit of IL-6 was the amount required to promote half maximal growth of the B9 cells.

319

4. R E S U L T S A N D D I S C U S S I O N Initial experiments were conducted to establish the lethality of LPS isolated from E. coli strain Bort for BCG primed C 3 H / H e N mice. These experiments demonstrated that 10 > g of Bort LPS represented a non-lethal dose (LDo), 800 > g represented a lethal dose (LDm0), and 4 0 0 / , g represented a dose of intermediate lethal-. ity (LD~) (Table 1). An interesting observation was that when the virulence of the whole organism and its isolated LPS were compared, the same time course to death was noted (see Table 1). This was particularly noticeable in the case of intermediate doses of LPS or whole organisms where the same percentage of animals died as was seen with higher doses of the LPS or whole organisms but just took longer to do so. In thecase of whole organisms one can easily hypothesise that the longer time to death results from the multiplication of the original inoculum. However, such a hypothesis is meaningless in the case of a non-replicating molecule such as LPS. An alternative explanation might be that death in the case of Gram-negative bacterial endotoxic shock is the end-result of a cascade event in which endot o x i n / L P S serves as the trigger. The shorter time to death in the case of higher concentrations of LPS (or whole organisms) may be an indication that LPS, depending on its concentration, can trigger multiple steps in the cascade. A series of experiments was conducted to establish whether the different pathologies presented by mice receiving non-lethal and lethal injections of LPS could be explained by different levels of circulating TNF. These experiments revealed no difference in the physical symptoms presented by mice receiving non-lethal and lethal doses of LPS in the first 8 h post-injection (Table 2). Mice receiving even the smallest dose of LPS became moribund to the extent that it was impossible to distinguish those receiving a 10 > g nonlethal dose from those receiving a 800 /~g lethal dose. In the second 8 h post-injection the difference became obvious as the mice that received a non-lethal LPS dose began to recover while those that received a lethal dose continued to deteriorate. When the serum was examined for T N F

Table 1 The lethality of E. coli strain Bort and its isolated LPS for C3H/HeN mice Injected agent a

Dose

Animal lethality (dead/alive) days post-injection; 1

2

3

E. coli

105 cfu 104 cfu 1()5 cfu l0 s' cfu 10 7 cfu

0/3 0/3 0/3 2/3 3/3

0/3 1/3 2/3 3/3 3/3

0/3 1/3 3/3 3/3 3/3

LPS b

10 ~,g 100 ug 200 Fzg 400/xg

0/4 0/4 0/3 0/4

0/4 0/4 1/3 3/4

0/4 0/4 3/3 3/4

800 ~,g

3/3

3/3

3/3

~ Injections were i.p. b Mice receiving LPS were pretreated with BCG.

activity the time course of circulating T N F was found to be the same for all three doses of LPS (Table 2). T N F activity was detected in the serum as early as 30 min after an LPS injection and was no longer detectable 4 h post-injection. This transient appearance of T N F has been noted by others after lethal injections of LPS into baboons, rabbits, rats, and mice [1], and after non-lethal injections of LPS into humans [1]. Our results suggest that the transient appearance of T N F in the serum is the same whether the animal has received a non-lethal or lethal dose of LPS. In examining the concentration of circulating T N F a five- to nine-fold increase was recorded in mice receiving the lethal 800 /.tg dose of LPS when compared with mice receiving the 10 /,g nonlethal dose (Fig. 1). However, as is evident in the experiment illustrated in Fig. 1, the time at which the blood sample is taken, in relation to the LPS injection, determines whether, or not, this differential is detected. For this reason, and in association with the very transient nature of circulating T N F in the LPS injected mouse, we suggest that T N F may be an unreliable prognostic marker in cases of endotoxic shock. The absence of circulating T N F leading up to death suggested that if T N F was contributing to the lethal event it must be doing so through another mediator, e.g. as part of a cascade sys-

32O Table 2 The timing of physical symptoms and circulating TNF in BCG stimulated C 3 H / H e N mice injected with ram-lethal and lethal doses of LPS LPS

Hours

Appearance of symptom

Dose

postinjection

Diarrhea

10/~g (LD~0

1 2 4 8 24-48 1 2 4-8 24-48 1 2 4-8 24 48

+

400 /zg (LD i)

800 ~ g (LDio~)

Detectable

Decreased Activity

Uveitis

Death

TNF in serum + +

+ +

" h

+

+ +

+ + +

+ +

+

-

+ + +

T N F was not detected. b The symptom did not appear.

