Infections Post Transplant

Viridans streptococcal septicaemia in neutropenic patients: role of proinflammatory cytokines


The immunostimulatory activity of viridans streptococcal strains isolated from neutropenic patients with severe sepsis (n=9) or uncomplicated bacteraemia (n=10) was compared. Peripheral blood mononuclear cells from healthy individuals were stimulated with heat-killed bacteria or culture supernatants, and cytokine production assessed. All strains were potent inducers of IL1β, IL8, and TNFα production. Heat-killed bacteria induced consistently higher IL1β and TNFα production than did the cell-free bacterial supernatants (P<0.01). The strains did not induce any proliferative response, nor any significant TNFβ or IFNγ production. No difference in cytokine-inducing capacity could be detected between the cohorts of severe and nonsevere isolates. Comparison of strains causing severe and nonsevere episodes in the same patient (n=2) revealed a significantly higher induction of IL1β by the severe episodes associated isolates as compared to the nonsevere (P<0.04). The study underscores the importance of the host–pathogen interplay in determining the level of inflammation, and hence the severity of disease.


Bacterial infection, especially bacteraemia, is one of the major complications during neutropenia following high-dose chemotherapy or stem cell transplantation. During the last decades, infections with Gram-negative bacteria have become less frequent, whereas an increasing incidence of bacteraemia due to Gram-positive bacteria has been reported by several haematological and stem cell transplant centres.1,2 Although the majority of these infections are caused by coagulase-negative staphylococci, a large number are due to viridans streptococci. In contrast to what is seen in septicaemia due to coagulase-negative staphylococci, it has been described that 3–33% of cases of viridans streptococcal bacteraemia among neutropenic patients are associated with rapid onset of severe septic syndrome including shock and acute respiratory distress syndrome (ARDS).3,4,5

Viridans streptococci are normally considered to be of low virulence even though they can occasionally cause infections such as subacute endocarditis and, some subgroups, in particular Streptococcus intermedius (milleri), even cause invasive pyogenic infections, including brain, pulmonary and abdominal abscesses, and osteomyelitis, in immunocompetent individuals.6,7 Relatively little is known about the immunopathogenesis of these diseases. A role for cell wall components of Streptococcus mitis in induction of shock through a TNFα-dependent pathway was suggested in an experimental murine model.8 Previous studies have also shown that viridans streptococci isolated from neutropenic patients can induce proinflammatory cytokines, among others IL1β, IL8, and TNFβ.9,10 This is of interest since IL8 production has been suggested to be associated with neutrophil accumulation and lung damage in patients with ARDS.11,12

In this study we investigated whether varying clinical manifestations of viridans streptococcal bacteraemia in neutropenic patients are related to differences between isolates in their capacity to induce proinflammatory responses.

Material and methods

Clinical isolates and patient characteristics

A retrospective screening of stem cell transplanted and leukaemic patients during the period 1995–2002 was performed in order to identify neutropenic patients with severe or nonsevere sepsis caused by viridans streptococci. Screening was only performed for the last 7 years since this is the current time limitation for storage of blood-culture isolates at the bacteriological laboratory at Huddinge University Hospital, Stockholm, Sweden. One case was included from Hospital for Sick Children, Toronto, Canada. The following definitions were utilized in the screening: neutropenia (ie profound neutropenia), neutrophils <0.1 × 109/l following either high-dose chemotherapy for acute myeloid leukaemia or stem cell transplantation; bacteraemia, growth of bacteria in at least one pair of blood culture-bottles taken in relation to a fever episode during neutropenia; severe sepsis, sepsis associated with shock and ARDS, or organ dysfunction, hypoperfusion, or hypotension; nonsevere sepsis, cases with bacteraemia only related to fever without shock, ARDS, or any organ dysfunction.

Nine patients were identified who, during profound neutropenia, had at least one episode of bacteraemia due to viridans streptococci and developed severe sepsis syndrome with or without ARDS. These strains were compared with 10 isolates from 10 nonsevere septic episodes in neutropenic patients. Cases 2 and 6 experienced two septic episodes, one severe and one nonsevere, caused by different viridans streptococcal strains. The patient characteristics are described in Table 1.

