Altered Synthesis of Interleukin-12 and Type 1 and Type 2 Cytokines in Rhesus Macaques during Measles and Atypical Measles

Transcription

1 13 Altered Synthesis of Interleukin-12 and Type 1 and Type 2 Cytokines in Rhesus Macaques during Measles and Atypical Measles Fernando P. Polack, 1,2 Scott J. Hoffman, 1 William J. Moss, 1 and Diane E. Griffin 1 1 W. Harry Feinstone Department of Molecular Microbiology and Immunology, Johns Hopkins Bloomberg School of Hygiene and Public Health, and 2 Department of Pediatrics, Johns Hopkins University School of Medicine, Baltimore, Maryland Immunosuppression during and after measles results in increased susceptibility to other infections and 1 million deaths annually. The mechanism by which measles virus (MV) induces immune suppression is incompletely understood, but a type 2 skewing of the cytokine response after infection has been documented. In vitro studies suggest that lack of interleukin (IL) 12 production by monocytes and dendritic cells plays an early role in the skewed response. In addition, immunization with an inactivated measles vaccine before measles develops appears to lead to an even stronger type 2 skewing of the cytokine response and atypical measles. In this study, the cytokine responses in rhesus macaques were compared after vaccination with live and formalin-inactivated vaccines and after challenge with MV. In vivo production of IL-12 was decreased during the viremic phase of the illness, and production of IL-4 was increased during and after atypical measles, compared with measles. Received 24 May 2001; revised 17 September 2001; electronically published 14 December The studies were approved by the Johns Hopkins University animal care and use committee. Reprints or correspondence: Dr. Fernando P. Polack, W. Harry Feinstone Dept. of Molecular Microbiology and Immunology, Johns Hopkins Bloomberg School of Hygiene and Public Health, 615 N. Wolfe St., W5102, Baltimore, MD (fpolack@jhsph.edu). The Journal of Infectious Diseases 2002;185:13 9 q 2002 by the Infectious Diseases Society of America. All rights reserved /2002/ $02.00 The outcome of infection with measles virus (MV) is largely determined by unique interactions between the virus and the immune system. Measles was the first disease to be recognized to be immunosuppressive, and most deaths associated with measles are due to infection with other pathogens during this phase of immune suppression [1, 2]. In 1907, von Pirquet described the loss of skin test reactivity to tuberculin during and after measles [3]. Subsequently, other abnormalities of lymphocyte function in vitro, such as loss of proliferative responses to mitogens and recall antigens and altered cytokine production, have been recognized [4 7]. In addition to the immunologic abnormalities associated with acute MV infection in MV vaccine naive (unvaccinated) persons, serious complications have been associated with MV infection in subjects immunized earlier with a poorly protective formalin-inactivated, alum-precipitated measles vaccine manufactured in the 1960s (FIMV; Pfizer). Up to 60% of children immunized with FIMV and exposed to MV after protective antibodies had declined developed atypical measles, a severe form of measles characterized by high fever, an acral petechial rash, and pneumonitis [8 10]. In both situations immune suppression in previously unvaccinated persons and atypical measles in those previously immunized with FIMV altered immune responses involving type 2 cytokine skewing have been documented [11, 12]. The immune response during measles in unvaccinated persons involves increases in IgE levels, eosinophilia, and prolonged activation of CD4 T cells producing interleukin (IL) 4 [12]. Atypical measles is associated with even more dramatic eosinophilia, sustained increases in IgE levels, and immune complex mediated vasculitis. The mechanisms by which MV affects the immune response and the influence of earlier priming with FIMV are incompletely understood. The type 2 cytokine pattern during measles is postulated, on the basis of in vitro data, to be due, at least in part, to MV suppression of macrophage and dendritic cell production of IL-12, a cytokine necessary for the differentiation of T cells into Th1 cells [13 16]. Absence of IL-12 early during the immune response to MV would be predicted to skew the T cell response toward type 2 cytokines. Macrophages are a primary target cell for MV in vivo [17]. Recent data from children in The Gambia suggest that IL-12 is suppressed during measles [18]. However, the role of macrophages and the effect of MV on IL-12 production in vivo before and during the entire period of infection, including viremia and induction of immune response, have not been assessed. Using rhesus macaques to study immune responses before and throughout the entire period of infection, including viremia, induction of immune response, rash, and recovery, we investigated the effect of MV infection on cytokine synthesis during the course of disease in macaques with measles and atypical measles. Materials and Methods Immunization and challenge. Samples from year-old rhesus macaques, collected during earlier vaccine studies [11],

