- Email: [email protected]
Systemic cytokine response in moribund mice of streptococcal toxic shock syndrome model
Microbial Pathogenesis 50 (2011) 109e113
Contents lists available at ScienceDirect
Microbial Pathogenesis journal homepage: www.elsevier.com/locate/micpath
Systemic cytokine response in moribund mice of streptococcal toxic shock syndrome model Mitsumasa Saito a, b, c, *, Hideko Kajiwara b, Ken-ichiro Iida b, Takayuki Hoshina c, Koichi Kusuhara c, d, Toshiro Hara c, Shin-ichi Yoshida b a
Molecular Structure & Function Program, Research Institute, The Hospital for Sick Children, 555 University Ave., Toronto, Ontario M5G 1X8, Canada Department of Bacteriology, Faculty of Medical Sciences, Kyushu University, Higashi-ku, Fukuoka, Fukuoka 812-8582, Japan Department of Pediatrics, Faculty of Medical Sciences, Kyushu University, Higashi-ku, Fukuoka, Fukuoka 812-8582, Japan d Department of Pediatrics, University of Occupational and Environmental Health, Yahata-nishi-ku, Kitakyushu, Fukuoka 807-8555, Japan b c
a r t i c l e i n f o
a b s t r a c t
Article history: Received 20 October 2009 Received in revised form 25 November 2010 Accepted 2 December 2010 Available online 10 December 2010
Streptococcus pyogenes causes severe invasive disease in humans, including streptococcal toxic shock syndrome (STSS). We previously reported a mouse model that is similar to human STSS. When mice were infected intramuscularly with 107 CFU of S. pyogenes, all of them survived acute phase of infection. After 20 or more days of infection, a number of them died suddenly accompanied by S. pyogenes bacteremia. We call this phenomenon “delayed death”. We analyzed the serum cytokine levels of mice with delayed death, and compared them with those of mice who died in the acute phase of intravenous S. pyogenes infection. The serum levels of TNF-a and IFN-g in mice of delayed death were more than 100 times higher than those in acute death mice. IL-10 and IL-12, which were not detected in acute death, were also significantly higher in mice of delayed death. IL-6 and MCP-1 (CCL-2) were elevated in both groups of mice. It was noteworthy that not only pro-inflammatory cytokines but also anti-inflammatory cytokines were elevated in delayed death. We also found that intravenous TNF-a injection accelerated delayed death, suggesting that an increase of serum TNF-a induced S. pyogenes bacteremia in our mouse model. Ó 2010 Elsevier Ltd. All rights reserved.
Keywords: Streptococcus pyogenes Streptococcal toxic shock syndrome Cytokine Mouse model
1. Introduction Streptococcal toxic shock syndrome (STSS), caused by Streptococcus pyogenes (S. pyogenes), is a severe and life-threatening condition characterized by sepsis, necrotizing fasciitis, and hypotension. In typical cases of human STSS, healthy people suddenly suffer from the above signs and symptoms, and die within 48 h after onset. Infection by causative S. pyogenes strains is mainly through the mucous membrane or skin, but specific portal of invasion can be pinpointed only in half of the patients [1]. The pathogenesis of shock in STSS is not fully understood. It is thought that an excessive production of various cytokines is responsible for severe systemic effects such as vasodilation and multiple organ failure in STSS patients [2]. Although there have been few reports on cytokine levels in human STSS [3e5], the treatments targeting cytokines, such as human polyspecific immunoglobulin [4,6e8],
* Corresponding author. Molecular Structure & Function Program, Research Institute, The Hospital for Sick Children, 555 University Ave., Toronto, Ontario M5G 1X8, Canada. Tel.: þ1 416 813 5998; fax: þ1 416 813 5993. E-mail address: [email protected] (M. Saito). 0882-4010/$ e see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.micpath.2010.12.001
hemodiafiltration [4], and plasma exchange [3,4] have led to improved survival in STSS patients. Animal models have been employed to analyze the pathogenesis of STSS. Most of the previous animal studies showed the death of mice in the acute phase of S. pyogenes infection [9e15]. When the organisms are injected intraperitoneally or directly into the blood vessels of mice, the mice fall into bacteremia immediately and die in the acute phase of infection. In contrast, intramuscular injection with 107 CFU of S. pyogenes did not cause bacteremia in the mice in the acute phase of infection [16,17]. This is why the intravenous route of S. pyogenes infection has been used in most STSS mouse model previously. Although the onset of STSS is acute, it is unlikely that a large amount of organisms of S. pyogenes directly enters into the bloodstream to cause STSS, judging from the clinical features of human STSS. There have been a few reports describing that non-intravenous injection with S. pyogenes could cause severe infection to mice by administration of inducers or by concomitant viral infection. In 2002, Diao et al. [16] reported that intramuscularly infected mice showed augmentation of bacterial growth, muscular necrosis and death when treated with Escherichia coli lipopolysaccharide (LPS). In 2003, Okamoto et al. [11] reported that prior infection with
110
M. Saito et al. / Microbial Pathogenesis 50 (2011) 109e113
influenza A virus induced a lethal synergism, resulting in the induction of invasive S. pyogenes infection in mice. In 2001, we reported a simple STSS mouse model [17]. When the mice were infected intramuscularly with 107 CFU of S. pyogenes, all of them survived the acute phase, became apparently healthy, and gained body weight. After 20 or more days of infection, some of them suddenly suffered from sickness such as akinesia and wet fur, and died within 24 h after the onset of illness accompanied by S. pyogenes bacteremia. We called this phenomenon “delayed death”. The organism obtained from the organs of the mice with delayed death showed thick capsule. We found that the hyaluronic acid (capsule) synthesis gene, hasA, is essential in causing delayed death [18]. This mouse model resembles the human STSS in the following three points: 1) approximately 20e30% of infected mice develop necrotizing fasciitis, 2) they suffer from bacteremia and a sudden onset of shock that leads to death, 3) onset is accelerated by bruising on their legs [19]. Our STSS mouse model is thought to be suitable for the analysis of the pathogenic mechanism because it requires no artificial trigger. Here we aimed to clarify the involvement of cytokines in the pathogenesis of the STSS model. We analyzed the serum cytokine levels of mice with delayed death in our model, comparing these levels with those in the acute phase of intravenous S. pyogenes infection. We may predict the pathophysiology of STSS by elucidating the differences in cytokine response between acute and delayed death. 2. Results 2.1. Morbidity and mortality of mice infected with S. pyogenes SP2 strain intramuscularly or intravenously S. pyogenes strain SP2 was injected intramuscularly into the right foreleg or intravenously through the lateral tail vein. All eleven mice inoculated intramuscularly survived the acute phase and became apparently healthy. After the 20th day of infection, however, the mice suddenly started to suffer from sickness observed in a shock state, such as akinesia, closed-eyes, rough fur, vascular dilatation of the ears, tachypnea, paleness and coldness. Our previous studies [17e19] showed that all the mice under this severe condition will die within 24 h and never recover from such a state. Therefore we took whole blood from each moribund mouse and then sacrificed it. All the mice were dead by the 55th day of infection (Fig. 1). In contrast, the mice inoculated intravenously with bacteria became moribund two days after infection. Whole blood was obtained from a total of 10 mice at this acute infectious phase (Fig. 1).