Table 1 Patient demographics, diagnosis, clinical strains, and outcome

In seven of the 19 septic episodes, patients received adequate empiric antibiotic therapy, three were treated with antibiotics that were not completely efficient against their infection, and nine patients had strains resistant against the empiric antibiotic therapy.

Bacterial typing

All 19 viridans streptococcal strains were typed based on the presence of glycoside hydrolases, phosphatases, esterases, and lipases determined using API ZYM (API System, La Balme-Les-Grottes, France). Production of the extracellular polysaccharide dextran was tested by ethanol precipitation.13

Pulse-field gel electrophoresis (PFGE)

Isolates from cases 2 and 6 who experienced two septic episodes (one severe and one nonsevere) were further characterized by PFGE. The PFGE procedure was adapted from the procedure described by Hermans et al.14 Chromo-somal restriction fragments obtained by cleavage with restriction enzyme ApaI were embedded in agarose and separated on a 1% agarose gel by pulse-field electrophoresis performed for 22 h at 14°C at 6 V/cm in 0.5 × Tris-borate-EDTA with pulse times of 2–30 s and an angle of 60°. After electrophoresis, the gel was stained with ethidium bromide for 30 min (1 mg/ml).

Preparation of bacterial culture supernatant and heat-killed bacteria

The isolates were cultured overnight in 15 ml Todd-Hewitt broth (Difco, Detroit, MI, USA) supplemented with 1.5% yeast extract (Difco) at 37°C. A β-haemolytic group-A streptococcal isolate of serotype T3/B3264 treated in the same manner as the viridans streptococcal isolates was used as a positive control at a dilution of 1:100.

Both bacterial culture supernatant and heat-killed bacteria were prepared from each isolate. The bacteria and supernatant were separated by centrifugation, and proteins in the culture supernatants were precipitated by ethanol as previously described in detail.15 The cell-free culture supernatants were stored at −20°C until used. The bacterial pellet was resuspended in 1 ml PBS, boiled for 10 min and washed three times with phosphate-buffered saline (PBS), resuspended in 1 ml PBS, and stored at –20°C until used.

Viable bacterial counts (CFU/ml) were determined for each isolate by serial dilution and plating on blood–agar plates. Dose–response experiments were performed to determine the optimal concentration of cell-free culture supernatants and heat-killed bacteria. A concentration of 6 × 108 CFU/ml was found to be an optimal standard concentration for all strains and for both bacterial preparations.

Lipopolysaccharide (LPS) from E. coli 055:B5 provided by Division of Tumor Biology, Karolinska Institutet, was used at a concentration of 100 ng/ml.

Preparation of peripheral blood mononuclear cells (PBMC) and proliferation assay

PBMC were isolated from heparinized blood of healthy donors by Ficoll–Hypaque gradient centrifugation. The PBMC were cultured in RPMI 1640 medium supplemented with 25 mM HEPES, 4 mM L-glutamine, 100 U/ml penicillin/streptomycin, and 5% heat-inactivated foetal calf serum (FCS). PBMC (1 × 106 cells/ml) were stimulated with the optimal concentrations of sterile culture supernatants or heat-killed bacteria. After 72 h, the cells were pulsed for 6 h with 1 μCi per well of [3H]thymidine (specific activity, 5.0 Ci/mmol; Amersham Pharmacia Biotech, UK). Phytohaemagglutinin-L (PHA) (Sigma, St Louis, MO, USA) was used as a positive control at a concentration of 1 μg/ml. All samples were assayed in triplicate. The experiments were repeated three times using cells from different individuals.