2 14 Polack et al. JID 2002;185 (1 January) Figure 1. Interferon (IFN) g levels (mean ^ SEM) in supernatant fluids of phytohemagglutinin-stimulated peripheral blood leukocytes from rhesus macaques after immunization with inactivated and live measles vaccines. White squares, formalin-inactivated measles vaccine (FIMV; n ¼ 3); shaded squares, live attenuated vaccine (LAV; n ¼ 2). were used for these investigations. Five monkeys received 3 doses of FIMV (Pfizer) [11]. Two monkeys received 1 dose of live attenuated MV vaccine (LAV [Attenuvax]; Merck) subcutaneously, and 6 monkeys were naive to MV vaccine (unvaccinated). Challenge was intratracheal, with 10 4 TCID 50 of the Bilthoven wild-type strain of MV administered 2 15 years after immunization. For all manipulations, monkeys were chemically restrained with ketamine (10 15 mg/kg). Cytokine assays. Peripheral blood leukocytes (PBL) were isolated from heparinized blood by gradient centrifugation on Ficoll- Paque (density, 1.077; Amersham Pharmacia). Cells were washed and were suspended at 10 6 cells/ml in RPMI with 10% fetal bovine serum, glutamine, and antibiotics. IL-4 and interferon (IFN) g were measured in supernatant fluids of 10 6 PBL cultured with 4 mg of phytohemagglutinin (PHA-P; Sigma) for 48 h. IL-12 (p70 and p40) and tumor necrosis factor (TNF) a were measured in supernatant fluids of similar cultures stimulated for 24 h with 10 ml of 0.75% Staphylococcus aureus Cowan strain I (SAC; Sigma). Plasma was stored in aliquots at 270 C. IL-12 p70 and p40, IFN-g, and IL-4 levels were also measured in plasma. Cytokine concentrations were determined by EIA, following the manufacturer s instructions (Biosource International). Flow cytometric analysis. Flow cytometric analysis was performed after incubation of 300 ml of whole blood for 18 h with 50 ng of lipopolysaccharide (from Salmonella typhimurinum; Sigma) and 15 U of recombinant human IFN-g (BD PharMingen). Erythrocytes were lysed with PharM Lyse solution (BD PharMingen). Fluorescein isothiocyanate conjugated mouse anti human CD14 monoclonal antibody (MAb), CD64 MAb (clones M5E2 and 10.1; BD PharMingen), or control mouse immunoglobulin (clone MOPC-21; BD PharMingen) was used (data not shown), in addition to side and forward scatter plots, to identify the monocyte and granulocyte populations. Phycoerythrin-conjugated mouse anti human IL-12 p70 and p40, rat anti human IL-10 MAb (clones C11.5 and 9D7; BioSource International), and control mouse or rat immunoglobulin (clones MOPC-21 and R3-34, respectively; BD PharMingen) were used for intracellular cytokine detection. All antibodies were used at saturating concentrations, as defined by the manufacturer. Data for stained cells were acquired on a flow cytometer (FACSCalibur; Becton Dickinson) and were analyzed with CellQuest software (Becton Dickinson Immunocytometry Systems). Statistical analysis. Data were analyzed with statistical software (Statview). Data are expressed as mean ^ SEM. Comparisons were made with the Mann-Whitney U test. Results Cytokine production after FIMV and LAV vaccination. To determine changes in cytokine responses associated with FIMV and LAV immunization in rhesus macaques, we first compared the levels of IL-4 and IFN-g produced by PHA-stimulated PBL taken at various times after immunization (figure 1). IL-4 was undetectable at all time points. IFN-g levels were low in both groups; they rose slightly after immunization with LAV but decreased after each FIMV dose. Because IFN-g production is affected by production of IL-12, we examined vaccine-induced changes in plasma IL-12 levels after immunization with FIMV and LAV. Changes were not significantly different between groups (P ¼ :083). Figure 2. Interleukin (IL) 12 levels in plasma from rhesus macaques after immunization with inactivated and live measles vaccines. White squares, formalin-inactivated measles vaccine (FIMV; n ¼ 5); shaded squares, live attenuated vaccine (LAV; n = 2). Data are mean ^ SEM. *P ¼ :083, Mann-Whitney U test.

