IN-VITRO STIMULATION OF TNF-a FROM HUMAN WHOLE BLOOD BY CELL-FREE SUPERNATANTS OF GRAM-POSITIVE BACTERIA Katherine
Bayston,
Mark Tomlinson,
Jonathan
Cohen
Gram-positive bacteria are being recognized increasingly as the cause of shock-like syndromes, clinically indistinguishable from those seen in association with Gram-negative endotoxic shock. Much clinical and experimental data link tumour necrosis factor-a (TNF-c~) with the pathogenesis of endotoxic shock, and a number of studies of individual Gram-positive species have also implicated TNF-a. We report here the first systematic study of the ability of cell-free supernatants of common Gram-positive bacteria to induce TNF-CY from human peripheral blood monocytes in vitro. Almost all the 63 strains were able to induce TNF-cu, although the levels were substantially lower than those obtained from supernatants of Gram-negative bacteria, used as controls. Streptococcuspneumoniue, S. pyogenes, viridans streptococci and coagulase-negative staphylococci were consistently more active than group B and D streptococci. TNF-(w induction did not correlate with conventional markers of pathogenicity; amongst strains of Staphylococcus aureus, commensal and blood culture isolates did not induce significantly different amounts of TNF. We conclude that cell-free supernatants of most Gram-positive bacteria are capable of inducing TNF-cw from human peripheral blood monocytes in vitro, but the significance of this finding remains to be determined.
In recent years much has been learnt about the pathogenesis of Gram-negative bacterial shock. However, a significant proportion of episodes of sepsis are caused by Gram-positive bacteria and it is apparent that these too are associated with a significant incidence of shock and death.i,2 Clinically, shock due to Gram-positive organisms is indistinguishable from that associated with Gram-negative septicaemia, suggesting that a final common pathway may be involved in the pathogenesis of septic shock, irrespective of the causative organism. Many of the pathological consequences of Gram-negative shock are related to endotoxin,
From the Department of Infectious Diseases & Bacteriology, Royal Postgraduate Medical School, Du Cane Road, London W12 ONN. Correspondence to: Prof J Cohen, Department of Infectious Diseases & Bacteriology, Royal Postgraduate Medical School, DU Cane Road, London-W12 ONN. Received 26 Februarv 1992: accented for nublication 20 March 1992 01992 Academic Press Limited 1043-4666/92/050397+06 $08.00/O KEY
WORDS:
CYTOKINE,
endotoxinlgram-positive/TNF-a
Vol.
4, No. 5 (September),
1992: pp 397-402
the lipopolysaccharide (LPS) component of the Gram-negative bacterial outer cell wall (reviewed in ref. 3). Much, if not all, of the toxicity of LPS is brought about by a series of host-derived mediators rather than by LPS itself. In particular, cytokines such as tumour necrosis factor-o (TNF) have been implicated. TNF is produced both in vitro and in vivo in response to LPS challenge.47 Gram-negative septicaemia in man and in animals is associated with elevated circulating levels of TNF and other cytokines,s,9 and the degree and duration of elevation of these cytokines has been associated with outcome.107ii Finally, we and others have shown that monoclonal antibodies to TNF protect animals from experimental Gram-negative shock.i2,13 The pathogenesis of Gram-positive shock is not so well understood. There are few animal models of Gram-positive shock and in vitro data are scarce. Elevated levels of TNF have been found in the blood (and CSF) of patients with Gram-positive septicaemia,l4,15 but the possible role of concomitant endotoxaemia has not always been addressed in these studies. Several investigators have proposed that endotoxin derived from bacteria in the gut may leak into the circulation and contribute to the pathogenesis of septic shock due to microorganisms 397
398
I Bayston
CYTOKINE,
et al.
Log dilutions of E. coli LPS, between 100 pg/ml and 1 pg/ml, were included on each plate to serve as a positive control. A representative dose response curve is illustrated in Fig. 1. Positive control supernatants (E. coli and K. pneumoniae) contained 14.2 l&/ml and 8 pg/ml endotoxin respectively, and consistently induced 2.0-2.5 rig/ml of TNF. All media controls failed to induce TNF.