Fig. 1. Survival curves of ddY mice infected with 107 CFU of S. pyogenes SP2 strain. The mice were inoculated with 107 CFU of S. pyogenes intramuscularly (i.m.) into the right foreleg (solid line, n ¼ 11) or intravenously (i.v.) through the lateral tail vein (dotted line, n ¼ 10). The survival days were monitored over a period of 55 days. The mice inoculated intravenously died in two days after infection. In contrast, all eleven mice inoculated intramuscularly survived the acute phase and became apparently healthy. After the 20th day of infection, the mice suddenly started to suffer from sickness observed in a shock state and died by the 55th day of infection.
of vasoactive and inflammatory mediators such as IL-1b and IL-6. To clarify when the serum level of TNF-a increases, we measured the concentration in sera, over time after infection. A total of fifty ddY mice, which were numbered before experiment, were injected with 4.2 106 CFU of S. pyogenes, intramuscularly, and the whole blood of two mice was taken in numerical order every two or three days after infection. Seven mice died without collection of whole blood by the 51st day of infection, and they were removed from this study. The serum TNF-a levels of all mice, except one, were under 50 pg/ml on the measurement day (data not shown). Only one mouse showed an extremely high concentration of TNF-a which was over 300 pg/ml on day 36. This mouse presented akinesia
2.2. Cytokine profile in sera of mice with acute or delayed death We analyzed the serum cytokine levels of mice with delayed death, and compared them with those in the acute phase of intravenous S. pyogenes infection. The mice with delayed death by intramuscular S. pyogenes infection showed significantly higher levels of cytokines than those with acute death by intravenous infection. As shown in Fig. 2, the serum levels of TNF-a and IFN-g were more than 100 times higher than those in acute death. IL-10 and IL-12p70 were detected at high concentrations in delayed, but not acute, death. The serum levels of IL-6 and MCP-1 were elevated in both groups of mice. IL-2, IL-4, and IL-5 were undetectable or of very low concentration in all mice. 2.3. Serial measurement of serum TNF-a in mice after intramuscular infection We focused on the high serum level of TNF-a in mice with delayed death, because it is known that TNF-a stimulates a cascade
Fig. 2. Cytokine profile in sera of mice infected with S. pyogenes SP2 strain intravenously or intramuscularly. Whole blood of each mouse was taken from the femoral artery under direct vision with skin incision just before death. The sera were collected by centrifuging the blood obtained from mice and frozen at 20 C until analysis. The concentrations of IL-2, IL-4, IL-5, IL-6, IL-10, IL-12p70, MCP-1, TNF-a and IFN-g were measured, and results were presented as means SD. The mice suffering from delayed death after intramuscular (i.m.) S. pyogenes infection (black bars, n ¼ 11) showed the higher serum levels of TNF-a and IFN-g than those in acute death caused by intravenous (i.v.) infection (white bars, n ¼ 10). IL-10 and IL-12p70 were detected at high concentrations in delayed (black bars), but not acute (white bars), death. The serum levels of IL-6 and MCP-1 were elevated in both groups of mice. IL-2, IL-4, and IL-5 were undetectable or of very low concentration in all mice. ND: Not detected.
M. Saito et al. / Microbial Pathogenesis 50 (2011) 109e113
when we took its blood. It might have been in the shock state just before delayed death. 2.4. Effect of administration of recombinant mouse TNF-a on the mortality of infected mice To investigate the effect of TNF-a on delayed death, we injected rTNF-a intravenously everyday after the 20th day of infection to maintain high serum level of TNF-a. As shown in Fig. 3, two out of ten control mice died by the 30th day of infection. Meanwhile, twelve out of twenty mice administrated with rTNF-a died of sepsis by the 30th day. Obviously, the death rate was higher in mice injected with TNF-a than in the control mice (P < 0.005). 2.5. Effect of TNF-a on bacterial growth and capsular formation in vitro To clarify whether TNF-a enhances the growth of S. pyogenes or not, the organisms were cultured in BHI broth or TNF-a-supplemented BHI broth. There was no significant difference in the increase of turbidity between them (data not shown). The nigrosine staining of capsule was negative in bacteria cultured with or without TNF-a (data not shown). 2.6. Serum LPS level of mice with delayed death Diao et al. [16] previously reported that LPS induced TNF-a production and death in the acute infectious phase of mice inoculated intramuscularly with 5 108 of S. pyogenes. To determine whether the delayed death was induced by LPS in our mouse model, we measured the serum LPS level just before delayed death by limulus amebocyte lysate assay. LPS was not detected in any serum of the eleven mice which died delayed death (data not shown). Using C3H/HeJ mice which lack response to LPS, we also
Fig. 3. Effect of administration of recombinant mouse TNF-a (rTNF-a) on the mortality of the mice infected with 107 CFU of S. pyogenes SP2 strain intramuscularly. After the 20th day of infection, twenty mice (white circles) were intravenously injected with 10 ng/mouse of rTNF-a through the lateral tail vein once everyday (arrows), while another ten mice (black circles) were injected with PBS as the control. The survival days were monitored over a period of 30 days. The death rate was higher in mice injected with TNF-a than in the control mice (P < 0.005).