3 JID 2002;185 (1 January) IL-12 Suppression in Measles 15 Figure 3. Levels of interleukin (IL) 12 (A) and tumor necrosis factor (TNF) a (B), as measured by EIA, in supernatant fluids of Staphylococcus aureus Cowan strain 1 stimulated peripheral blood leukocytes from unvaccinated and previously immunized rhesus macaques after measles virus challenge. White squares, preimmunized with formalin-inactivated measles vaccine (n ¼ 3); shaded squares, unvaccinated (n ¼ 6); half-shaded squares, preimmunized with live attenuated vaccine (LAV; n ¼ 2). Rules at bottom of panels A and B show period of viremia. Data are mean ^ SEM. *P, :001, Mann-Whitney U test. FIMV immunization was associated with decreases in IL-12 levels after the first and second doses, whereas LAV immunization elicited an increase in plasma IL-12 levels that was maximal 30 days after vaccination (figure 2). IFN-g and IL-4 were undetectable in plasma in both groups. Cytokine production after infection with wild-type MV. To determine the effect of previous priming by FIMV or LAV on the cytokine production by PBL after challenge with wild-type MV, we measured IL-12 in supernatant fluids of SAC-stimulated PBL at various times after infection (figure 3A). Production of IL-12 in unvaccinated monkeys with measles and in FIMVimmunized monkeys with atypical measles decreased beginning 2 days after infection, coincident with the period of viremia, and then rose above baseline levels during the time of the rash, 9 13 days after infection (P, :001 compared with baseline levels for both groups). IL-12 production remained elevated for 2 weeks after clinical disease was apparent and then returned to baseline levels. Macaques previously immunized with LAV were protected from disease after MV challenge [11] and had no evidence of suppression of IL-12. To determine whether IL-12 suppression reflected a generalized failure of monocyte function during the period of viremia, we measured levels of TNF-a in supernatant fluids of PBL stimulated with SAC (figure 3B). Production of TNF-a was suppressed for 15 days after challenge in macaques with measles but increased during this period in macaques with atypical measles. The levels of TNF-a in macaques with atypical measles peaked at the time of rash and remained high for 3 weeks. Changes in TNF-a synthesis similar to those of unvaccinated macaques with measles also occurred in macaques protected by LAV, but the response returned to baseline levels more quickly. To determine whether suppression of IL-12 production by PBL was associated with changes in levels of IL-12 in plasma during the viremic period, we measured plasma IL-12 levels in unvaccinated macaques and in monkeys previously primed with FIMV (figure 4). Plasma IL-12 levels were lower during the period of viremia in both groups, correlating with decreased production by stimulated PBL (P ¼ :006 for measles and P ¼ :049 for atypical measles, compared with baseline levels). Figure 4. Interleukin (IL) 12 levels in plasma from rhesus macaques after measles virus challenge. White squares, preimmunized with formalin-inactivated measles vaccine (macaques with atypical measles; n ¼ 5); shaded squares, unvaccinated (macaques with measles; n ¼ 6). Rule at bottom shows period of viremia. Data are mean ^ SEM. *P ¼ :006; **P ¼ :049, both Mann-Whitney U test.