~'

tern. With the idea that IL-6, a cytokine known to be induced by T N F as well as LPS [7] might be one such mediator, we extended our study to include an assessment of circulating IL-6 in mice receiving lethal and non-lethal doses of LPS. The

results showed a good correlation between the concentration of LPS injected into the mouse and the concentration of circulating IL-6 (Fig. 2). Furthermore, the presence of IL-6 in the circulation correlated with the presence of physical

300-

10---0--10 ---0---400

ug LPS ug LPS

+ 8 0 0

ug L P S

200

>,

E

? d

E 100-

u

2

o

1

2

3

Hrs Post Injection

Fig. 1. Levels of circulating TNF in BCG stimulated C 3 H / H e N mice injected with non-lethal and lethal doses of LPS. Each point on the graph represents the mean of two separate experiments. Qualitatively similar results were obtained in each of six repeat experiments.

0

1'O Hrs

2'0

3 I0

Post I n j e c t i o n

Fig. 2. Levels of circulating IL-6 in BCG stimulated C 3 H / H e N mice injected with non-lethal and lethal doses of LPS. Each point on the graph represents the mean of two separate experiments. Qualitatively similar results were obtained in each of four repeat experiments.

321

symptoms such that the relative disappearance of IL-6 after 10 h in mice receiving non-lethal doses of LPS corresponded with their symptomatic recovery; in contrast, IL-6 continued to circulate at easily detectable levels up to 24 h after the injection of lethal doses of LPS (i.e. up to the time of death). Our results, using a mouse model of nonlethal and lethally induced endotoxic shock, suggest that IL-6 may be a more realistic marker of impending lethality than T N F in cases of endotoxic shock. Research from other laboratories suggests that IL-6 may in fact be responsible for the down-regulation of the T N F response [8,9]. The early disappearance of IL-6 in response to non-lethal doses of LPS has been previously recorded in mice [10] and in humans [11,12] while high levels of IL-6 have been recorded in humans in association with a diagnosis of septicemia [13,14]. Furthermore, a correlation has been made between high levels of circulating IL-6 and fatality in human cases of septic shock [13,14]. Whether this high level of circulating IL-6 merely reflects the role of this cytokine as a key mediator of the inflammatory response to the endotoxin or whether it suggests a role for IL-6 in the lethal complications of endotoxic shock is a question currently being addressed in our laboratory, and in others [8]. Starnes et al. [8] have recently shown that anti-IL-6 protects mice challenged with lethal doses of E. coli or T N F suggesting that IL-6 is involved in the lethal pathology. Our study did not address the issue as to whether the IL-6 induced into the circulation was a direct result of the LPS challenge, an indirect result arising from the induction of other cytokines such as T N F or IL-1, or a combination of both. The appearance of large quantities of IL-6 (in the order of ] 0 6 units per ml of serum) within the first 30 min of the LPS challenge suggests a direct role for LPS in the initial IL-6 response. The peak of IL-6 activity observed in the lethally injected mice at 3 h post-injection, which was absent in the non-lethally injected mice (see Fig. 2), may be a result of the excess LPS in these mice a n d / o r a result of the increased T N F production (see Fig. 1). Since IL-6 is known to be produced by various cell types [7], Helle et al. [15] have suggested that LPS and T N F may act in

concert on different cell types to induce high levels of circulating IL-6. Given a reported half life for circulating IL-6 of approximately 1 h [7] its continued presence in the circulation of lethally injected mice suggests that it was being continuously produced (see Fig. 2). The absence of circulating T N F beyond 4 h rules out a role for circulating T N F in the continued production of IL-6. It remains possible, however, that local production of T N F in one of the organ systems, e.g. the liver, may have contributed to the induction of local IL-6 which entered the circulation. In contrast to the short half life of the cytokines, lethal doses of LPS injected intravenously into mice have been reported as having a half life approximating 10 h [16]. It seems reasonable, therefore, to suggest that the continuous presence of circulating LPS in our lethally infected C 3 H / H e N mice may have resulted in the continuous production of circulating IL-6.

ACKNOWLEDGEMENTS N.K. was a recipient of an N.R.C. Research Associateship. We are grateful to Dr. Richard Nordan for supplying B9 cells, IL-6, and helpful advice.