111
examined whether delayed death occurred with or without translocating LPS. All of C3H/HeJ mice inoculated with 107 CFU of S. pyogenes intramuscularly died between the 31st day and the 91st day of infection. There was no difference in the time course of mortality between C3H/HeJ mice and control C3H/HeN mice (data not shown). 3. Discussion Our present study showed that the mice inoculated intramuscularly with S. pyogenes died after more than 20 days of infection accompanied by the hypercytokinemia. On the other hand, all the mice inoculated intravenously with S. pyogenes died in the acute infectious phase. We found that there were prominent differences in the serum cytokine profiles between mice with “acute” and “delayed” death. In delayed death, the serum levels of TNF-a and IFN-g were more than 100 times higher than those in acute death.IL-10 and IL-12, which were not detected in acute death, were also high in delayed death. These cytokine profiles in the delayed death are similar to that in human STSS described in the following reports. Drenger et al. [3] showed that concentrations of TNF-a and IL-1 were significantly elevated in the plasma of septic patients caused by S. pyogenes. Norrby-Teglund et al. [5] reported that IL-6 was detected in the serum of STSS patients. Kawaguchi et al. [4] reported that IL-6 levels were remarkably elevated, while TNF-a was not detected in two out of three STSS cases. Wang et al. [20] determined the expression of inflammatory cytokines in children with different severity of S. pyogenes infections. Their patients with invasive diseases had significantly higher IFN-g, IL-1b, IL-6, IL-8, IL-10 and IL-18 than those with noninvasive disease, colonization, and healthy control, and there were no differences in TNF-a, IL-2 and IL-12 levels among the groups. Thus, high serum levels of IFN-g, IL-6 and IL-10 in patients with STSS have been reported. Not all human cases showed the high serum level of TNF-a.This is probably because TNF-a has a short life time, and the response is transient. Cytokines are often classified as pro-inflammatory (TNF-a, IFN-g, and IL-6) or anti-inflammatory cytokines (IL-10) based on their biologic effects [20]. It was noteworthy that not only pro-inflammatory cytokines but also anti-inflammatory cytokines were elevated in the serum of mice with delayed death. In contrast, no mice with acute death showed the high serum level of anti-inflammatory cytokines. Recently, the term “systemic cytokine response” is used to describe an immune system that has over-reacted and is damaging the body, causing failure of multiple organ systems. In such a condition, a systemic release of various cytokines is not properly regulated, and the high blood levels of the pro-inflammatory cytokines induce an autodestructive generalized inflammatory reaction. It is known that both pro-inflammatory and anti-inflammatory cytokines are elevated in the condition of systemic cytokine response. Sekine et al. [21] investigated the plasma levels of various cytokines, in patients with systemic inflammatory response syndrome (SIRS) and sepsis, and showed both pro-inflammatory cytokines (TNF-a, IL-6) and anti-inflammatory cytokines (IL-10) are elevated in the serum of these patients. The delayed death in our STSS mouse model showed the characteristic cytokine profiles found in systemic cytokine response. We suspect that the development of S. pyogenes bacteremia caused the systemic cytokine response, and this leads to death. Then how do the latent organisms localized in the muscle disseminate in the blood? Diao et al. [16,22] provided important insights into this question. They reported that the number of organisms in the muscles after inoculation was significantly higher when the mice were infected with 5 108 CFU S. pyogenes and treated with rTNF-a [16]. Therefore we investigated the effect of TNF-a on delayed death
112
M. Saito et al. / Microbial Pathogenesis 50 (2011) 109e113
in our mouse model, and found that the death rate was higher in mice injected with TNF-a intravenously everyday after the 20th day of infection than in control mice. It was also found that the serum level of TNF-a increased just before death in our delayed death mouse model. These results suggest that a sudden increase in the serum level of TNF-a may trigger the bacteremia. As we previously reported [17e19], the disseminated organisms express thick capsules and multiply rapidly. In this study, we showed that TNF-a was not directly involved in these phenomena, that is, TNF-a had no effect on bacterial growth and capsular formation in vitro. After all, the mechanisms that cause bacterial dissemination remain to be clarified. What triggered the production of TNF-a in our STSS model? Diao et al. [16] reported that LPS treatment elevated the death rate of mice infected with S. pyogenes while anti-TNF-a monoclonal antibody decreased the mortality of mice. In our study, however, LPS was not detected in the sera of mice which died a delayed death, and the delayed death of C3H/HeJ mice, which lack response to LPS, was also observed. Although TNF-a induces muscular necrosis and sepsis, and LPS is one of the triggers of TNF-a production, LPS is not essential for delayed death in our mouse model. Our STSS mouse model reminds us of Brill-Zinsser disease (BZD) which is a relapsed form of epidemic typhus, caused by Rickettsia prowazekii. The concept of BZD is remarkably similar to our STSS model in that the infected individuals control the infection for a certain time after acute phase until reactivation phase, but there are some differences between them. R. prowazekii that causes initial infection remains viable for many years in the host, however, the length of asymptomatic period after acute S. pyogenes infection in our model is much shorter (approximately 20 days). This might be due to the difference between human and mice, and the human patient with STSS could be infected initially several months or years ago. In terms of trigger for the reactivation, it is known that stress or waning immune system directly reactivate the persistent pathogen in BZD. Our STSS model showed S. pyogenes increased the expression of capsule and induced the systemic cytokine response during the reactivation phase. It is not clear whether the waning immune system plays a role as the trigger of reactivation in our STSS model, however, the high level of anti-inflammatory cytokines could cause the immunocompromised status. It should be noted that BZD is milder than typhus due to the presence of IgG antibody which is developed during the convalescent phase of initial infection. In contrast, the reactivation phase in our STSS model is more severe than acute phase and often causes delayed death. The measurement of IgG level against S. pyogenes of mice during the reactivation phase may be needed. In conclusion, the mice with delayed death after intramuscular S. pyogenes infection showed significantly higher levels of cytokines than those with acute death by intravenous infection. Not only pro-inflammatory cytokines but also anti-inflammatory cytokines were elevated in the serum of mice with S. pyogenes bacteremia. This condition, so-called “systemic cytokine response”, may cause multiple organ failure and lead to death. It was suggested that a TNF-a increase triggered the bacteremia. Our simple STSS mouse model is suitable to further analyze pathogenic mechanisms of STSS because it requires no artificial trigger.
mouse model [17e19]. The organisms were cultured in brain heart infusion (BHI) broth (Eiken Chemical, Tokyo) at 37 C with shaking under aerobic conditions. Exponential-phase organisms grown in BHI broth were harvested by low-speed centrifugation. The cells were washed twice and resuspended in sterile phosphate-buffered saline (PBS) at a density of 108 CFU/ml. The cell density was routinely adjusted by measuring the turbidity at 660 nm, which had been calibrated against CFU determined by plating on a bloodsupplemented BHI agar medium. 4.2. Mouse experiments Six-week-old male mice of ddY strain, purchased from Japan SLC (Hamamatsu, Shizuoka, Japan) were used. The mice were inoculated with 107 CFU of S. pyogenes in 0.1 ml of PBS intramuscularly into the right foreleg or intravenously through the lateral tail vein. Whole blood of each mouse was taken from the femoral artery under direct vision with skin incision just before death. All these procedures were carried out under diethylether anesthesia. To clarify the role of LPS, we used six-week-old male mice of the C3H/HeN and C3H/HeJ strains, purchased from Japan SLC. The experimental protocol was reviewed by the Committee of Ethics on Animal Experiments of the Faculty of Medical Sciences, Kyushu University, and carried out in accordance with the Animal Experiment Guidelines of the Faculty, Law No. 105 and the Notification No. 6 of the Japanese Government. 4.3. Measurement of serum cytokines The serum was collected by centrifuging the blood obtained from the mice and frozen at 20 C until analysis. The concentrations of IL-2, IL-4, IL-5, IL-6, IL-10, IL-12p70, MCP-1, TNF-a and IFN-g were measured using BD Cytometric Bead Array (CBA) Mouse TH1/TH2 Cytokine Kit (BD, Franklin Lakes, NJ), according to the manufacturer’s instructions. 4.4. Administration of recombinant mouse TNF-a (rTNF-a) to the infected mice Twenty days after intramuscular infection, the mice were injected with 10 ng/mouse of rTNF-a (R&D Systems Inc., Minneapolis, MN), intravenously, through the lateral tail vein once everyday. PBS was injected as a control for TNF-a. 4.5. Effects of rTNF-a on bacterial growth and capsule formation in vitro S. pyogenes strain SP2 were inoculated into the BHI broth (Eiken Chemical, Tokyo, Japan) and divided into two tubes. rTNF-a was added to one tube to a final concentration of 5 ng/ml. These samples were cultured at 37 C with shaking under aerobic conditions. To monitor the growth at intervals, OD660 of bacterial cultures was measured. To observe the capsule of the bacteria, the cultured samples were stained with nigrosine, which stained the background black while the capsule remained unstained.