4 16 Polack et al. JID 2002;185 (1 January) MV viremia was associated with a transient lymphopenia that resolved 2 weeks after challenge (figure 5A). Since MV infects monocytes in vivo [17], we determined whether IL-12 suppression was associated with a monocytopenia by performing monocyte counts in peripheral circulation of monkeys with measles and atypical measles (figure 5B). No significant change in numbers of circulating monocytes occurred in monkeys with measles, but a transient monocytosis that reached its peak at the time of rash was detected in macaques with atypical measles. Recent reports have indicated that neutrophils are an important source of IL-12 in other infectious diseases [19 21]. To determine whether decreased IL-12 production was associated with changes in neutrophil counts, we performed neutrophil counts on peripheral blood (figure 5C). Monkeys with measles exhibited a transient decrease in neutrophil counts during the viremic phase of illness, coinciding with the period of IL-12 suppression, and counts returned to baseline levels during the time of the rash. The number of circulating neutrophils in macaques with atypical measles increased after infection. To examine whether neutrophils could be a source of IL-12 during measles, we stained PBL to identify intracellular IL-12 and analyzed the cells by flow cytometry (figure 6). Both granulocytes and monocytes produced IL-12 when stimulated. To determine the types of cytokines being produced by T cells in monkeys with measles and atypical measles, we compared the levels of IL-4 and IFN-g in supernatant fluids from PHAstimulated PBL (figure 7). PBL from macaques that had received FIMV (and therefore had developed the changes associated with atypical measles) produced higher levels of IL-4 than PBL from unvaccinated monkeys that developed measles (P ¼ :02; day 55 levels), and these levels continued to increase for weeks. Monkeys that developed measles and monkeys protected from measles by previous immunization with LAV had a less pronounced increase in IL-4 production after challenge. IFN-g production was also higher in PBL from monkeys that Figure 5. Changes in absolute leukocyte counts of rhesus macaques after measles virus challenge. Shown are changes in lymphocytes (A), monocytes (B), and neutrophils (C). White squares, preimmunized with formalin-inactivated measles vaccine (macaques with atypical measles; n ¼ 5); shaded squares, unvaccinated (macaques with measles; n ¼ 6); half-shaded squares, preimmunized with live attenuated vaccine (LAV; macaques protected by LAV; n ¼ 2). Rules at bottom of panels A, B, and C show period of viremia. Data are mean ^ SEM.

5 JID 2002;185 (1 January) IL-12 Suppression in Measles 17 Figure 6. Detection of intracellular interleukin (IL) 12 in monocytes and neutrophils from rhesus macaques after measles virus (MV) challenge. A, Dot plot of representative data from unvaccinated macaque (17L) 28 days after MV challenge. FSC, forward scatter; SSC, side scatter. B, Histogram of representative data from the same unvaccinated macaque 28 days after MV challenge. developed the changes associated with atypical measles, compared with those from unvaccinated and LAV-protected macaques, but these differences were not statistically significant (P ¼ :18; day 55 levels). Discussion The immune response is an important determinant of the clinical manifestations of MV infection. The rash is immunologically mediated, type 2 polarization of the cytokine response is postulated to be an important factor in the immune suppression associated with measles, and the severe disease atypical measles is mediated by immune complex formation and eosinophilia. The factors leading to these outcomes are difficult to study in humans, since the earliest time of study is usually at the onset of clinical illness, typically 7 14 days after infection. In this study of rhesus macaques, production of IL-12 was suppressed during the viremic phase of measles, followed by an increase in IL-4 that was greatly exaggerated when the immune system was previously primed with FIMV. Figure 7. Levels of interleukin (IL) 4 (A) and interferon (IFN) g (B), as measured by EIA, in supernatant fluids of phytohemagglutinin-stimulated peripheral blood leukocytes from rhesus macaques after measles virus challenge. White squares, preimmunized with formalin-inactivated measles vaccine (macaques with atypical measles; n ¼ 3); shaded squares, unvaccinated (macaques with measles; n ¼ 6); half-shaded squares, preimmunized with live attenuated vaccine (LAV; macaques protected by LAV; n ¼ 2). Data are mean ^ SEM. *P ¼ :02; **P ¼ :18, both Mann- Whitney U test.