Induction of TNF by Bacterial Supernatants To ascertain the optimal time for incubation of bacterial supernatants with blood, a series of time curves were performed (Fig. 2). In all cases the optimal time for TNF induction was 8 h and this was used routinely thereafter. TNF was induced in human whole blood by the majority of Gram-positive bacterial supernatants. Although there was occasional variability in the ability of individual strains within groups of bacteria to induce TNF, some consistent patterns emerged (Fig. 3). Coagulase-negative staphylococci were the
Bacteria Sixty-three strains of Gram-positive bacteria were used in stimulation experiments. Of the S. aureuS strains one was positive for TSST-1 and enterotoxin
TABLE 1. Details of isolates studied. Density indicates number of organisms present (see Methods for details). Type strain designations refer to NCTC accession number. Density
Coagulase-negative
S. pyogenes
staphylococci
NO.
Blood/pus Nose/throat Type strains
9 3 2
5.7 k 3.7 x 10x
Blood
3.3 k 0.9 x 105
Skin Type strain Blood Throat Type strain
7 4 I 2 1 5
Blood Type strain Blood Type strain
5 1 9 2
Blood Type strain
5 2
Blood Type strain
4 1
lo7
S. agalactiae
8.0 + 4.9 x 10’
Gp D streptococci
4.8 + 1.3 x lo”
Viridans
streptococci
S. pneumoniae
(cfu/ml)
Site
1.8+0.9x
aweus
4.0 i- 2.1 x 10’
5.7 & 1.6 x 10’
397-402)
Induction of TNF by L.PS
RESULTS
Staphylococcus
1992:
A, three for enterotoxin A alone, two for enterotoxin D and one for enterotoxin C and D. The density of growth (expressed as &/ml) after overnight culture was similar within groups of bacteria, but differed considerably between groups (Table 1) All bacterial supernatants were sterile and contained < 50 pg/ml endotoxin. No cell wall fragments were identified in supernatants examined by electron microscopy.
other than Gram-negative bacteria.r6 However, in the absence of endotoxaemia, non-toxic shock syndrome toxin 1 (TSST-1) producing strains of Staphylococcus aureu~ can induce the same cardiovascular abnormalities of septic shock in experimental animals as Escherichiu coZi.17 TSST-1 ,is staphylococcal (Ytoxin,19 certain staphylococcal enterotoxins and streptococcal pyrogenic exotoxin2’J have all been shown to induce the production of TNF and/or interleukin 1 (IL-l) by monocytes in vitro. Pneumococcal cell wall preparations failed to induce TNF (although IL-l was released) in vitro, 21 but in an experimental model of Gram-positive meningitis anti-TNF antibody reduced inflammation induced by heat-killed Streptococcus pneumoniae.22 In this study we have investigated the ability of cell-free bacterial supernatants derived from cultures of Staphylococcus aureus, coagulase-negative staphylococci, Streptococcus pneumoniae and groups A, B, D and viridans type streptococci, to induce TNF in human whole blood. We have shown that TNF is induced by the majority of organisms tested, and that the amount of TNF released does not appear to correlate with conventional markers of virulence.
