SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

Objective

Patients treated with antibodies to tumor necrosis factor α (TNFα) have an increased susceptibility to intracellular infections. We describe 2 patients with rheumatoid arthritis (RA) who developed Salmonella septicemia during anti-TNF treatment. The aim of this study was to identify the mechanisms involved in the increased susceptibility of anti-TNF–treated patients to intracellular microorganisms.

Methods

We evaluated an additional 6 RA patients receiving anti-TNF antibodies, 5 RA patients not receiving anti-TNF therapy, and 6 age- and sex-matched healthy volunteers. The in vitro production of cytokines (interleukin-1β [IL-1β], IL-6, interferon-γ [IFNγ], and IL-10) upon bacterial stimulation of whole blood and the expression of Toll-like receptor 4 (TLR-4) on dendritic cells from RA patients treated with infliximab, RA patients not treated with infliximab, and healthy controls were compared.

Results

Stimulation with heat-killed Salmonella typhimurium or Candida albicans led to a significantly decreased production of IFNγ, but not to a decreased production of IL-10, IL-β, or IL-6, in anti-TNF–treated RA patients compared with RA patients who were not receiving anti-TNF antibodies and compared with healthy controls. TNF-blocking treatment ex vivo significantly inhibited TLR-4 expression on dendritic cells from RA patients and healthy controls.

Conclusion

Since recognition of microorganisms by TLR-4 and activation of phagocytes by IFNγ are essential mechanisms for the defense against intracellular and fungal pathogens, we propose that this pathway is crucial for the increased susceptibility to these microorganisms in patients receiving anti-TNF therapy.

Anti–tumor necrosis factor (anti-TNF) strategies with either specific monoclonal antibodies or TNF receptors (TNFRs) result in significant clinical benefits in a large group of patients with rheumatoid arthritis (RA) or Crohn's disease, and application of this therapeutic option in other inflammatory diseases is to be expected. An increased incidence of tuberculosis among patients treated with the TNFα-neutralizing antibody infliximab has recently been described (1), and serious systemic infections with fungi, such as Candida albicans, Aspergillus fumigatus, Pneumocystis carinii, or Histoplasma capsulatum, have also been reported (2). In this report, we describe 2 RA patients treated with anti-TNF antibodies who developed sepsis caused by the intracellular gram-negative bacterium Salmonella enterica.

Since most of these pathogens are so-called facultative intracellular organisms, it may be assumed that the defect induced by anti-TNF resides in the cellular immunity, but the precise mechanisms through which anti-TNF treatment renders patients more susceptible to infections have not been elucidated. The Th1 cytokine interferon-γ (INFγ) is crucial for proper activation of phagocytosis and killing of intracellular bacteria, and the production of IFNγ is stimulated by TNF. We therefore hypothesized that TNF neutralization with monoclonal antibodies may result in a decreased production of IFNγ, subsequently leading to a defective cellular immune response. In addition, both TNF and IFNγ are known to potentiate the expression of Toll-like receptor 4 (TLR-4) on the cell membrane (3, 4), and TLR-4 is important for the recognition of Salmonella species (5), Mycobacterium tuberculosis (6), C albicans (7), and A fumigatus (8) by host cells such as dendritic cells and macrophages. We therefore also assessed the impact of TNF blockade on the expression of TLR-4 on dendritic cells.

PATIENTS AND METHODS

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

Patients with Salmonella septicemia.

Patient 1.

The patient, a 62-year-old woman, had refractory RA despite treatment with antirheumatic drugs and low-dose corticosteroids. She had been enrolled in a clinical trial of the fully humanized anti-TNF antibody adalimumab (Abbott Netherlands, Hoofddorp, The Netherlands). Four months after the start of anti-TNF therapy, she was admitted to the hospital because of fever, chills, and severe synovitis of the left knee, following a period of bloody diarrhea. Blood cultures and a knee aspirate yielded S enterica group D. She was treated with 400 mg of ciprofloxacin twice a day, and she recovered from the septicemia and septic arthritis.

Patient 2.

The patient, a 60-year-old man with RA, was admitted to the hospital because of high fever, chills, nausea, and vomiting. He had been treated with infliximab for 3 years, until 6 weeks before admission. Blood cultures yielded S enterica group D, and the patient was treated with intravenous ciprofloxacin, 400 mg twice a day. Imaging studies during admission showed an aneurysm of the abdominal aorta with a diameter of 4.5 cm, which progressed to 8 cm over the following 2 weeks. The patient underwent surgical resection of the aneurysm and placement of an axillobifemoral bypass. Cultures of the resected aneurysm revealed infection with S enterica group D and C albicans. Fluconazole, 400 mg/day, was added to the treatment regimen. Despite clinical improvement and discharge from the hospital, the patient was admitted to the emergency department 3 weeks later because of refractory hypotension, and he died of hypovolemic shock. Postmortem investigation demonstrated the presence of an aortoduodenal fistula as the source of massive gastrointestinal bleeding.