6 18 Polack et al. JID 2002;185 (1 January) Decreased IL-12 levels in plasma and decreased IL-12 production by SAC-stimulated PBL occurred early after infection, primarily during the viremic period. Suppression of IL-12 during the period of viremia supports the in vitro observation that exposure of monocytes to MV suppresses IL-12 production through a CD46-mediated process [13]. However, both monocytes and neutrophils produce IL-12 after infection, and the source of IL-12 in these cultures is not clear. Granulocytes were present in ficoll-hypaque gradient separated PBL, probably because this separation technique does not achieve a pure mononuclear cell population in rhesus macaques [22, 23]. Flow cytometry showed that both neutrophils and monocytes were producing IL-12, and the decrease in neutrophil counts during measles suggests the possibility that the decrease in IL-12 production is due to fewer neutrophils in the PBL cultures. However, monkeys with atypical measles have an even greater decrease in IL-12, whereas neutrophil counts are increasing. It is possible that complement activation and immune complex formation during atypical measles, which are initiated during the viremic period [11], may also contribute to suppression of IL-12 production [13]. This early suppression of IL-12 may explain, in part, the low IFN-g levels observed in macaques with measles and supports a potential role for a lack of IL-12 in subsequent well-documented deficits in cellular immune responses during measles. Responses such as delayed-type hypersensitivity reactions are dependent on production of type 1 cytokines, such as IFN-g. Although the IL-12 suppression in vivo is followed by a rebound above baseline levels after MV has been cleared, no subsequent change is observed in IFN-g levels, which remain low for months after wild-type infection. IFN-g production is decreased in T cells stimulated by anti-cd3 antibodies during measles [12] and is undetectable in plasma in a large proportion of patients with measles [24]. This suggests that IL-12 suppression early during infection commits the cellular immune response during measles to a type 2 phenotype, further downregulating type 1 responses and rendering lymphocytes unable to produce IFN-g in response to IL-12 produced later. Production of TNF-a, another cytokine secreted by monocytes, was increased during MV viremia in macaques with atypical measles, suggesting that MV infection does not cause a generalized suppression of monocyte function. IL-4 levels were elevated for months after atypical measles and correlated with the persistent elevation of IgE levels in these animals reported elsewhere [11]. The greater increase in production of IL-4 in macaques with atypical measles may be associated with increased immune activation in PBL of persons with atypical measles, compared with those of persons with measles [25]. Increased production of IL-4 may also promote the strong anamnestic antibody response observed in FIMVimmunized macaques after MV challenge. High levels of nonprotective antibodies during viremia may promote disease mediated by immune complexes [26]. In summary, this study demonstrates suppression of IL-12 production during the viremic period of MV infection, both in unvaccinated and in FIMV-immunized macaques. This early decrease in IL-12 levels may have longstanding effects in the immune response and may play an important role on both measles-associated immune suppression and enhanced disease by promoting a type 2 polarization of the cytokine response. References 1. Akramuzzaman SM, Cutts FT, Wheeler JG, Hossain MJ. Increased childhood morbidity after measles is short-term in urban Bangladesh. Am J Epidemiol 