Description
Vol. 4, No. 5 (September
after overnight
incubation
Comment
See legend All MRSA NC0 8532 NC0 6571
NC 11047
NCO:
8195; 8324;
8191; 8198; 10877
NC0
8181
NC0 NC0
0775 S. faecalis 7171 S. faecium
NC0 NC0
0775 S. mitis 7863 S. sanguis
NC0
746
Stimulation
most potent inducers of TNF, followed by a group which included viridans type streptococci, group A streptococci and S. pneumoniae. Group A streptococci produced significantly more TNF than group B and D streptococci (P=O.OOOS and P 338. 10. Waage A, Halstensen A, Espevik T (1987) Association between tumour necrosis factor in serum and fatal outcome in patients with meningococcal disease. Lancet 1:355-357. 11. Girardin E, Grau GE, Dayer J-M, Roux-Lombard P, Lambert P-H (1988) Tumor necrosis factor and interleukin-1 in the serum of children with severe infectious purpura. N Engl J Med 319:397-400. 12. Silva AT, Bayston KF, Cohen J (1990) Prophylactic and therapeutic effects of a monoclonal antibody to tumor necrosis factor - alpha in experimental gram negative shock. J Infect Dis 1621421-427. 13. Tracey KJ, Fong Y, Hesse DG, Manogue KR, Lee AT, Kuo GC. Lowrv SF. Cerami A (1987) Anti-cachectin/TNF monoclonal antibodies prevent septic shock during lethal bacteraemia. Nature 330:662-664. 14. Marks JD , Marks CB , Lute JM, Montgomery AB , Turner J, Metz CA, Murray JF (1990) Plasma tumor necrosis factor in patients with septic shock. Mortality rate, incidence of adult respiratory distress syndrome, and effects of methyl prednisolone administration. Am Rev Respir Dis 141:94-97. 15. Baud L, Cadranel J, Offenstadt G, Luquel L, Guidet B, Amstutz P (1990) Tumor necrosis factor and septic shock. Crit Care Med 18:349-350. 16. Stone RL, Schlievert PM (1987) Evidence for the involvement of endotoxin in toxic shock syndrome. J Infect Dis 155:682. 17. Natanson C, Danner RL, Elin RJ, Hosseini JM, Peart KW, Banks SM, MacVittie TJ, Walker RI, Parrillo JE (1989) Role of endotoxemia in cardiovascular dysfunction and mortality. Escherichia coli and Staphylococcus aweus challenges in a canine model of human septic shock. J Clin Invest 83:243-251. 18. Ikejima T, Okusawa S, Van der Meer JW, Dinarello CA (1988) Induction by toxic-shock-syndrome toxin-l of a circulating tumor necrosis factor-like substance in rabbits and of immunoreactive tumor necrosis factor and interleukin-1 from human mononuclear cells. J Infect Dis 158:1017-1025. 19. Fischer H, Dohlsten M, Andersson U, Hedlund G, Ericsson P, Hansson J, Sjogren HO (1990) Production of TNF-CY and TNF-B by staphyloccal enterotoxin A activated human T cells. J Immunol144:4663-4669. 20. Fast DJ, Schlievert PM, Nelson RD (1989) Toxic shock syndrome-associated staphylococcal and streptococcal pyrogenic toxins are potent inducers of tumor necrosis factor production. Infect Immun 57:291-294. 21. Riesenfeld-Orn I, Wolpe S, Garcia-Bustos JF, Hoffman MK, Tuomanen E (1989) Production of interleukin-1 but not tumor necrosis factor by human monocytes stimulated with pneumoncoccal cell surface components. Infect Immun 57:189(&1893. 22. Saukkonen K, Sande S, Cioffe C, Wolpe S, Sherry B, Cerami A, Tuomanen E (1990) The role of cytokines in the generation of inflammation’ and tissue damage in experimental gram positive meningitis. J Exp Med 171:439-448. 23. Cohen J, McConnell JS (1984) Observations on the measurement and evaluation of endotoxemia by a quantitative Limulus lysate microassay. J Infect Dis 150:91&924. ,
\
,
24. Silva AT, Appelmelk BJ, Buurman WA, Bayston KF, Cohen J (1990) Monoclonal antibody to endotoxin core protects mice from Escherichia coli sepsis by a mechanism independent of tumor necrosis factor and interleukin-6. J Infect Dis 162:454-459.
402 I Bayston et al.
CYTOKINE,
25. Kiener PA, Marek F, Rodgers G, Lin P-F, Warr G, Desiderio J (1988) Induction of tumor necrosis factor, IFN-gamma, and acute lethality in mice by toxic and non-toxic forms of Lipid A. J Immunol141:870-874. tions.
26. Stevens Clin Infect
DL (1992) Dis 14:2-13.
Invasive
group
A streptococcus
infec-
27. Debets JM, Kampmijer R, Van der Linden MP, Buurman WA, Van der Linden CJ (1989) Plasma tumor necrosis factor and mortality in critically ill septic patients. Crit Care Med 17:489-494.
Vol.
4, No.
5 (September
28. Stevens DL, Tanner MH, Winship Schlievert PM, Kaplan E (1989) Severe infections associated with a toxic shock-like fever toxin A. N Engl J Med 321:1-7.
J, Swarts group A syndrome
1992: 397-402)
R, Ries KM, streptococcal and scarlet
29. Cohen J, Donnelly JP, Worsley AM, Catovsky D, Goldman JM, Galton DAG (1983) Septicaemia caused by viridans streptococci in neutropenic patients with leukaemia. Lancet 2:1452-1454.
Bayston,
Mark Tomlinson,
Jonathan
Cohen
Gram-positive bacteria are being recognized increasingly as the cause of shock-like syndromes, clinically indistinguishable from those seen in association with Gram-negative endotoxic shock. Much clinical and experimental data link tumour necrosis factor-a (TNF-c~) with the pathogenesis of endotoxic shock, and a number of studies of individual Gram-positive species have also implicated TNF-a. We report here the first systematic study of the ability of cell-free supernatants of common Gram-positive bacteria to induce TNF-CY from human peripheral blood monocytes in vitro. Almost all the 63 strains were able to induce TNF-cu, although the levels were substantially lower than those obtained from supernatants of Gram-negative bacteria, used as controls. Streptococcuspneumoniue, S. pyogenes, viridans streptococci and coagulase-negative staphylococci were consistently more active than group B and D streptococci. TNF-(w induction did not correlate with conventional markers of pathogenicity; amongst strains of Staphylococcus aureus, commensal and blood culture isolates did not induce significantly different amounts of TNF. We conclude that cell-free supernatants of most Gram-positive bacteria are capable of inducing TNF-cw from human peripheral blood monocytes in vitro, but the significance of this finding remains to be determined.
In recent years much has been learnt about the pathogenesis of Gram-negative bacterial shock. However, a significant proportion of episodes of sepsis are caused by Gram-positive bacteria and it is apparent that these too are associated with a significant incidence of shock and death.i,2 Clinically, shock due to Gram-positive organisms is indistinguishable from that associated with Gram-negative septicaemia, suggesting that a final common pathway may be involved in the pathogenesis of septic shock, irrespective of the causative organism. Many of the pathological consequences of Gram-negative shock are related to endotoxin,
From the Department of Infectious Diseases & Bacteriology, Royal Postgraduate Medical School, Du Cane Road, London W12 ONN. Correspondence to: Prof J Cohen, Department of Infectious Diseases & Bacteriology, Royal Postgraduate Medical School, DU Cane Road, London-W12 ONN. Received 26 Februarv 1992: accented for nublication 20 March 1992 01992 Academic Press Limited 1043-4666/92/050397+06 $08.00/O KEY
WORDS:
CYTOKINE,
endotoxinlgram-positive/TNF-a
Vol.
4, No. 5 (September),
1992: pp 397-402
the lipopolysaccharide (LPS) component of the Gram-negative bacterial outer cell wall (reviewed in ref. 3). Much, if not all, of the toxicity of LPS is brought about by a series of host-derived mediators rather than by LPS itself. In particular, cytokines such as tumour necrosis factor-o (TNF) have been implicated. TNF is produced both in vitro and in vivo in response to LPS challenge.47 Gram-negative septicaemia in man and in animals is associated with elevated circulating levels of TNF and other cytokines,s,9 and the degree and duration of elevation of these cytokines has been associated with outcome.107ii Finally, we and others have shown that monoclonal antibodies to TNF protect animals from experimental Gram-negative shock.i2,13 The pathogenesis of Gram-positive shock is not so well understood. There are few animal models of Gram-positive shock and in vitro data are scarce. Elevated levels of TNF have been found in the blood (and CSF) of patients with Gram-positive septicaemia,l4,15 but the possible role of concomitant endotoxaemia has not always been addressed in these studies. Several investigators have proposed that endotoxin derived from bacteria in the gut may leak into the circulation and contribute to the pathogenesis of septic shock due to microorganisms 397
398
I Bayston
CYTOKINE,
et al.
Log dilutions of E. coli LPS, between 100 pg/ml and 1 pg/ml, were included on each plate to serve as a positive control. A representative dose response curve is illustrated in Fig. 1. Positive control supernatants (E. coli and K. pneumoniae) contained 14.2 l&/ml and 8 pg/ml endotoxin respectively, and consistently induced 2.0-2.5 rig/ml of TNF. All media controls failed to induce TNF.