4. Materials and methods 4.6. Assay for serum LPS 4.1. Bacterial strain and culture media The S. pyogenes strain SP2, originally isolated from an STSS patient (M type unknown, T type 11, SpeA negative, and SpeB positive) was used in this study. This SP2 is one of the strains which are known to cause delayed death in mice, and has been used in our
The serum was collected by centrifuging the blood obtained from mice and frozen at 20 C until analysis. The LPS concentration in the serum was measured by limulus amebocyte lysate assay with Endotoxin Single Test (Wako Pure Chemical Industries, Ltd., Osaka, Japan), according to the manufacturer’s instructions.
M. Saito et al. / Microbial Pathogenesis 50 (2011) 109e113
4.7. Statistics KaplaneMeier survival analysis was applied to verify the effect of TNF-a and the statistical differences were evaluated by the logrank test. Student’s t-test was used to determine the significance of differences of cytokine titers between the two groups. Acknowledgements We thank Sharon Y.A.M. Villanueva and Linda Saza for valuable editorial advice on the manuscript. This research was supported by Grants-in-Aid for Young Scientists (B) No. 16790554 and Scientific Research (C) No. 18591112 from the Ministry of Education, Culture, Sports, Science and Technology of Japan. References [1] Weiss KA, Laverdiere M. Group A Streptococcus invasive infections: a review. Can J Surg 1997;40:18e25. [2] Pahlman LI, Olin AI, Darenberg J, Morgelin M, Kotb M, Herwald H, et al. Soluble M1 protein of Streptococcus pyogenes triggers potent T cell activation. Cell Microbiol 2008;10:404e14. [3] Drenger B, Israeli A, Or R, Leitersdorf E. Plasmapheresis for streptococcal sepsis? Lancet 1985;2:943. [4] Kawaguchi T, Igaki N, Kinoshita S, Matsuda T, Kida A, Moriguchi R, et al. A new therapeutic strategy for streptococcal toxic shock syndrome: a key target for cytokines. Intern Med 2003;42:211e8. [5] Norrby-Teqlund A, Pauksens K, Norgren M, Holm SE. Correlation between serum TNF alpha and IL6 levels and severity of group A streptococcal infections. Scand J Infect Dis 1995;27:125e30. [6] Darenberg J, Ihendyane N, Sjölin J, Aufwerber E, Haidl S, Follin P, et al. Intravenous immunoglobulin G therapy in streptococcal toxic shock syndrome e a European randomized double-blind placebo-controlled trial. Clin Infect Dis 2003;37:333e40. [7] Kaul R, McGeer A, Norrby-Teglund A, Kotb M, Schwartz B, O’Rourke K, et al. Intravenous immunoglobulin therapy for streptococcal toxic shock syndrome e a comparative observational study. Clin Infect Dis 1999;28:800e7. [8] Norrby-Teglund A, Muller MP, McGeer A, Gan BS, Guru V, Bohnen J, et al. Successful management of severe group A streptococcal soft tissue infections using intravenous polyspecific immunoglobulin and a conservative surgical approach. Scand J Infect Dis 2005;37:166e72.
113
[9] Ashbaugh CD, Warren HB, Carey VJ, Wessels MR. Molecular analysis of the role of the group A streptococcal cysteine protease, hyaluronic acid capsule, and M protein in a murine model of human invasive soft-tissue infection. J Clin Invest 1998;102:550e60. [10] Boyle MD, Raeder R, Flosdorff A, Podbielski A. Role of emm and mrp genes in the virulence of group A streptococcal isolate 64/14 in a mouse model of skin infection. J Infect Dis 1998;177:991e7. [11] Okamoto S, Kawabata S, Nakagawa I, Okuno Y, Goto T, Sano K, et al. Influenza A virus-infected hosts boost invasive type of Streptococcus pyogenes infection in mice. J Virol 2003;77:4104e12. [12] Raeder R, Boyle MD. Association between expression of immunoglobulin G-binding proteins by group A streptococci and virulence in a mouse skin infection model. Infect Immun 1993;61:1378e84. [13] Schmidt KH, Podbielski A, Raeder R, Boyle MD. Inactivation of single genes within the virulence regulon of an M2 group A streptococcal isolate result in differences in virulence for chicken embryos and for mice. Microb Pathog 1997;23:347e55. [14] Shiseki M, Miwa K, Nemoto Y, Kato H, Suzuki J, Sekiya K, et al. Comparison of pathogenic factors expressed by group A streptococci isolated from patients with streptococcal toxic shock syndrome and scarlet fever. Microb Pathog 1999;27:243e52. [15] Sriskandan S, Unnikrishnan M, Krausz T, Cohen J. Molecular analysis of the role of streptococcal pyrogenic exotoxin A (SPEA) in invasive soft-tissue infection resulting from Streptococcus pyogenes. Mol Microbiol 1999;33:778e90. [16] Diao H, Kohanawa M, Yimin, Nakajima H, Sato Y, Minagawa T. Lipopolysaccharide triggers invasive streptococcal disease in mice through a tumour necrosis factor-alpha-dependent mechanism. Immunology 2002;105:344e9. [17] Saito M, Kajiwara H, Ishikawa T, Iida K, Endo M, Hara T, et al. Delayed onset of systemic bacterial dissemination and subsequent death in mice injected intramuscularly with Streptococcus pyogenes strains. Microbiol Immunol 2001;45:777e86. [18] Iida K, Seki M, Saito M, Kawamura Y, Kajiwara H, Yoshida S. Capsule of Streptococcus pyogenes is essential for delayed death of mice in a model of streptococcal toxic shock syndrome. Microbiol Immunol 2006;50:127e30. [19] Seki M, Saito M, Iida K, Taniai H, Soejima T, Nakayama H, et al. Onset of streptococcal toxic shock syndrome is accelerated by bruising in a mouse model. Microb Pathog 2008;44:339e43. [20] Wang SM, Lu IH, Lin YL, Lin YS, Wu JJ, Chuang WJ, et al. The severity of Streptococcus pyogenes infections in children is significantly associated with plasma levels of inflammatory cytokines. Diagn Microbiol Infect Dis 2008;61: 165e9. [21] Sekine K, Fujishima S, Aikawa N. Plasma hepatocyte growth factor is increased in early-phase sepsis. J Infect Chemother 2004;10:110e4. [22] Diao H, Kohanawa M. Endogenous interleukin-6 plays a crucial protective role in streptococcal toxic shock syndrome via suppression of tumor necrosis factor alpha production. Infect Immun 2005;73:3745e8.