2000;151: Kipps A, Kaschula ROC. Virus pneumonia following measles: a virological and histological study of autopsy material. S Afr Med J 1976;50: Pirquet CV. Das Verhalten der kutanen Tuberkulin-reaktion wahrend der Masern. Dtsch Med Wochenschr 1908;30: Griffin DE, Ward BJ, Jauregui E, Johnson RT, Vaisberg A. Immune activation in measles. N Engl J Med 1989;320: Burnet FM. Measles as an index of immunological function. Lancet 1968; 2: Ward BJ, Johnson RT, Vaisberg A, Jauregui E, Griffin DE. Cytokine production in vitro and the lymphoproliferative defect of natural measles infection. Clin Immunol Immunopathol 1991;61: Whittle HC, Dossetor J, Oduloju A, Bryceson AD, Greenwood BM. Cellmediated immunity during natural measles infection. J Clin Invest 1978; 62: Carter CH, et al. Serologic response of children to inactivated measles vaccine. JAMA 1962;179: Fulginiti VA, Eller JJ, Downie AW, Kempe CH. Altered reactivity to measles virus: atypical measles in children previously immunized with inactivated measles virus vaccines. JAMA 1967;202: Nader PR, Honwitz MS, Rousseau J. Atypical exanthem following exposure to natural measles: eleven cases in children previously inoculated with killed vaccine. J Pediatr 1968;72: Polack FP, Auwaerter PG, Lee SH, et al. Atypical measles: evidence for disease mediated by immune complex formation and eosinophils in the presence of fusion inhibiting antibodies. Nat Med 1999;5: Griffin DE, Ward BJ. Differential CD4 T cell activation in measles. J Infect Dis 1993;168: Karp CL, Wysocka M, Wahl LM, et al. Mechanism of suppression of cellmediated immunity by measles virus. Science 1996;273: Fugier-Vivier I, Servet-Delprat C, Rivailler P, et al. Measles virus suppresses cell-mediated immunity by interfering with the survival and functions of dendritic and T cells. J Exp Med 1997;186: Grosjean I, Caux C, Bella C, et al. Measles virus infects human dendritic cells and blocks their allostimulatory properties for CD4 + T cells. J Exp Med 1997;186: Schnorr JJ, Xanthakos S, Keikavoussi P, et al. Induction of maturation of human blood dendritic cell precursors by measles virus is associated with immunosuppression. Proc Natl Acad Sci USA 1997;94: Esolen LM, Ward BJ, Moench TR, Griffin DE. Infection of monocytes during measles. J Infect Dis 1993;168: Atabani SF, Byrnes AA, Jaye A, et al. Natural measles causes prolonged suppression of interleukin-12 production. J Infect Dis 2001;184: Romani L, Mencacci A, Cenci E, Spaccapelo R. Neutrophil production of IL-12 and IL-10 in candidiasis and efficacy of IL-12 therapy in neutropenic mice. J Immunol 1997;158: Takeda K, Moore TA, Deng JC, et al. Early recruitment of neutrophils determines subsequent T1/T2 host responses in a murine model of Legionella pneumophila pneumonia. J Immunol 2001;166:

7 JID 2002;185 (1 January) IL-12 Suppression in Measles Bliss SK, Butcher BA, Denkers EY. Rapid recruitment of neutrophils containing prestored IL-12 during microbial infection. J Immunol 2000;165: Budzko D, Madden DL, London WT, Sever JL. Improved separation of rhesus monkey lymphocytes with Percoll. Lab Invest 1985;53: Taniuchi S. Isolation of functional polymorphonuclear leukocytes from rhesus monkeys using discontinuous ficoll-hypaque density gradients. J Med Primatol 1991;20: Griffin DE, Ward BJ, Jauregui E, Johnson RT. Immune activation during measles: interferon-g and neopterin in plasma and cerebrospinal fluid in complicated and uncomplicated disease. J Infect Dis 1990;161: Krause PJ, Cherry JD, Carney JM, Naiditch MJ, O Connor K. Measlesspecific lymphocyte reactivity and serum antibody in subjects with different measles histories. Am J Dis Child 1980;134: Carter PM. Immune complex disease. Ann Rheum Dis 1973;32: