Cellular Immunity and Active Human Cytomegalovirus Infection in ...

5 downloads 0 Views 246KB Size Report
Oct 1, 2007 - ... Manfred Weiss,2 Heide Suger-Wiedeck,2 Marion Schneider,2 ..... Barry SM, Johnson MA, Janossy G. Cytopathology or immunopath- ology?
MAJOR ARTICLE

Cellular Immunity and Active Human Cytomegalovirus Infection in Patients with Septic Shock Lutz von Mu¨ller,1,a Anke Klemm,1,2 Nilgu¨n Durmus,1 Manfred Weiss,2 Heide Suger-Wiedeck,2 Marion Schneider,2 Walter Hampl,1 and Thomas Mertens1 1

Institute of Virology and 2Department of Anesthesiology, University Hospital Ulm, Ulm, Germany

(See the editorial commentary by Cook, on pages 1273–5.)

Background. Human cytomegalovirus (CMV) is an important opportunistic pathogen after transplantations. In the present study, monitoring of CMV in patients with septic shock was used to discover whether T helper cell type 1 (Th1) cell and natural killer (NK) cell functions interact with CMV reactivation in patients not undergoing immunosuppressive therapy. Methods. Thirty-eight patients with septic shock were monitored, and the 23 CMV-seropositive patients were included in this prospective study. Results. Seven patients (30.4%) developed an active CMV infection despite the detection of CMV-reactive Th1 cells. After active CMV infection, the frequency of CMV-reactive Th1 cells increased from a median of 0.52% to 5.04% (P p .009). Active CMV infections were terminated without antiviral therapy within 2 weeks. In parallel, the frequency of staphylococcal enterotoxin B (SEB; superantigen)–reactive Th1 cells increased from a median of 1.11% to 8.48% (P p .027). In patients without active CMV infection, the frequency of CMV-reactive (median, 0.39%) and SEB-reactive (median, 1.11%) Th1 cells did not increase. Cytotoxic NK cell activity was persistently suppressed despite the presence of CD56+CD16+ NK cells. Moreover, interleukin-2 application in vitro did not restore NK cell activity. Conclusions. A proinflammatory immune response may contribute to CMV reactivation in patients with septic shock. Adaptive T cell immunity, more likely than NK cell immunity, may contribute to termination of active CMV infection without antiviral therapy in these patients. Cytomegalovirus (CMV) is a human betaherpesvirus with a seroprevalence of ∼80% in German adults [1]. Usually, primary CMV infection is asymptomatic and is followed by lifelong CMV latency. The history of previous CMV infection is demonstrated by the detection of CMV-specific IgG antibodies and of CMV-spe-

Received 10 January 2007; accepted 17 April 2007; electronically published 1 October 2007. Potential conflicts of interest: none reported. Presented in part: Jahrestagung der Gesellschaft fu¨r Virologie 2006, Munich, 15–18 March 2006 (abstract 233). Financial support: Deutsche Forschungsgemeinschaft, Sonderforschungsbereich 451; Roche Diagnostics. a Present affiliation: Institute of Medical Microbiology and Hygiene, University of Saarland Hospital, Homburg/Saar, Germany. Reprints or correspondence: Dr. Lutz von Mu¨ller, Institute of Medical Microbiology and Hygiene, University of Saarland Hospital, Kirrberger Str., Bldg. 43, 66421 Homburg/Saar, Germany ([email protected]). The Journal of Infectious Diseases 2007; 196:1288–95  2007 by the Infectious Diseases Society of America. All rights reserved. 0022-1899/2007/19609-0006$15.00 DOI: 10.1086/522429

1288 • JID 2007:196 (1 November) • von Mu¨ller et al.

cific memory T cells [2]. Innate NK cell immunity counteracts active CMV infection after hemopoietic stem cell transplantation in vivo and in vitro [3–5]. However, the importance of NK cell immunity to CMV infections in patients not undergoing immunosuppressive therapy is not clear. Active CMV infection with high viral load is a wellknown opportunistic infection in immunocompromised patients (e.g., after solid organ and hemopoietic stem cell transplantations) [1, 6–8]. However, CMV reactivation also occurs in patients with sepsis and septic shock who are not undergoing immunosuppressive therapy [9–11]. We recently showed that one-third of patients with septic shock developed an active CMV infection with a low viral load as detected on the basis of quantitative pp65 antigenemia [12]. In contrast to the situation in patients after transplantation, active CMV infection was terminated without antiviral therapy in these patients. Thus, monitoring of patients with septic shock provides a unique model to investigate the