Contents lists available at ScienceDirect
Microbial Pathogenesis journal homepage: www.elsevier.com/locate/micpath
Systemic cytokine response in moribund mice of streptococcal toxic shock syndrome model Mitsumasa Saito a, b, c, *, Hideko Kajiwara b, Ken-ichiro Iida b, Takayuki Hoshina c, Koichi Kusuhara c, d, Toshiro Hara c, Shin-ichi Yoshida b a
Molecular Structure & Function Program, Research Institute, The Hospital for Sick Children, 555 University Ave., Toronto, Ontario M5G 1X8, Canada Department of Bacteriology, Faculty of Medical Sciences, Kyushu University, Higashi-ku, Fukuoka, Fukuoka 812-8582, Japan Department of Pediatrics, Faculty of Medical Sciences, Kyushu University, Higashi-ku, Fukuoka, Fukuoka 812-8582, Japan d Department of Pediatrics, University of Occupational and Environmental Health, Yahata-nishi-ku, Kitakyushu, Fukuoka 807-8555, Japan b c
a r t i c l e i n f o
a b s t r a c t
Article history: Received 20 October 2009 Received in revised form 25 November 2010 Accepted 2 December 2010 Available online 10 December 2010
Streptococcus pyogenes causes severe invasive disease in humans, including streptococcal toxic shock syndrome (STSS). We previously reported a mouse model that is similar to human STSS. When mice were infected intramuscularly with 107 CFU of S. pyogenes, all of them survived acute phase of infection. After 20 or more days of infection, a number of them died suddenly accompanied by S. pyogenes bacteremia. We call this phenomenon “delayed death”. We analyzed the serum cytokine levels of mice with delayed death, and compared them with those of mice who died in the acute phase of intravenous S. pyogenes infection. The serum levels of TNF-a and IFN-g in mice of delayed death were more than 100 times higher than those in acute death mice. IL-10 and IL-12, which were not detected in acute death, were also significantly higher in mice of delayed death. IL-6 and MCP-1 (CCL-2) were elevated in both groups of mice. It was noteworthy that not only pro-inflammatory cytokines but also anti-inflammatory cytokines were elevated in delayed death. We also found that intravenous TNF-a injection accelerated delayed death, suggesting that an increase of serum TNF-a induced S. pyogenes bacteremia in our mouse model. Ó 2010 Elsevier Ltd. All rights reserved.
Keywords: Streptococcus pyogenes Streptococcal toxic shock syndrome Cytokine Mouse model
1. Introduction Streptococcal toxic shock syndrome (STSS), caused by Streptococcus pyogenes (S. pyogenes), is a severe and life-threatening condition characterized by sepsis, necrotizing fasciitis, and hypotension. In typical cases of human STSS, healthy people suddenly suffer from the above signs and symptoms, and die within 48 h after onset. Infection by causative S. pyogenes strains is mainly through the mucous membrane or skin, but specific portal of invasion can be pinpointed only in half of the patients [1]. The pathogenesis of shock in STSS is not fully understood. It is thought that an excessive production of various cytokines is responsible for severe systemic effects such as vasodilation and multiple organ failure in STSS patients [2]. Although there have been few reports on cytokine levels in human STSS [3e5], the treatments targeting cytokines, such as human polyspecific immunoglobulin [4,6e8],
* Corresponding author. Molecular Structure & Function Program, Research Institute, The Hospital for Sick Children, 555 University Ave., Toronto, Ontario M5G 1X8, Canada. Tel.: þ1 416 813 5998; fax: þ1 416 813 5993. E-mail address: [email protected] (M. Saito). 0882-4010/$ e see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.micpath.2010.12.001
hemodiafiltration [4], and plasma exchange [3,4] have led to improved survival in STSS patients. Animal models have been employed to analyze the pathogenesis of STSS. Most of the previous animal studies showed the death of mice in the acute phase of S. pyogenes infection [9e15]. When the organisms are injected intraperitoneally or directly into the blood vessels of mice, the mice fall into bacteremia immediately and die in the acute phase of infection. In contrast, intramuscular injection with 107 CFU of S. pyogenes did not cause bacteremia in the mice in the acute phase of infection [16,17]. This is why the intravenous route of S. pyogenes infection has been used in most STSS mouse model previously. Although the onset of STSS is acute, it is unlikely that a large amount of organisms of S. pyogenes directly enters into the bloodstream to cause STSS, judging from the clinical features of human STSS. There have been a few reports describing that non-intravenous injection with S. pyogenes could cause severe infection to mice by administration of inducers or by concomitant viral infection. In 2002, Diao et al. [16] reported that intramuscularly infected mice showed augmentation of bacterial growth, muscular necrosis and death when treated with Escherichia coli lipopolysaccharide (LPS). In 2003, Okamoto et al. [11] reported that prior infection with
110
M. Saito et al. / Microbial Pathogenesis 50 (2011) 109e113
influenza A virus induced a lethal synergism, resulting in the induction of invasive S. pyogenes infection in mice. In 2001, we reported a simple STSS mouse model [17]. When the mice were infected intramuscularly with 107 CFU of S. pyogenes, all of them survived the acute phase, became apparently healthy, and gained body weight. After 20 or more days of infection, some of them suddenly suffered from sickness such as akinesia and wet fur, and died within 24 h after the onset of illness accompanied by S. pyogenes bacteremia. We called this phenomenon “delayed death”. The organism obtained from the organs of the mice with delayed death showed thick capsule. We found that the hyaluronic acid (capsule) synthesis gene, hasA, is essential in causing delayed death [18]. This mouse model resembles the human STSS in the following three points: 1) approximately 20e30% of infected mice develop necrotizing fasciitis, 2) they suffer from bacteremia and a sudden onset of shock that leads to death, 3) onset is accelerated by bruising on their legs [19]. Our STSS mouse model is thought to be suitable for the analysis of the pathogenic mechanism because it requires no artificial trigger. Here we aimed to clarify the involvement of cytokines in the pathogenesis of the STSS model. We analyzed the serum cytokine levels of mice with delayed death in our model, comparing these levels with those in the acute phase of intravenous S. pyogenes infection. We may predict the pathophysiology of STSS by elucidating the differences in cytokine response between acute and delayed death. 2. Results 2.1. Morbidity and mortality of mice infected with S. pyogenes SP2 strain intramuscularly or intravenously S. pyogenes strain SP2 was injected intramuscularly into the right foreleg or intravenously through the lateral tail vein. All eleven mice inoculated intramuscularly survived the acute phase and became apparently healthy. After the 20th day of infection, however, the mice suddenly started to suffer from sickness observed in a shock state, such as akinesia, closed-eyes, rough fur, vascular dilatation of the ears, tachypnea, paleness and coldness. Our previous studies [17e19] showed that all the mice under this severe condition will die within 24 h and never recover from such a state. Therefore we took whole blood from each moribund mouse and then sacrificed it. All the mice were dead by the 55th day of infection (Fig. 1). In contrast, the mice inoculated intravenously with bacteria became moribund two days after infection. Whole blood was obtained from a total of 10 mice at this acute infectious phase (Fig. 1).