REFERENCES [1] Sherry, B. and Cerami, A. (1988) Cachectin tumour necrosis factor exerts endocrine, paracrine, and autocrine control of inflammatory responses. J. Cell. Biol. 107, 1269-1277. [2] Cerami, A. and Beutler, B. (1988) The role of cachrectin/TNF in endotoxic shock and cachexia. Immunol Today 9, 28-31. [3] Bortolussi, R., Ferrieri, P., Bjorksten, B. and Quie, P.G. (1979) Capsular KI polysaccharide of Escherichia coli: relationship to virulence in newborn rats and resistance to phagocytosis. Infect. Immun. 25, 293-298. [4] Westphal, O. and Jann, K. (1965) Bacterial lipopolysaccharides: Extraction with phenol-water and further application of the procedure. In: Methods in Carbohydrate Chemistry (Whistler, R.L., Ed.), pp. 83-91. Academic Press, New York. [5] Cross, A.S., Sadoff, J.C., Kelly, N., Bernton, E. and Gemski, P. (1988) Pretreatment with recombinant murine

322

[6]

[7] [8]

[9]

[10]

[11]

tumour necrosis factor and alpha cachectin and murine interleukin 1 alpha protects mice from lethal bacterial infection. J. Exp. Med. 169, 2021 2027. Aarden, L.A., De Groot, E.R., Schara, O.L and Lansdorp, P.M. (1987) Production of hybridoma growth factor by human monocytes. Eur. J. lmmun. 17, 1411-1416. Van Snick, J. (1990) Interleukin-6: an overview. Ann. Rev. Immunol. 8, 253-278. Starnes, Jr., H.F., Pearce, M.K., Tewari, A., Yim, J.H., Zou, J.C. and Abrams, J.S. (1990) Anti IL-6 monoclonal antibodies protect against lethal Escherichia coli infection and lethal tumour necrosis factor challenge in mice. J. lmmun. 145, 4185 4191. Ulich, T.R., Yin, S., Guo, K., Yi, E.S., Remick, D. and del Castilto, J. (1991) Intratracheal injection of endotoxin and cytokines. II. Interleukin-6 and transforming growth factor beta inhibit acute inflammation. Am. J. Pathol. 138, 1097-1101. Coulie, P.G., Cayphas, S., Visk, A., Uyltenhove, C. and Van Snick, J. (1987) Interleukin-H Pl-related hybridoma and plasmacytoma growth factors induced by lypopolysaccharide in vivo. Eur. J. lmmun. 17, 1217-1220. Fong, Y., Moldawer, L.L., Marano, M.A., Wei, H., Tatter, S.B., ClarJck, R.H., Santherem, U., Sherris, D., May,

[12]

[13]

[14]

[15]

[16]

L.T., Sehgal, P.B. and Lowry, S.F. (1989) Endotoxemia elicits increased circulating /32-IFN/IL-6 in man. J. lmmun. 142, 2321-2324. Zabel, P., Schoenharting, M.M., Wolter, D.T. and Schade, U.F. (t989) Oxpentifylline in Endotoxaemia. Lancet 11, 1474 1477. Waage, A., Brandtzaeg, P., Halstensen, A., Kierulf, P. and Espevik, T. (1989) The complex pattern of cytokincs in serum from patients with meningococcal septic shock. J. Exp. Med. 169, 333 338. Hack, C.E., De Groot, E.R., Felt-Bersma, R.J.F., Nuijens, J.H., Strack von Schijndel, R.J.M., Cerenberg-Beliner, A.J.M., Thits, L.G. and Aarden, L.A. (1989) Increased plasma levels of interleukin-6 in sepsis. Blood 74. 17(/4 1710. Helle, M., Boeije, L. and Aarden, L. (1989) Differential induction of IL-6 production by monocytes and endothelial cells. Cytokine 1, 76. Morrison, D.C., Duncan, Jr, R.L. and Goodman, S.A. (1985) In vivo biological activities of endotoxin, In: Bacterial endotoxins: structure. Biomedical significance, and detection with the Limulus Amebocyte Lysate Test (ten Care, J.W., Bueller, H.R., Sturk, A. and Levin, J., (Eds.), pp. 81-98. Alan R. Liss, New York.

Interleukin-6 is a better marker of lethality than tumor necrosis factor in endotoxin treated mice.

We established a mouse model to differentiate between a lethal and non-lethal presentation of endotoxic shock. The model involved injecting different ...
375KB Sizes 0 Downloads 0 Views