Cytokine analyses

Simultaneously with the proliferation assay, the cell cultures were tested for cytokine production. PBMC (1 × 106 cells/ml), prepared and stimulated as previously described above, were harvested after 4 h of cultures for TNFα analysis, after 24 h for IL1β and IL8, and after 72 h for IL8, TNFβ, and IFNγ analysis. The stimulated cells and cell culture supernatants were separated by centrifugation, and the supernatants were stored at –20°C. Cytokine production was assessed at the single cell level by intracellular immunohistochemical staining, as previously described in detail.16 The following cytokine-specific mAbs were used at concentrations of 2 μg/ml: IL8 (NAP-1, mouse IgG1), IL1β (cocktail of 2.D.8 and 1437.96.85, mouse IgG, R&D, Immunokontact, UK), TNFα (Mab1 and Mab11, mouse IgG1, Pharmingen, San Diego, CA, USA), TNFβ (LTX-22, mouse IgG1, Bender Medsystem, Vienna, Austria), and IFNγ (cocktail of 1-DIK and 7-B6-1, mouse IgG1, MabTech, Stockholm, Sweden). Biotinylated secondary goat anti-mouse IgG1antibody (Caltag Lab., San Francisco, CA, USA) was used at a dilution of 1:300, and immunohistochemical staining was achieved by use of the avidin/peroxidase solution (Vectastain Elite-ABC-Peroxidase, Vector Laboratories, Burlingame, CA, USA) and diaminobenzidine (DAB) substrate for peroxidase (Vector Laboratories). Duplicate fields were stained for each cytokine. Cytokine production was assessed by direct microscopy, and 500–700 cells were counted per field.

The concentrations of IL1β and IL8 in cell culture supernatants were determined by Luminex cytokine multiplex analyses using the Fluorokine MAP kits (R&D, Minneapolis, MN, USA) and the Luminex100 instrument (Luminex Corp., Austin, TX, USA).

Serum samples collected during the septic episode were available from seven patients, three samples from severe sepsis episodes and five from nonsevere sepsis episodes. The sera had been stored frozen at −20°C until analysis. IL1β and IL8 levels were determined by automated chemoluminescence immunoassay (IMMULITE®, DPC, CA, USA). The lower limits of detection were 5 pg/ml for both assays. Sensitivity of the assays were 1.5 and 2.0 pg/ml, respectively.


The Mann–Whitney U-test was used to analyse differences between the severe and nonsevere cohorts. Comparison of matched isolates was performed by the Wilcoxon matched pairs test. A P-value <0.05 was considered as significant.


Cytokine induction profile of viridans streptococci resembles that of LPS

After 24 h of stimulation with cell-free bacterial supernatant or heat-killed bacteria from all 19 viridans streptococcal strains, production of IL1β and IL8 by human PBMC could be detected (Table 2). IL8 induction was seen also after 72 h of stimulation (data not shown). Both bacterial preparations induced cytokine production, but significantly higher levels of IL1β were demonstrated after stimulation with heat-killed bacteria as compared to supernatants (P<0.01) (Table 2).

Table 2 IL1β and IL8 production induced by viridans streptococcal strains isolated from severe and nonsevere cases

Surprisingly, none of the isolates induced TNFα production after 24 h of stimulation (data not shown). Since TNFα production had previously been demonstrated in cell culture supernatants following stimulation with viridans streptococci,9 we assumed that we might have missed the production peak. Therefore, we analysed TNFα production in PBMC stimulated for 4 h with either cell-free bacterial supernatant or heat-killed bacteria, and found high frequencies of TNFα-producing cells, ranging from 1 to 7%, in cultures stimulated with heat-killed bacteria (Table 3). The noted frequencies were in the same range as the positive control LPS, which induced TNFα production in 3% of the cells (data not shown). A majority (90%) of the cell-free viridans supernatants induced production of TNFα, but at significantly lower frequencies than heat-killed bacteria (P<0.01) (Table 3).

Table 3 TNFα production induced by viridans streptococcal strains isolated from severe and nonsevere cases

Viridans streptococcal strains induced no or only very low production of TNFβ or IFNγ, whereas these cytokines were produced at high frequencies (3.5–5.5%) following stimulation with supernatant from a culture of β-haemolytic group A streptococci (data not shown). Similarly, bacterial supernatants and heat-killed bacteria from viridans streptococci did not induce a significant proli-ferative response, whereas culture supernatant from β-haemolytic streptococci group A showed high mitogenic activity (data not shown).

Cytokine-inducing activity of isolates that caused sepsis episodes of varying severity in the same patient

Comparison of the strains isolated from either severe sepsis episodes or nonsevere sepsis showed that there was no significant difference in cytokine-inducing capacity between the two groups (Tables 2 and 3). The material included two cases (2 and 6), who each experienced two episodes, one severe and one nonsevere, of bacteraemia caused by viridans streptococci, following chemotherapy including high-dose cytosine arabinoside (Table 1). The severe episode in both cases included shock and ARDS, whereas only fever and elevated CRP but no signs of organ involvement was seen during the nonsevere episodes. The strains were typed both by biochemical typing and by PFGE to verify that the two episodes in each case were indeed caused by different strains. Matched comparisons of these strains causing infections of starkly varying severity in the same individual revealed that in both cases, the isolates causing severe sepsis induced significantly higher IL1β production as compared to the isolate causing the nonsevere sepsis (P<0.04) (Figure 1). A similar trend was also noted for induction of IL8 (Figure 1).

Figure 1

IL1β and IL8 production induced by viridans streptococcal strains that caused septic episodes of varying severity in the same patient. PBMC from healthy individuals were stimulated with bacterial culture supernatants prepared from the strains causing severe (filled symbols) or nonsevere sepsis (open symbol), respectively, in the same case: (a) case 2 and (b) case 6. Cells were harvested after 24 h of culture and stained for IL1β and IL8 cytokines by immunohistochemistry. The experiments were repeated six times using cells from different individuals. Statistically significant differences between isolates causing severe or nonsevere sepsis were determined by the Wilcoxon matched pairs test, and P-values are indicated in the figure.

Higher IL8 levels in acute phase sera from patients with severe sepsis as compared to nonsevere

Sera collected during the septic episode were available from seven of the patients, three severe and five nonsevere septic episodes, including case 2 who experienced two episodes, one severe and one nonsevere. These samples were analysed for IL1β and IL8 levels by ELISA. In agreement with a previous study,17 IL1β could not be detected in serum from any of these patients. In contrast, IL8 could be detected in all serum samples, and the highest levels were consistently found in sera from severe episodes as compared to nonsevere (Figure 2). Sera were available from the two sepsis episodes of case 2, and markedly higher IL8 levels were found in serum collected during the severe episode as compared to the nonsevere, 4510 (day 4 after onset of sepsis) and 10 (day 2 after onset of sepsis) pg/ml, respectively (Figure 2).

Figure 2

IL8 levels in sera from patients with viridans streptococal bacteraemia. Sera were available from seven cases, including three severe episodes and five nonsevere episodes, as indicated in the figure. The samples were analysed for IL8 levels by ELISA. The sera were collected on days 0–7 from onset of symptoms of infection where day 0 refers to the first day of fever and positive blood cultures. Sera from the severe cases were obtained on days 1, 7, and 4, and sera from nonsevere cases on days 1–4. †10 pg/ml (not shown in the figure).


Viridans streptococcal infection causes different clinical manifestations depending on the host's immune status. In immunocompetent individuals, infections are rare, but when they do occur they usually present with mild clinical symptoms even in bacteraemia and subacute endocarditis, although Streptococcus milleri can cause invasive pyogenic infections.7 In contrast, in neutropenic patients, bacteraemia is frequently seen and septic shock syndrome including ARDS has been described.3,5,18 No cases of viridans streptococcal septic shock syndrome in immunocompetent hosts have been reported, thus, suggesting that host factors must be crucial in the development of septic shock/ARDS. However, in the majority of neutropenic patients, viridans streptococcal bacteraemia resolves without further complications, indicating that bacterial factors may also be relevant. Furthermore, ARDS has not been described in relation to bacteraemia caused by other low-virulent Gram-positive bacteria such as coagulase-negative staphylococci, which are at least as common as viridans streptococci in this patient setting.19

To investigate whether bacterial factors are strong determinants of the severity of sepsis, we identified cohorts of patients who experienced either severe or nonsevere episodes of viridans streptococcal sepsis. The two cohorts were well matched with respect to clinical diagnosis, treatment, and prevalence of mucositis. The latter was found in approximately 50% of the cases in the respective cohort. As expected, the mortality rate was high in the severe cohort (56%), whereas all survived in the nonsevere cohort. The mortality was not related to strain sensitivity to the antibiotic therapy, since the mortality was 33 and 28% in patients receiving ineffective or adequate empiric antibiotic therapy, respectively. This is in agreement with previous studies reporting that the susceptibility of viridans streptococci against empirical antibiotic therapy does not correlate with better prognosis in this patient category.1,4,18,20,21

We chose to test the isolates for induction of IL1β and TNFα since these cytokines' relation to severe sepsis and septic shock is well described:22,23,24 IL8 due to its association with ARDS;11,12,25 and TNFβ and IFNγ as markers for superantigen triggered T-cell activation.26 In order to examine both cell wall components and potential exotoxin activity, we analysed the proinflammatory response induced by heat-killed bacteria as well as supernatants from overnight cultures containing extracellular proteins. The study showed that the strains, both heat-killed and bacterial supernatants, were potent inducers of proinflammatory cytokines. However, the most potent cytokine induction was consistently seen with heat-killed bacteria as compared to bacterial supernatant. The cytokine induction profile of viridans streptococcal strains strongly resembled that previously described for LPS,16 with early TNFα production peaking at 4–6 h, potent IL1β and IL8 after 24 h of stimulation, and lack of or only very low induction of T-cell cytokines. It seems likely that the proinflammatory activity induced by heat-killed bacteria may be mediated by peptidoglycan and lipoteichoic acid, since they have been shown to induce proinflammatory cytokine production.27,28,29 Also cell-free supernatants from overnight cultures of the isolates were potent proinflammatory inducers. The pro-inflammatory inducing molecules in the supernatants may be either shedded cell wall components and/or exotoxins. One report has described an extracellular product of S. mitis that caused selective activation of T cells expressing Vβ2 and Vβ5.1, and hence was suggested as a potential superantigen.30 In this study, we could not find any evidence for secreted factors with superantigenic activity among the 19 tested isolates, since the supernatants did not have mitogenic activity nor did they induce the characteristic T-cell cytokines TNFβ and IFNγ. This is consistent with other reports of lack of mitogenic activity by viridans streptococcal culture supernatants.9

Comparison of strains that caused either severe or nonsevere sepsis did not reveal a difference with respect to proinflammatory activity. However, the patient material contained two cases who had experienced two episodes of sepsis, one severe and one nonsevere, and comparison of these matched isolates from the same patient revealed that the isolates causing the severe sepsis were more potent cytokine inducers than the isolates causing nonsevere sepsis. Furthermore, the finding that serum collected during the severe sepsis episode contained more than 400-fold IL8 levels than serum from the nonsevere sepsis episode further supported this noted difference between the strains.

Most probably, the development of severe sepsis/ARDS due to viridans streptococcal bacteraemia in neutropenic patients is dependent on both host and bacterial factors, and most importantly the interplay between the host and the pathogen. Cell wall components as peptidoglycan and lipotheichoic acid have been shown to activate the innate immunity through activation of Toll-like receptors (TLR) in particular TLR 2, which subsequently induces expression of proinflammatory genes by triggering activation of NF-κB.31,32,33 Thus, variations in expression of these receptors may explain interindividual variations in response to specific microbial infections. In addition, differences in characteristics of cell wall components or exotoxins can lead to variations in the activation of innate immune responses. Interactions between microbes and epithelial and endothelial cell surface receptors may also be important with regard to bacterial as well as host-dependent factors.34 Further studies are required to understand fully the interplay between host and pathogen in viridans streptococcal sepsis and shock, and the results of this study indicate that success of such studies may require matched host and pathogen samples.


  1. 1

    Bochud PY, Eggiman P, Calandra T et al. Bacteremia due to viridans streptococcus in neutropenic patients with cancer: clinical spectrum and risk factors. Clin Infect Dis 1994; 18: 25–31.