Cytokine stimulation studies.

Venous blood was obtained from 6 RA patients who had been receiving treatment with infliximab for at least 4 months (3 men and 3 women, age range 35–57 years), 5 RA patients who had not been treated with infliximab (3 men and 2 women, age range 31–55 years), and 6 age- and sex-matched healthy volunteers. In addition, blood samples were obtained from patient 1 (index case) on day 4 after admission and 3 weeks after discontinuation of the anti-TNF treatment. Informed consent was obtained from all study subjects.

To investigate cytokine production ex vivo, whole blood was diluted 1:5 with RPMI 1640 in 24-well plates and stimulated with heat-killed (30 minutes at 100°C) Salmonella typhimurium or C albicans microorganisms (107/ml). After incubation for 24 or 48 hours at 37°C, plasma was obtained by centrifugation and stored at −80°C until assayed. Concentrations of interleukin-1β (IL-1β) and IL-6 (after 24 hours of stimulation) and of IFNγ and IL-10 (after 48 hours of stimulation) were measured using commercial enzyme-linked immunosorbent assay kits (Pelikine; Central Laboratory of The Netherlands Red Cross Blood Transfusion Service, Amsterdam, The Netherlands).

Generation of monocyte-derived dendritic cells.

Venous blood was obtained from 4 patients with RA who had not been treated with anti-TNF and from 4 age- and sex-matched healthy volunteers. The peripheral blood mononuclear cell fraction was obtained by density-gradient centrifugation over Ficoll-Hypaque (Amersham Biosciences, Roosendaal, The Netherlands), resuspended in RPMI 1640 (the Dutch modification with 2% heat-inactivated human serum), and allowed to adhere for 1 hour at 37°C in 6-well culture plates (Costar, Badhoeverdorp, The Netherlands). Adherent monocytes were cultured in RPMI 1640 (the Dutch modification supplemented with 10% fetal calf serum [FCS]) and antibiotics/antimycotics in the presence of IL-4 (500 units/ml) and granulocyte–macrophage colony-stimulating factor (GM-CSF; 800 units/ml) for 6 days. On day 6, the immature dendritic cells were harvested.

To generate mature dendritic cells, the cells were transferred to new 6-well culture plates and cultured for 2 days in RPMI 1640 supplemented with 10% FCS, IL-4 (500 units/ml), GM-CSF (800 units/ml), and 2 μg/ml of lipopolysaccharide (LPS; from Escherichia coli) (Sigma, St. Louis, MO).

Role of endogenous TNF in the expression of TLR-4 on dendritic cells.

To assess the effect of blocking TNF on the expression of TLR-4, dendritic cells from RA patients and healthy volunteers were coincubated with 10 μg/ml soluble TNFR (sTNFR) p55 (a kind gift from Dr. Carl K. Edwards, Amgen, Thousand Oaks, CA) during their 6 days of maturation. The membrane expression of TLR-4 on mature dendritic cells was measured by indirect immunofluorescence staining. The first layer was attached after incubation of 1 × 105 monocyte-derived dendritic cells for 30 minutes at 4°C with monoclonal anti-human antibodies against TLR-4 (Santa Cruz Biotechnology, Santa Cruz, CA). After a washing step, the cells were incubated with fluorescein isothiocyanate–conjugated goat anti-mouse IgG for 30 minutes at 4°C in complete darkness. Subsequently, the cells were washed and analyzed with a FACScan flow cytometer (Becton Dickinson, Mountain View, CA).

Cells were gated according to their forward- and side-scatter patterns, and the expression of CD83, the mature dendritic cell marker. For each marker, 104 cells were counted. Mouse IgG2b was used as an isotype control. The same settings were used for all fluorescence-activated cell sorting procedures, and the amount and position of viable cells were checked using propidium iodide staining.

RESULTS

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

Cytokine production.

Upon stimulation of whole blood with S typhimurium or C albicans microorganisms, patients treated with anti-TNF antibodies responded with significantly less IFNγ production than did healthy controls or RA patients who were not receiving infliximab, whereas IL-10 production was not affected (Figures 1A and B). The synthesis of IL-1 and IL-6 did not differ between the anti-TNF–treated patients and controls (results not shown). In patient 1, the index case of Salmonella septicemia treatment with the anti-TNF antibody was discontinued, and the cytokine profile was reassessed after 3 weeks. Although IFNγ production was still suppressed 3 weeks after discontinuation of adalimumab, there was a clear trend toward recovery (Figure 1C).