Fig. 1. Survival curves of ddY mice infected with 107 CFU of S. pyogenes SP2 strain. The mice were inoculated with 107 CFU of S. pyogenes intramuscularly (i.m.) into the right foreleg (solid line, n ¼ 11) or intravenously (i.v.) through the lateral tail vein (dotted line, n ¼ 10). The survival days were monitored over a period of 55 days. The mice inoculated intravenously died in two days after infection. In contrast, all eleven mice inoculated intramuscularly survived the acute phase and became apparently healthy. After the 20th day of infection, the mice suddenly started to suffer from sickness observed in a shock state and died by the 55th day of infection.
of vasoactive and inflammatory mediators such as IL-1b and IL-6. To clarify when the serum level of TNF-a increases, we measured the concentration in sera, over time after infection. A total of fifty ddY mice, which were numbered before experiment, were injected with 4.2 106 CFU of S. pyogenes, intramuscularly, and the whole blood of two mice was taken in numerical order every two or three days after infection. Seven mice died without collection of whole blood by the 51st day of infection, and they were removed from this study. The serum TNF-a levels of all mice, except one, were under 50 pg/ml on the measurement day (data not shown). Only one mouse showed an extremely high concentration of TNF-a which was over 300 pg/ml on day 36. This mouse presented akinesia
2.2. Cytokine profile in sera of mice with acute or delayed death We analyzed the serum cytokine levels of mice with delayed death, and compared them with those in the acute phase of intravenous S. pyogenes infection. The mice with delayed death by intramuscular S. pyogenes infection showed significantly higher levels of cytokines than those with acute death by intravenous infection. As shown in Fig. 2, the serum levels of TNF-a and IFN-g were more than 100 times higher than those in acute death. IL-10 and IL-12p70 were detected at high concentrations in delayed, but not acute, death. The serum levels of IL-6 and MCP-1 were elevated in both groups of mice. IL-2, IL-4, and IL-5 were undetectable or of very low concentration in all mice. 2.3. Serial measurement of serum TNF-a in mice after intramuscular infection We focused on the high serum level of TNF-a in mice with delayed death, because it is known that TNF-a stimulates a cascade
Fig. 2. Cytokine profile in sera of mice infected with S. pyogenes SP2 strain intravenously or intramuscularly. Whole blood of each mouse was taken from the femoral artery under direct vision with skin incision just before death. The sera were collected by centrifuging the blood obtained from mice and frozen at 20 C until analysis. The concentrations of IL-2, IL-4, IL-5, IL-6, IL-10, IL-12p70, MCP-1, TNF-a and IFN-g were measured, and results were presented as means SD. The mice suffering from delayed death after intramuscular (i.m.) S. pyogenes infection (black bars, n ¼ 11) showed the higher serum levels of TNF-a and IFN-g than those in acute death caused by intravenous (i.v.) infection (white bars, n ¼ 10). IL-10 and IL-12p70 were detected at high concentrations in delayed (black bars), but not acute (white bars), death. The serum levels of IL-6 and MCP-1 were elevated in both groups of mice. IL-2, IL-4, and IL-5 were undetectable or of very low concentration in all mice. ND: Not detected.
M. Saito et al. / Microbial Pathogenesis 50 (2011) 109e113
when we took its blood. It might have been in the shock state just before delayed death. 2.4. Effect of administration of recombinant mouse TNF-a on the mortality of infected mice To investigate the effect of TNF-a on delayed death, we injected rTNF-a intravenously everyday after the 20th day of infection to maintain high serum level of TNF-a. As shown in Fig. 3, two out of ten control mice died by the 30th day of infection. Meanwhile, twelve out of twenty mice administrated with rTNF-a died of sepsis by the 30th day. Obviously, the death rate was higher in mice injected with TNF-a than in the control mice (P < 0.005). 2.5. Effect of TNF-a on bacterial growth and capsular formation in vitro To clarify whether TNF-a enhances the growth of S. pyogenes or not, the organisms were cultured in BHI broth or TNF-a-supplemented BHI broth. There was no significant difference in the increase of turbidity between them (data not shown). The nigrosine staining of capsule was negative in bacteria cultured with or without TNF-a (data not shown). 2.6. Serum LPS level of mice with delayed death Diao et al. [16] previously reported that LPS induced TNF-a production and death in the acute infectious phase of mice inoculated intramuscularly with 5 108 of S. pyogenes. To determine whether the delayed death was induced by LPS in our mouse model, we measured the serum LPS level just before delayed death by limulus amebocyte lysate assay. LPS was not detected in any serum of the eleven mice which died delayed death (data not shown). Using C3H/HeJ mice which lack response to LPS, we also
Fig. 3. Effect of administration of recombinant mouse TNF-a (rTNF-a) on the mortality of the mice infected with 107 CFU of S. pyogenes SP2 strain intramuscularly. After the 20th day of infection, twenty mice (white circles) were intravenously injected with 10 ng/mouse of rTNF-a through the lateral tail vein once everyday (arrows), while another ten mice (black circles) were injected with PBS as the control. The survival days were monitored over a period of 30 days. The death rate was higher in mice injected with TNF-a than in the control mice (P < 0.005).
111
examined whether delayed death occurred with or without translocating LPS. All of C3H/HeJ mice inoculated with 107 CFU of S. pyogenes intramuscularly died between the 31st day and the 91st day of infection. There was no difference in the time course of mortality between C3H/HeJ mice and control C3H/HeN mice (data not shown). 3. Discussion Our present study showed that the mice inoculated intramuscularly with S. pyogenes died after more than 20 days of infection accompanied by the hypercytokinemia. On the other hand, all the mice inoculated intravenously with S. pyogenes died in the acute infectious phase. We found that there were prominent differences in the serum cytokine profiles between mice with “acute” and “delayed” death. In delayed death, the serum levels of TNF-a and IFN-g were more than 100 times higher than those in acute death.IL-10 and IL-12, which were not detected in acute death, were also high in delayed death. These cytokine profiles in the delayed death are similar to that in human STSS described in the following reports. Drenger et al. [3] showed that concentrations of TNF-a and IL-1 were significantly elevated in the plasma of septic patients caused by S. pyogenes. Norrby-Teglund et al. [5] reported that IL-6 was detected in the serum of STSS patients. Kawaguchi et al. [4] reported that IL-6 levels were remarkably elevated, while TNF-a was not detected in two out of three STSS cases. Wang et al. [20] determined the expression of inflammatory cytokines in children with different severity of S. pyogenes infections. Their patients with invasive diseases had significantly higher IFN-g, IL-1b, IL-6, IL-8, IL-10 and IL-18 than those with noninvasive disease, colonization, and healthy control, and there were no differences in TNF-a, IL-2 and IL-12 levels among the groups. Thus, high serum levels of IFN-g, IL-6 and IL-10 in patients with STSS have been reported. Not all human cases showed the high serum level of TNF-a.This is probably because TNF-a has a short life time, and the response is transient. Cytokines are often classified as pro-inflammatory (TNF-a, IFN-g, and IL-6) or anti-inflammatory cytokines (IL-10) based on their biologic effects [20]. It was noteworthy that not only pro-inflammatory cytokines but also anti-inflammatory cytokines were elevated in the serum of mice with delayed death. In contrast, no mice with acute death showed the high serum level of anti-inflammatory cytokines. Recently, the term “systemic cytokine response” is used to describe an immune system that has over-reacted and is damaging the body, causing failure of multiple organ systems. In such a condition, a systemic release of various cytokines is not properly regulated, and the high blood levels of the pro-inflammatory cytokines induce an autodestructive generalized inflammatory reaction. It is known that both pro-inflammatory and anti-inflammatory cytokines are elevated in the condition of systemic cytokine response. Sekine et al. [21] investigated the plasma levels of various cytokines, in patients with systemic inflammatory response syndrome (SIRS) and sepsis, and showed both pro-inflammatory cytokines (TNF-a, IL-6) and anti-inflammatory cytokines (IL-10) are elevated in the serum of these patients. The delayed death in our STSS mouse model showed the characteristic cytokine profiles found in systemic cytokine response. We suspect that the development of S. pyogenes bacteremia caused the systemic cytokine response, and this leads to death. Then how do the latent organisms localized in the muscle disseminate in the blood? Diao et al. [16,22] provided important insights into this question. They reported that the number of organisms in the muscles after inoculation was significantly higher when the mice were infected with 5 108 CFU S. pyogenes and treated with rTNF-a [16]. Therefore we investigated the effect of TNF-a on delayed death
112
M. Saito et al. / Microbial Pathogenesis 50 (2011) 109e113
in our mouse model, and found that the death rate was higher in mice injected with TNF-a intravenously everyday after the 20th day of infection than in control mice. It was also found that the serum level of TNF-a increased just before death in our delayed death mouse model. These results suggest that a sudden increase in the serum level of TNF-a may trigger the bacteremia. As we previously reported [17e19], the disseminated organisms express thick capsules and multiply rapidly. In this study, we showed that TNF-a was not directly involved in these phenomena, that is, TNF-a had no effect on bacterial growth and capsular formation in vitro. After all, the mechanisms that cause bacterial dissemination remain to be clarified. What triggered the production of TNF-a in our STSS model? Diao et al. [16] reported that LPS treatment elevated the death rate of mice infected with S. pyogenes while anti-TNF-a monoclonal antibody decreased the mortality of mice. In our study, however, LPS was not detected in the sera of mice which died a delayed death, and the delayed death of C3H/HeJ mice, which lack response to LPS, was also observed. Although TNF-a induces muscular necrosis and sepsis, and LPS is one of the triggers of TNF-a production, LPS is not essential for delayed death in our mouse model. Our STSS mouse model reminds us of Brill-Zinsser disease (BZD) which is a relapsed form of epidemic typhus, caused by Rickettsia prowazekii. The concept of BZD is remarkably similar to our STSS model in that the infected individuals control the infection for a certain time after acute phase until reactivation phase, but there are some differences between them. R. prowazekii that causes initial infection remains viable for many years in the host, however, the length of asymptomatic period after acute S. pyogenes infection in our model is much shorter (approximately 20 days). This might be due to the difference between human and mice, and the human patient with STSS could be infected initially several months or years ago. In terms of trigger for the reactivation, it is known that stress or waning immune system directly reactivate the persistent pathogen in BZD. Our STSS model showed S. pyogenes increased the expression of capsule and induced the systemic cytokine response during the reactivation phase. It is not clear whether the waning immune system plays a role as the trigger of reactivation in our STSS model, however, the high level of anti-inflammatory cytokines could cause the immunocompromised status. It should be noted that BZD is milder than typhus due to the presence of IgG antibody which is developed during the convalescent phase of initial infection. In contrast, the reactivation phase in our STSS model is more severe than acute phase and often causes delayed death. The measurement of IgG level against S. pyogenes of mice during the reactivation phase may be needed. In conclusion, the mice with delayed death after intramuscular S. pyogenes infection showed significantly higher levels of cytokines than those with acute death by intravenous infection. Not only pro-inflammatory cytokines but also anti-inflammatory cytokines were elevated in the serum of mice with S. pyogenes bacteremia. This condition, so-called “systemic cytokine response”, may cause multiple organ failure and lead to death. It was suggested that a TNF-a increase triggered the bacteremia. Our simple STSS mouse model is suitable to further analyze pathogenic mechanisms of STSS because it requires no artificial trigger.
mouse model [17e19]. The organisms were cultured in brain heart infusion (BHI) broth (Eiken Chemical, Tokyo) at 37 C with shaking under aerobic conditions. Exponential-phase organisms grown in BHI broth were harvested by low-speed centrifugation. The cells were washed twice and resuspended in sterile phosphate-buffered saline (PBS) at a density of 108 CFU/ml. The cell density was routinely adjusted by measuring the turbidity at 660 nm, which had been calibrated against CFU determined by plating on a bloodsupplemented BHI agar medium. 4.2. Mouse experiments Six-week-old male mice of ddY strain, purchased from Japan SLC (Hamamatsu, Shizuoka, Japan) were used. The mice were inoculated with 107 CFU of S. pyogenes in 0.1 ml of PBS intramuscularly into the right foreleg or intravenously through the lateral tail vein. Whole blood of each mouse was taken from the femoral artery under direct vision with skin incision just before death. All these procedures were carried out under diethylether anesthesia. To clarify the role of LPS, we used six-week-old male mice of the C3H/HeN and C3H/HeJ strains, purchased from Japan SLC. The experimental protocol was reviewed by the Committee of Ethics on Animal Experiments of the Faculty of Medical Sciences, Kyushu University, and carried out in accordance with the Animal Experiment Guidelines of the Faculty, Law No. 105 and the Notification No. 6 of the Japanese Government. 4.3. Measurement of serum cytokines The serum was collected by centrifuging the blood obtained from the mice and frozen at 20 C until analysis. The concentrations of IL-2, IL-4, IL-5, IL-6, IL-10, IL-12p70, MCP-1, TNF-a and IFN-g were measured using BD Cytometric Bead Array (CBA) Mouse TH1/TH2 Cytokine Kit (BD, Franklin Lakes, NJ), according to the manufacturer’s instructions. 4.4. Administration of recombinant mouse TNF-a (rTNF-a) to the infected mice Twenty days after intramuscular infection, the mice were injected with 10 ng/mouse of rTNF-a (R&D Systems Inc., Minneapolis, MN), intravenously, through the lateral tail vein once everyday. PBS was injected as a control for TNF-a. 4.5. Effects of rTNF-a on bacterial growth and capsule formation in vitro S. pyogenes strain SP2 were inoculated into the BHI broth (Eiken Chemical, Tokyo, Japan) and divided into two tubes. rTNF-a was added to one tube to a final concentration of 5 ng/ml. These samples were cultured at 37 C with shaking under aerobic conditions. To monitor the growth at intervals, OD660 of bacterial cultures was measured. To observe the capsule of the bacteria, the cultured samples were stained with nigrosine, which stained the background black while the capsule remained unstained.
4. Materials and methods 4.6. Assay for serum LPS 4.1. Bacterial strain and culture media The S. pyogenes strain SP2, originally isolated from an STSS patient (M type unknown, T type 11, SpeA negative, and SpeB positive) was used in this study. This SP2 is one of the strains which are known to cause delayed death in mice, and has been used in our
The serum was collected by centrifuging the blood obtained from mice and frozen at 20 C until analysis. The LPS concentration in the serum was measured by limulus amebocyte lysate assay with Endotoxin Single Test (Wako Pure Chemical Industries, Ltd., Osaka, Japan), according to the manufacturer’s instructions.
M. Saito et al. / Microbial Pathogenesis 50 (2011) 109e113
4.7. Statistics KaplaneMeier survival analysis was applied to verify the effect of TNF-a and the statistical differences were evaluated by the logrank test. Student’s t-test was used to determine the significance of differences of cytokine titers between the two groups. Acknowledgements We thank Sharon Y.A.M. Villanueva and Linda Saza for valuable editorial advice on the manuscript. This research was supported by Grants-in-Aid for Young Scientists (B) No. 16790554 and Scientific Research (C) No. 18591112 from the Ministry of Education, Culture, Sports, Science and Technology of Japan. References [1] Weiss KA, Laverdiere M. Group A Streptococcus invasive infections: a review. Can J Surg 1997;40:18e25. [2] Pahlman LI, Olin AI, Darenberg J, Morgelin M, Kotb M, Herwald H, et al. Soluble M1 protein of Streptococcus pyogenes triggers potent T cell activation. Cell Microbiol 2008;10:404e14. [3] Drenger B, Israeli A, Or R, Leitersdorf E. Plasmapheresis for streptococcal sepsis? Lancet 1985;2:943. [4] Kawaguchi T, Igaki N, Kinoshita S, Matsuda T, Kida A, Moriguchi R, et al. A new therapeutic strategy for streptococcal toxic shock syndrome: a key target for cytokines. Intern Med 2003;42:211e8. [5] Norrby-Teqlund A, Pauksens K, Norgren M, Holm SE. Correlation between serum TNF alpha and IL6 levels and severity of group A streptococcal infections. Scand J Infect Dis 1995;27:125e30. [6] Darenberg J, Ihendyane N, Sjölin J, Aufwerber E, Haidl S, Follin P, et al. Intravenous immunoglobulin G therapy in streptococcal toxic shock syndrome e a European randomized double-blind placebo-controlled trial. Clin Infect Dis 2003;37:333e40. [7] Kaul R, McGeer A, Norrby-Teglund A, Kotb M, Schwartz B, O’Rourke K, et al. Intravenous immunoglobulin therapy for streptococcal toxic shock syndrome e a comparative observational study. Clin Infect Dis 1999;28:800e7. [8] Norrby-Teglund A, Muller MP, McGeer A, Gan BS, Guru V, Bohnen J, et al. Successful management of severe group A streptococcal soft tissue infections using intravenous polyspecific immunoglobulin and a conservative surgical approach. Scand J Infect Dis 2005;37:166e72.
113
[9] Ashbaugh CD, Warren HB, Carey VJ, Wessels MR. Molecular analysis of the role of the group A streptococcal cysteine protease, hyaluronic acid capsule, and M protein in a murine model of human invasive soft-tissue infection. J Clin Invest 1998;102:550e60. [10] Boyle MD, Raeder R, Flosdorff A, Podbielski A. Role of emm and mrp genes in the virulence of group A streptococcal isolate 64/14 in a mouse model of skin infection. J Infect Dis 1998;177:991e7. [11] Okamoto S, Kawabata S, Nakagawa I, Okuno Y, Goto T, Sano K, et al. Influenza A virus-infected hosts boost invasive type of Streptococcus pyogenes infection in mice. J Virol 2003;77:4104e12. [12] Raeder R, Boyle MD. Association between expression of immunoglobulin G-binding proteins by group A streptococci and virulence in a mouse skin infection model. Infect Immun 1993;61:1378e84. [13] Schmidt KH, Podbielski A, Raeder R, Boyle MD. Inactivation of single genes within the virulence regulon of an M2 group A streptococcal isolate result in differences in virulence for chicken embryos and for mice. Microb Pathog 1997;23:347e55. [14] Shiseki M, Miwa K, Nemoto Y, Kato H, Suzuki J, Sekiya K, et al. Comparison of pathogenic factors expressed by group A streptococci isolated from patients with streptococcal toxic shock syndrome and scarlet fever. Microb Pathog 1999;27:243e52. [15] Sriskandan S, Unnikrishnan M, Krausz T, Cohen J. Molecular analysis of the role of streptococcal pyrogenic exotoxin A (SPEA) in invasive soft-tissue infection resulting from Streptococcus pyogenes. Mol Microbiol 1999;33:778e90. [16] Diao H, Kohanawa M, Yimin, Nakajima H, Sato Y, Minagawa T. Lipopolysaccharide triggers invasive streptococcal disease in mice through a tumour necrosis factor-alpha-dependent mechanism. Immunology 2002;105:344e9. [17] Saito M, Kajiwara H, Ishikawa T, Iida K, Endo M, Hara T, et al. Delayed onset of systemic bacterial dissemination and subsequent death in mice injected intramuscularly with Streptococcus pyogenes strains. Microbiol Immunol 2001;45:777e86. [18] Iida K, Seki M, Saito M, Kawamura Y, Kajiwara H, Yoshida S. Capsule of Streptococcus pyogenes is essential for delayed death of mice in a model of streptococcal toxic shock syndrome. Microbiol Immunol 2006;50:127e30. [19] Seki M, Saito M, Iida K, Taniai H, Soejima T, Nakayama H, et al. Onset of streptococcal toxic shock syndrome is accelerated by bruising in a mouse model. Microb Pathog 2008;44:339e43. [20] Wang SM, Lu IH, Lin YL, Lin YS, Wu JJ, Chuang WJ, et al. The severity of Streptococcus pyogenes infections in children is significantly associated with plasma levels of inflammatory cytokines. Diagn Microbiol Infect Dis 2008;61: 165e9. [21] Sekine K, Fujishima S, Aikawa N. Plasma hepatocyte growth factor is increased in early-phase sepsis. J Infect Chemother 2004;10:110e4. [22] Diao H, Kohanawa M. Endogenous interleukin-6 plays a crucial protective role in streptococcal toxic shock syndrome via suppression of tumor necrosis factor alpha production. Infect Immun 2005;73:3745e8.