  2. 2

    Persson L, Vikerfors T, Sjoberg L et al. Increased incidence of bacteraemia due to viridans streptococci in an unselected population of patients with acute myeloid leukaemia. Scand J Infect Dis 2000; 32: 615–621.

  3. 3

    Bochud PY, Calandra T, Francioli P . Bacteremia due to viridans streptococci in neutropenic patients: a review. Am J Med 1994; 97: 256–264.

  4. 4

    Cohen J, Donnelly JP, Worsley AM et al. Septicaemia caused by viridans streptococci in neutropenic patients with leukaemia. Lancet 1983; 2: 1452–1454.

  5. 5

    Razonable RR, Litzow MR, Khaliq Y et al. Bacteremia due to viridans group Streptococci with diminished susceptibility to Levofloxacin among neutropenic patients receiving levofloxacin prophylaxis. Clin Infect Dis 2002; 34: 1469–1474.

  6. 6

    Belko J, Goldmann DA, Macone A et al. Clinically significant infections with organisms of the Streptococcus milleri group. Pediatr Infect Dis J 2002; 21: 715–723.

  7. 7

    Murray HW, Gross KC, Masur H et al. Serious infections caused by Streptococcus milleri. Am J Med 1978; 64: 759–764.

  8. 8

    Le Roy D, Morand P, Lengacher S et al. Streptococcus mitis cell walls and lipopolysaccharide induce lethality in D-galactosamine-sensitized mice by a tumor necrosis factor-dependent pathway. Infect Immun 1996; 64: 1846–1849.

  9. 9

    Soto A, McWhinney PH, Kibbler CC et al. Cytokine release and mitogenic activity in the viridans streptococcal shock syndrome. Cytokine 1998; 10: 370–376.

  10. 10

    Hanage WP, Cohen J . Stimulation of cytokine release and adhesion molecule expression by products of viridans streptococci. J Infect Dis 2002; 185: 357–367.

  11. 11

    Hack CE, Hart M, van Schijndel RJ et al. Interleukin-8 in sepsis: relation to shock and inflammatory mediators. Infect Immun 1992; 60: 2835–2842.

  12. 12

    Martin TR . Lung cytokines and ARDS: Roger S. Mitchell Lecture. Chest 1999; 116: 2S–8S.

  13. 13

    Kilian M, Mikkelsen L, Henrichsen J . Taxonomic study of viridans streptococci: description of Streptoccoccus gordonii sp. nov. and emended descriptions of Streptococcus sanguis (White and Niven 1946), Streptococcus oralis (Bridge and Sneath 1982), and Streptococcus mitis (Andrewes and Horder 1906). Int J Syst Bacteriol 1989; 39: 471–484.

  14. 14

    Hermans PW, Sluijter M, Hoogenboezem T et al. Comparative study of five different DNA fingerprint techniques for molecular typing of Streptococcus pneumoniae strains. J Clin Microbiol 1995; 33: 1606–1612.

  15. 15

    Norrby-Teglund A, Kaul R, Low DE et al. Plasma from patients with severe invasive group A streptococcal infections treated with normal polyspecific IgG inhibits streptococcal superantigen-induced T cell proliferation and cytokine production. J Immunol 1996; 156: 3057–3064.

  16. 16

    Andersson J, Nagy S, Bjork L et al. Bacterial toxin-induced cytokine production studied at the single-cell level. Immunol Rev 1992; 127: 69–96.

  17. 17

    Engel A, Kern P, Kern WV . Levels of cytokines and cytokine inhibitors in the neutropenic patient with alpha-hemolytic streptococcus shock syndrome. Clin Infect Dis 1996; 23: 785–789.

  18. 18

    Elting LS, Bodey GP, Keefe BH . Septicemia and shock syndrome due to viridans streptococci: a case–control study of predisposing factors. Clin Infect Dis 1992; 14: 1201–1207.

  19. 19

    Sparrelid E, Hagglund H, Remberger M et al. Bacteraemia during the aplastic phase after allogeneic bone marrow transplantation is associated with early death from invasive fungal infection. Bone Marrow Transplant 1998; 22: 795–800.