thumbnail image

Figure 1. Concentrations of interferon-γ (IFNγ) and interleukin-10 (IL-10) in whole blood obtained from 6 healthy volunteers (open bars), 5 rheumatoid arthritis (RA) patients not receiving anti–tumor necrosis factor (anti-TNF) therapy (striped bars), and 6 RA patients treated with infliximab for at least 4 months (solid bars). IFNγ and IL-10 were measured after stimulation with heat-killed Salmonella typhimurium (A) or Candida albicans (B) microorganisms (107/ml) for 48 hours at 37°C. IFNγ production in whole blood obtained from index patient 1 during treatment with anti-TNF antibodies and 3 weeks after discontinuation of treatment was also measured (C), after stimulation with heat-killed S typhimurium (open bars) or C albicans (striped bars) microorganisms (107/ml). Values are the mean and SD concentration of IFNγ and IL-10 (A and B) or the percentage of IFNγ production in blood from healthy volunteers (C). ∗ = P < 0.05 versus healthy volunteers and versus RA patients not receiving infliximab.

Download figure to PowerPoint

TLR-4 expression on dendritic cells.

The percentage of cells positive for TLR-4 expression was only slightly lower in the mature dendritic cell population (79.7%) than in the immature dendritic cell population (88.7%), both in healthy volunteers and in RA patients. However, the addition of sTNFR p55 during the maturation period to cells from patients and healthy volunteers reduced these percentages to 14% and 16%, respectively (P < 0.05) (Figure 2).

thumbnail image

Figure 2. The effect of anti–tumor necrosis factor (anti-TNF) antibodies on Toll-like receptor 4 (TLR-4) expression on mature dendritic cells obtained from rheumatoid arthritis (RA) patients who had not received anti-TNF therapy (n = 4) and from healthy volunteers (n = 4). Cells were treated (striped bars) or were not treated (open bars) with anti-TNF and TLR-4 expression was measured. Values are the mean and SD of 4 experiments. ∗ = P < 0.05 versus cells not treated with anti-TNF.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

We report on 2 patients treated with anti-TNF antibodies who were diagnosed as having Salmonella septicemia. In patient 2, secondary infection of an abdominal aortic aneurysm with C albicans was also diagnosed.

This pattern of infectious complications is strongly suggestive of a defect in the activation of the cellular immune system in these patients. On the one hand, it is to be expected that blocking TNF activity itself would lead to impairment of important mechanisms, such as leukocyte recruitment, activation of phagocytosis and killing by neutrophils and macrophages, and granuloma formation (9, 10). On the other hand, TNF is also crucial for the production of other important cytokines, such as IFNγ. It is known that IFNγ plays a pivotal role in the host defense against both mycobacterial and Salmonella infections (11), as well as against disseminated fungal infections (12). In the present study, we show that treatment of RA patients with anti-TNF monoclonal antibodies severely reduced their production of IFNγ upon challenge with specific stimuli, such as heat-killed Salmonella or C albicans. Discontinuation of anti-TNF therapy in patient 1 resulted in a partial recovery of IFNγ production.

Recently, severe infections by Salmonella and Mycobacterium species have been described in patients with mutations that lead to defective receptors for IL-12 or IFNγ (13), supporting a pivotal role of IFNγ in the defense against these intracellular pathogens. This suggests that IFNγ is crucial for host defense against Salmonella and Mycobacterium species only, whereas mechanisms of defense against other pathogens utilize redundant pathways. In addition, experimental data strongly suggest an important role of IFNγ for the host defense against disseminated fungal infections (12). We therefore propose that inhibition of IFNγ production by anti-TNF antibodies is the central event in the increased susceptibility to infections in general, and that it is most likely central to the development of mycobacterial infection and Salmonella septicemia in particular.

For activation of cellular defense against intracellular pathogens, an adequate balance between Th1 proinflammatory cytokines (such as IFNγ) and Th2 antiinflammatory cytokines (such as IL-10) is essential. Whereas production of the proinflammatory cytokine IFNγ was impaired in the patients treated with anti-TNF antibodies, production of the antiinflammatory cytokine IL-10 remained unaffected, resulting in an imbalance toward a Th2 cytokine profile. We hypothesize that this imbalance is even more important for the increased susceptibility to infections than is the absolute deficiency in the production of IFNγ.

A central event in the defense against bacteria is the recognition of pathogen-associated molecular patterns by pattern recognition receptors. TLRs are the most important class of pattern recognition receptors, and TLR-4 recognizes LPS from gram-negative bacteria (including Salmonella species) (5), components of M tuberculosis (6), C albicans (7), and A fumigatus (8). In critically ill patients, TLR-4 mutations resulting in hyporesponsiveness to LPS are associated with an increased incidence of gram-negative infections (14, 15). It has been recently shown that the expression of TLR-4 is up-regulated by both IFNγ and TNF (4), and we demonstrate here that anti-TNF antibodies can significantly reduce the percentage of TLR-4–positive dendritic cells. Since dendritic cells are crucial for the recognition of bacterial pathogens and antigen presentation to T cells, we hypothesize that down-modulation of TLR-4 expression by anti-TNF treatment may represent an important mechanism of increased susceptibility to infection.

In conclusion, Salmonella infections may represent a severe complication of anti-TNF treatment. Since recognition of microorganisms by TLR-4 and activation of phagocytes by IFNγ are crucial mechanisms for the defense against intracellular and fungal pathogens, we propose that this pathway is crucial for the increased susceptibility to these microorganisms in anti-TNF–treated patients.

REFERENCES

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES
  • 1
    Keane J, Gershon S, Wise RP, Mirabile-Levens E, Kasznica J, Schwieterman WD, et al. Tuberculosis associated with infliximab, a tumor necrosis factor alpha-neutralizing agent. N Engl J Med 2001; 345: 1098104.
  • 2
    Keenan GF, Schaible TF, Boscia JA. Reply to Invasive pulmonary aspergillosis associated with infliximab therapy. N Engl J Med 2001; 344: 1099100.
  • 3
    Bosisio D, Polentarutti N, Sironi M, Bernasconi S, Miyake K, Webb GR, et al. Stimulation of toll-like receptor 4 expression in human mononuclear phagocytes by interferon-gamma: a molecular basis for priming and synergism with bacterial lipopolysaccharide. Blood 2002; 99: 342731.
  • 4
    Wolfs TGAM, Buurman WA, van Schadewijk A, de Vries B, Daemen MARC, Hiemstra PS, et al. In vivo expression of toll like receptor 2 and 4 by renal epithelial cells: IFNgamma and TNF mediated up-regulation during inflammation. J Immunol 2002; 168: 128693.
  • 5
    Tapping RI, Akashi S, Miyake K, Godowski PJ, Tobias PS. Toll-like receptor 4, but not toll-like receptor 2, is a signaling receptor for Escherichia and Salmonella lipopolysaccharides. J Immunol 2000; 165: 57807.
  • 6
    Means TK, Wang S, Lien E, Yoshimura A, Golenbock DT, Fenton MJ. Human toll-like receptors mediate cellular activation by Mycobacterium tuberculosis. J Immunol 1999; 163: 39207.
  • 7
    Netea MG, de Graaf C, Vonk A, Verschueren I, Van der Meer JWM, Kullberg BJ. The role of toll-like receptors in the defense against disseminated candidiasis. J Infect Dis 2002; 185: 14839.
  • 8
    Wang JE, Warris A, Ellingsen EA, Jorgensen PF, Flo TH, Espevik T, et al. Involvement of CD14 and toll-like receptors in activation of human monocytes by Aspergillus fumigatus hyphae. Infect Immun 2001; 69: 24026.
  • 9
    Flynn JL, Goldstein MM, Chan J, Triebold KJ, Pfeffer K, Lowenstein CJ, et al. Tumor necrosis factor is required in the protective immune response against Mycobacterium tuberculosis in mice. Immunity 1995; 2: 56172.
  • 10
    Senaldi G, Yin S. Corynebacterium parvum and Mycobacterium bovis bacillus Calmette-Guerin-induced granuloma formation is inhibited in TNF receptor I (TNF-RI) knock-out mice and by treatment with soluble TNF-RI. J Immunol 1996; 157: 50226.
  • 11
    Mastroeni P, Villarreal-Ramos B, Hormaeche CE. Role of T cells, TNF and IFNgamma in recall of immunity to oral challenge with virulent salmonellae in mice vaccinated with live attenuated aro-Salmonella vaccines. Microb Pathog 1992; 13: 47791.
  • 12
    Kullberg BJ, Van't Wout JW, Hoogstraten C, Van Furth R. Recombinant interferon-γ enhances resistance to acute disseminated Candida albicans infection in mice. J Infect Dis 1993; 168: 43643.
  • 13
    Ottenhoff TH, Verreck FA, Lichtenauer-Kaligis EG, Hoeve MA, Sanal O, van Dissel JT. Genetics, cytokines and human infectious disease: lessons from weakly pathogenic mycobacteria and salmonellae. Nat Genet 2002; 32: 97105.
  • 14
    Agnese DM, Calvano JE, Hahm SJ, Coyle SM, Corbett SA, Calvano SE, et al. Human toll-like receptor 4 mutations but not CD14 polymorphisms are associated with an increased risk of gram-negative infections. J Infect Dis 2002; 186: 15225.
  • 15
    Lorenz E, Mira JP, Frees KL, Schwartz DA. Relevance of mutations in the TLR4 receptor in patients with gram-negative septic shock. Arch Intern Med 2002; 162: 102832.