Induction of TNF by Bacterial Supernatants To ascertain the optimal time for incubation of bacterial supernatants with blood, a series of time curves were performed (Fig. 2). In all cases the optimal time for TNF induction was 8 h and this was used routinely thereafter. TNF was induced in human whole blood by the majority of Gram-positive bacterial supernatants. Although there was occasional variability in the ability of individual strains within groups of bacteria to induce TNF, some consistent patterns emerged (Fig. 3). Coagulase-negative staphylococci were the
Bacteria Sixty-three strains of Gram-positive bacteria were used in stimulation experiments. Of the S. aureuS strains one was positive for TSST-1 and enterotoxin
TABLE 1. Details of isolates studied. Density indicates number of organisms present (see Methods for details). Type strain designations refer to NCTC accession number. Density
Coagulase-negative
S. pyogenes
staphylococci
NO.
Blood/pus Nose/throat Type strains
9 3 2
5.7 k 3.7 x 10x
Blood
3.3 k 0.9 x 105
Skin Type strain Blood Throat Type strain
7 4 I 2 1 5
Blood Type strain Blood Type strain
5 1 9 2
Blood Type strain
5 2
Blood Type strain
4 1
lo7
S. agalactiae
8.0 + 4.9 x 10’
Gp D streptococci
4.8 + 1.3 x lo”
Viridans
streptococci
S. pneumoniae
(cfu/ml)
Site
1.8+0.9x
aweus
4.0 i- 2.1 x 10’
5.7 & 1.6 x 10’
397-402)
Induction of TNF by L.PS
RESULTS
Staphylococcus
1992:
A, three for enterotoxin A alone, two for enterotoxin D and one for enterotoxin C and D. The density of growth (expressed as &/ml) after overnight culture was similar within groups of bacteria, but differed considerably between groups (Table 1) All bacterial supernatants were sterile and contained < 50 pg/ml endotoxin. No cell wall fragments were identified in supernatants examined by electron microscopy.
other than Gram-negative bacteria.r6 However, in the absence of endotoxaemia, non-toxic shock syndrome toxin 1 (TSST-1) producing strains of Staphylococcus aureu~ can induce the same cardiovascular abnormalities of septic shock in experimental animals as Escherichiu coZi.17 TSST-1 ,is staphylococcal (Ytoxin,19 certain staphylococcal enterotoxins and streptococcal pyrogenic exotoxin2’J have all been shown to induce the production of TNF and/or interleukin 1 (IL-l) by monocytes in vitro. Pneumococcal cell wall preparations failed to induce TNF (although IL-l was released) in vitro, 21 but in an experimental model of Gram-positive meningitis anti-TNF antibody reduced inflammation induced by heat-killed Streptococcus pneumoniae.22 In this study we have investigated the ability of cell-free bacterial supernatants derived from cultures of Staphylococcus aureus, coagulase-negative staphylococci, Streptococcus pneumoniae and groups A, B, D and viridans type streptococci, to induce TNF in human whole blood. We have shown that TNF is induced by the majority of organisms tested, and that the amount of TNF released does not appear to correlate with conventional markers of virulence.
Description
Vol. 4, No. 5 (September
after overnight
incubation
Comment
See legend All MRSA NC0 8532 NC0 6571
NC 11047
NCO:
8195; 8324;
8191; 8198; 10877
NC0
8181
NC0 NC0
0775 S. faecalis 7171 S. faecium
NC0 NC0
0775 S. mitis 7863 S. sanguis
NC0
746
Stimulation
most potent inducers of TNF, followed by a group which included viridans type streptococci, group A streptococci and S. pneumoniae. Group A streptococci produced significantly more TNF than group B and D streptococci (P=O.OOOS and P 338. 10. Waage A, Halstensen A, Espevik T (1987) Association between tumour necrosis factor in serum and fatal outcome in patients with meningococcal disease. Lancet 1:355-357. 11. Girardin E, Grau GE, Dayer J-M, Roux-Lombard P, Lambert P-H (1988) Tumor necrosis factor and interleukin-1 in the serum of children with severe infectious purpura. N Engl J Med 319:397-400. 12. Silva AT, Bayston KF, Cohen J (1990) Prophylactic and therapeutic effects of a monoclonal antibody to tumor necrosis factor - alpha in experimental gram negative shock. J Infect Dis 1621421-427. 13. Tracey KJ, Fong Y, Hesse DG, Manogue KR, Lee AT, Kuo GC. Lowrv SF. Cerami A (1987) Anti-cachectin/TNF monoclonal antibodies prevent septic shock during lethal bacteraemia. Nature 330:662-664. 14. Marks JD , Marks CB , Lute JM, Montgomery AB , Turner J, Metz CA, Murray JF (1990) Plasma tumor necrosis factor in patients with septic shock. Mortality rate, incidence of adult respiratory distress syndrome, and effects of methyl prednisolone administration. Am Rev Respir Dis 141:94-97. 15. Baud L, Cadranel J, Offenstadt G, Luquel L, Guidet B, Amstutz P (1990) Tumor necrosis factor and septic shock. Crit Care Med 18:349-350. 16. Stone RL, Schlievert PM (1987) Evidence for the involvement of endotoxin in toxic shock syndrome. J Infect Dis 155:682. 17. Natanson C, Danner RL, Elin RJ, Hosseini JM, Peart KW, Banks SM, MacVittie TJ, Walker RI, Parrillo JE (1989) Role of endotoxemia in cardiovascular dysfunction and mortality. Escherichia coli and Staphylococcus aweus challenges in a canine model of human septic shock. J Clin Invest 83:243-251. 18. Ikejima T, Okusawa S, Van der Meer JW, Dinarello CA (1988) Induction by toxic-shock-syndrome toxin-l of a circulating tumor necrosis factor-like substance in rabbits and of immunoreactive tumor necrosis factor and interleukin-1 from human mononuclear cells. J Infect Dis 158:1017-1025. 19. Fischer H, Dohlsten M, Andersson U, Hedlund G, Ericsson P, Hansson J, Sjogren HO (1990) Production of TNF-CY and TNF-B by staphyloccal enterotoxin A activated human T cells. J Immunol144:4663-4669. 20. Fast DJ, Schlievert PM, Nelson RD (1989) Toxic shock syndrome-associated staphylococcal and streptococcal pyrogenic toxins are potent inducers of tumor necrosis factor production. Infect Immun 57:291-294. 21. Riesenfeld-Orn I, Wolpe S, Garcia-Bustos JF, Hoffman MK, Tuomanen E (1989) Production of interleukin-1 but not tumor necrosis factor by human monocytes stimulated with pneumoncoccal cell surface components. Infect Immun 57:189(&1893. 22. Saukkonen K, Sande S, Cioffe C, Wolpe S, Sherry B, Cerami A, Tuomanen E (1990) The role of cytokines in the generation of inflammation’ and tissue damage in experimental gram positive meningitis. J Exp Med 171:439-448. 23. Cohen J, McConnell JS (1984) Observations on the measurement and evaluation of endotoxemia by a quantitative Limulus lysate microassay. J Infect Dis 150:91&924. ,
\
,
24. Silva AT, Appelmelk BJ, Buurman WA, Bayston KF, Cohen J (1990) Monoclonal antibody to endotoxin core protects mice from Escherichia coli sepsis by a mechanism independent of tumor necrosis factor and interleukin-6. J Infect Dis 162:454-459.
402 I Bayston et al.
CYTOKINE,
25. Kiener PA, Marek F, Rodgers G, Lin P-F, Warr G, Desiderio J (1988) Induction of tumor necrosis factor, IFN-gamma, and acute lethality in mice by toxic and non-toxic forms of Lipid A. J Immunol141:870-874. tions.
26. Stevens Clin Infect
DL (1992) Dis 14:2-13.
Invasive
group
A streptococcus
infec-
27. Debets JM, Kampmijer R, Van der Linden MP, Buurman WA, Van der Linden CJ (1989) Plasma tumor necrosis factor and mortality in critically ill septic patients. Crit Care Med 17:489-494.
Vol.
4, No.
5 (September
28. Stevens DL, Tanner MH, Winship Schlievert PM, Kaplan E (1989) Severe infections associated with a toxic shock-like fever toxin A. N Engl J Med 321:1-7.
J, Swarts group A syndrome
1992: 397-402)
R, Ries KM, streptococcal and scarlet
29. Cohen J, Donnelly JP, Worsley AM, Catovsky D, Goldman JM, Galton DAG (1983) Septicaemia caused by viridans streptococci in neutropenic patients with leukaemia. Lancet 2:1452-1454.