  20. 20

    Marron A, Carratala J, Gonzalez-Barca E et al. Serious complications of bacteremia caused by Viridans streptococci in neutropenic patients with cancer. Clin Infect Dis 2000; 31: 1126–1130.

  21. 21

    Burden AD, Oppenheim BA, Crowther D et al. Viridans streptococcal bacteraemia in patients with haematological and solid malignancies. Eur J Cancer 1991; 27: 409–411.

  22. 22

    Bone RC . A critical evaluation of new agents for the treatment of sepsis. JAMA 1991; 266: 1686–1691.

  23. 23

    Casey LC, Balk RA, Bone RC . Plasma cytokine and endotoxin levels correlate with survival in patients with the sepsis syndrome. Ann Intern Med 1993; 119: 771–778.

  24. 24

    Cohen J, Carlet J . INTERSEPT: an international, multicenter, placebo-controlled trial of monoclonal antibody to human tumor necrosis factor-alpha in patients with sepsis. International Sepsis Trial Study Group. Crit Care Med 1996; 24: 1431–1440.

  25. 25

    Martin TR . Cytokines and the acute respiratory distress syndrome (ARDS): a question of balance [news; comment]. Nat Med 1997; 3: 272–273.

  26. 26

    Kotb M . Bacterial pyrogenic exotoxins as superantigens. Clin Microbiol Rev 1995; 8: 411–426.

  27. 27

    Heumann D, Barras C, Severin A et al. Gram-positive cell walls stimulate synthesis of tumor necrosis factor alpha and interleukin-6 by human monocytes. Infect Immun 1994; 62: 2715–2721.

  28. 28

    English BK, Patrick CC, Orlicek SL et al. Lipoteichoic acid from viridans streptococci induces the production of tumor necrosis factor and nitric oxide by murine macrophages. J Infect Dis 1996; 174: 1348–1351.

  29. 29

    Standiford TJ, Arenberg DA, Danforth JM et al. Lipoteichoic acid induces secretion of interleukin-8 from human blood monocytes: a cellular and molecular analysis. Infect Immun 1994; 62: 119–125.

  30. 30

    Matsushita K, Fujimaki W, Kato H et al. Immunopathological activities of extracellular products of Streptococcus mitis, particularly a superantigenic fraction. Infect Immun 1995; 63: 785–793.

  31. 31

    Medzhitov R, Janeway C . Innate immunity. N Engl J Med 2000; 343: 338–344.

  32. 32

    Ozinsky A, Underhill DM, Fontenot JD et al. The repertoire for pattern recognition of pathogens by the innate immune system is defined by cooperation between toll-like receptors. Proc Natl Acad Sci USA 2000; 97: 13766–13771.

  33. 33

    Takeuchi O, Kawai T, Muhlradt PF et al. Discrimination of bacterial lipoproteins by Toll-like receptor 6. Int Immunol 2001; 13: 933–940.

  34. 34

    Vernier A, Diab M, Soell M et al. Cytokine production by human epithelial and endothelial cells following exposure to oral viridans streptococci involves lectin interactions between bacteria and cell surface receptors. Infect Immun 1996; 64: 3016–3022.

Download references


This work was supported by grants awarded by the Swedish Foundation for Strategic Research, Children's Cancer Foundation (1996/079 to ES, 2001/012 to MR, and 1999/004 to JA), grants from the Swedish Cancer Foundation (to OR), the Karolinska Institutet (to AN-T) and the Swedish Medical Research Council (012610 to AN-T and 10850 to JA), and Tobias Foundation (to AN-T). This study was conducted in accordance with guidelines for human experimentation as specified by the Ethical committee at Huddinge University Hospital, Stockholm, Sweden.

Author information

Correspondence to A Norrby-Teglund.

Rights and permissions

Reprints and Permissions

About this article


  • viridans streptococci
  • bacteraemia
  • acute respiratory failure syndrome
  • neutropenia
  • cytokines

Further reading