natural course of CMV reactivation and of antiviral immunity in patients not undergoing immunosuppressive therapy. Sepsis is characterized by clinical symptoms [13], by a severe inflammatory response syndrome (SIRS), and by increased levels of proinflammatory cytokines (e.g., tumor necrosis factor– a and interleukin [IL]–6), arachidonic acid derivates (e.g., prostaglandine E1 and thromboxane), and chemokines (e.g., IL-8 and monocyte chemotactic protein–1) [14] as reflected in the PIRO (predisposition/infection/response/organ dysfunction) staging system [15]. The hyperreactive inflammatory phase of sepsis can be followed by an anti-inflammatory phase (compensatory anti-inflammatory response syndrome [CARS]) [16], which is characterized by lymphocyte depletion and leukocyte dysfunction [17].

The present prospective study was performed to monitor the natural course of active CMV infection and to clarify the roles played by innate and acquired cellular immunity in patients with septic shock who did not receive immunosuppressive therapy. PATIENTS, MATERIALS, AND METHODS Patients and study design. The patients with septic shock in the anesthesiological intensive care unit (ICU) of the University Hospital Ulm were monitored for a period of 9 consecutive months by use of a prospective observational study design. The CMV-seropositive patients who were not undergoing immunosuppressive therapy and who had an ICU stay of at least 7

Table 1. Characteristics of patients with and those without active human cytomegalovirus (CMV) infection. Active CMV infection Characteristic, parameter Patients, no. Sex, male/female, no. (%) Age, years Primary disease Abdominal surgery Abdominal tumor Pancreatitis Trauma Infection Bacteremia, no. (%) Gram negative, no. (%) Gram positive, no. (%) Candidemia, no. (%) Bacterial peritonitis, no. (%) Candida peritonitis, no. (%) Laboratory parameters Lactate level, mmol/L WBC count, 109 cells/L C-reactive protein level, mg/L Serum creatinine level, mmol/L AST level, U/L Serum bilirubin level, mmol/L Severity of disease SOFA scorec d ICU stay, days Mechanical ventilation, daysd Mortality rate, no. (%) Cortisone, no. (%) Catecholamines, days

With (n p 7)

Without (n p 16)

P

7 5 (71)/2 (29) 68 (40–78)

16 9 (56)/7 (44) 60 (44–78)

NSa b NS

2 0 3 2

7 3 1 5

ND ND ND ND

(59) (19) (38) (6) (50) (31)

NS NSa a NS NSa a NS

3 1 4 1 5 3

(42) (14) (57) (14) (71) (43)

10 3 6 1 8 5

a

b

2.5 (1.9–3.7) 18.8 (8.1–38.4) 241 (20.3–320) 205 (100–304) 23 (6–42) 25 (6–279)

2.8 (1.2–14.4) 22.2 (7–33) 182 (85–396) 136 (69–307) 48 (9–120) 51 (10–253)

NS b NS b NS b NS NSb b NS

10 (7–13) 54 (16–87) 42 (15–80) 4 (57) 5 (71) 7 (4–41)

10 (7–16) 19 (10–42) 16 (6–38) 6 (38) 10 (63) 7 (1–35)

NSa b .0025 .0025b a NS a NS NSa

NOTE. Data are median (range) values, unless otherwise specified. For the laboratory parameters, the highest values during the observation periods are given. AST, aspartate aminotransferase; ICU, intensive care unit; ND, not done; NS, not significant (P 1 .05); SOFA, sepsis-related organ failure assessment; WBC, white blood cell. a b c d

Fisher’s exact test. Mann-Whitney U test. Highest values. After onset of septic shock.

CMV, Cellular Immunity, and Septic Shock • JID 2007:196 (1 November) • 1289

Table 2. Detection of human cytomegalovirus (CMV)–reactive and staphylococcal enterotoxin B (SEB)–reactive Th1 cells in patients with and those without active CMV infection. Active CMV infection

Category, parameter

With (n p 7)

Without (n p 16)

P

a

Patients with CMV-reactive Th1 cells ⭓0.3% Initial evaluation

5 (71)

6 (38)

NS

7 (100)

7 (44)

.0189

Initial evaluation

7 (100)

11 (69)

NS

Following examinations

7 (100)

14 (88)

NS

Following examinations Patients with SEB-reactive Th1 cells ⭓0.3%

NOTE. Data are no. (%) of patients, unless otherwise indicated. NS, not significant (P 1 .05); a

Fisher’s exact test.

days were included. Cytotoxic NK cell activity, pp65 antigenemia, and the frequencies of CMV-reactive and staphylococcal enterotoxin B (SEB)–reactive Th1 cells were monitored twice during the first week and once during the following weeks until discharge from the ICU. Sepsis and septic shock was diagnosed on the basis of the criteria of the International Sepsis Definitions Conference [15]. Antiviral therapy was not administered to any patient. The study was approved by the local ethics committee and was conducted in accordance with the Helsinki Declaration. Written informed consent was provided by the patients. Virological testing. Anti-CMV IgG antibodies were analyzed by ELISA (Medac); pp65 antigenemia was determined in EDTA-anticoagulated blood, as described elsewhere [18]. In short, blood cells were isolated by dextrane sedimentation (1% dextrane in PBS), and pp65 antigen–positive cells were detected in duplicates of 5 ⫻ 10 5 blood cells by use of 2 anti-pp65 monoclonal antibodies (20:1; Virion, Argene Biosoft, and Viva Diagnostika) and immunofluorescence microscopy. Frequency of CMV-reactive and SEB-reactive Th1 cells. For lymphocyte stimulation, CMV and human foreskin fibroblast (HFF) antigens were produced using glycine extraction [19]. In short, CMV-infected (AD169 strain) and uninfected HFFs were harvested, washed in 0.15 mol/L glycin PBS buffer, suspended in 0.5 mol/L glycin PBS buffer, chilled on ice, and sonified at 200 W (Branson Sonifier). After 3 rounds of freezing and thawing, cell lysates were incubated overnight at 4C (glycine extraction). Supernatants were harvested after centrifugation (500 g for 30 min), and aliquots were frozen (⫺80C). The frequency of CMV-reactive and SEB-reactive Th1 cells (CD4+CD69+interferon [IFN]-g+) was determined using multiparameter flow cytometry [2, 20]. Na-heparinized whole blood (450 mL) was stimulated with 50 mL of HFF antigen (negative control antigen), CMV antigen, and SEB (1 mg/mL; Sigma) in the presence of costimulatory CD28 antibody (0.5 mg; Becton Dickinson). Two hours after stimulation (37C), the 1290 • JID 2007:196 (1 November) • von Mu¨ller et al.

transgolgi transport was inhibited (brefeldin at 10 mg/mL; Sigma), and, after an additional 4 h, leukocyte activation was stopped using EDTA (20 mmol/L). Erythrocyte lysis and leukocyte fixation was performed using FACS Lysing Solution (Becton Dickinson). After permeabilization (FACS Permeabilizing Solution; Becton Dickinson), the reactive cells were detected by multiparameter flow cytometry (FACSCalibur and CellQuest Pro 4.0.2 software; Becton Dickinson) using fluorochrome-labeled monoclonal antibodies (anti-CD4–Cy5.5, antiCD69–phycoerythrin, anti–IFN-g–fluorescein isothiocyanate [FITC]; Becton Dickinson), and the net response (experimental minus background signal) was analyzed. In healthy volunteers, the frequency of CMV-reactive CD4+CD69+IFN-g+ cells was ⭓0.3%, with an average frequency of 2% (as described in the literature [20]). Because of the lack of CMV-specific memory T cells, the CMV-seronegative probands are found without CMVreactive Th1 cells, reflecting the assay’s specificity. Cytotoxic NK cell assay. Cytotoxic NK cell activity against K562 target cells was assessed using a nonradioactive 3-h cytotoxic assay [21]. In short, K562 target cells were labeled with a permanent green fluorescence dye (10 ng/mL DiOC18 for 15 min at 37C; Molecular Probes). Peripheral blood mononuclear cells from the patients and healthy volunteers (positive control) were freshly isolated by ficoll-hypaque gradient centrifugation, grown in complete medium (RPMI 1640 supplemented with 2 mmol/L glutamine, 0.1 mg/mL penicillin, 0.1 mg/mL streptomycin, 1% nonessential amino acids, and 10% heat-inactivated fetal calf serum [Gibco]), and coincubated with the green target cells (K562) at various effector to target cell (E:T) ratios (50:1, 25:1, 12.5:1, 6:1, 3:1, and target cells only). One additional sample supplemented with IL-2 (1000 U/mL; Chiron) was used as well (E:T ratio, 25:1). The frequency of actively killed K562 target cells was determined by propidium iodide uptake (3 mmol/L; Sigma) after 3 h of coincubation (37C) by flow cytometry (FACSCalibur). Cytotoxic NK cell activity was considered to be positive if 110% of target cells were killed at an E:T ratio of 50:1. After IL-2 application (E:T ratio, 25:1), cytotoxic activity increased 110% in healthy volunteers [22]. The frequency of CD56+CD16+ NK cells was determined by flow cytometry using FITC-conjugated anti-CD56/CD16 antibodies (Becton Dickinson). Statistics. Statistical analysis was performed by Fisher’s exact test, Mann-Whitney U test, and paired t test, using GraphPad Prism software (GraphPad Software). Differences were considered to be significant at P ⭐ .05. RESULTS Patients. Thirty-eight consecutive patients with septic shock were eligible during the observation period. CMV-seronegative patients (n p 6), patients undergoing immunosuppressive therapy (n p 1), and patients who died or were discharged

Figure 1. Human cytomegalovirus (CMV)–reactive and staphylococcal enterotoxin B (SEB)–reactive Th1 cells in patients with and those without active CMV infection. The frequency of CMV-reactive (A) and SEB-reactive (B) Th1 cells was monitored in groups with (gray bars) and without (white bars) active CMV infection. At the initial evaluation (time 0), the frequency of CMV-reactive and SEB-reactive Th1 cells was not different between the groups. In the following examinations, the frequency of CMV-reactive and SEB-reactive Th1 cells increased in the group with active CMV infection but not in the group without active CMV infection. Statistical evaluation was performed using the Mann-Whitney U test. Differences were considered to be significant at P ⭐ .05. P values for groups with significant differences are given above the bars.

within 7 days after the onset of septic shock (n p 6) were excluded. Two additional patients could not be included because samples were missing. Patient characteristics, primary diseases, serum lactate levels, white blood cell counts, C-reactive protein levels, serum creatinine levels, aspartate aminotransferase (AST) levels, serum bilirubin levels, bacterial and fungal infections, treatment with hydrocortisone (200 mg/day) and catecholamines, severity of multiple organ dysfunction (sepsisrelated organ failure assessment [SOFA] score) [23], and mortality rates were not different between the patients with and those without active CMV infection (table 1). However, time of mechanical ventilation and stay in the ICU was markedly

prolonged in the group with active CMV infection, as has been recently described [12]. Active CMV infections in patients with septic shock. Active CMV infection with pp65 antigenemia developed in 7 (30.4%) of 23 patients and was detected for the first time during the initial 2 weeks of septic shock (median, 7 days; range, 0–14 days). The quantitative values of pp65 antigenemia remained low in all cases (median, 3 positive/5 ⫻ 10 5 blood cells; range, 1–17), and active CMV infection stopped without antiviral therapy within 3–4 weeks (range, 1–8 weeks) after the onset of active CMV infection. Qualitative results from CMV-reactive and SEB-reactive Th1 cells. At the first evaluation, CMV-reactive Th1 cells could be detected in 5 patients with (71%) and 6 patients without (38%) active CMV infection (table 2). In the following samples, CMV-reactive Th1 cells were detected in all patients with active CMV infection (100%) but in only 7 patients without active CMV infection (44%) (P p .0189). Similarly, SEBreactive Th1 cells could be detected at the beginning of septic shock in all patients with active CMV infection (100%) but in only 11 patients without active CMV infection (69%). Also, in the following examinations, 2 patients in the group without active CMV infection (22%) remained without SEB-reactive Th1 cells. As expected, the 6 CMV-seronegative patients analyzed at the initial evaluation did not have CMV-reactive Th1 cells. Frequency of CMV-reactive and SEB-reactive Th1 cells. At the beginning of septic shock, the frequency of CMV-reactive Th1 cells was not different between the group with (median, 0.52%; range, 0.02%–3.45%) and the group without (median, 0.08%; range, 0%–2.86%) active CMV infection (figure 1; P 1 .05). For the comparison of values at the beginning of septic shock (time 0) with the highest values for CMV-reactive Th1 cells during follow-up, the frequency of CMV-reactive Th1 cells increased in patients with (median 5.04%; range 1.01%–7.82%; P p .0086) (figure 2A) but not in patients without (median, 0.39%; range, 0%–4.47%) active CMV infection. Significant differences between patients with and those without active CMV infection were demonstrated 7 days after the beginning of septic shock (P p .0046). At the beginning of septic shock, the frequency of SEBreactive Th1 cells was not different between the group with (median, 1.11%; range, 0.02%–8.48%) and the group without (median, 0.81%; range, 0%–3.63%) active CMV infection (figure 1). Later on, the frequency of SEB-reactive Th1 cells increased markedly in patients with active CMV infection (median, 4.5%; range, 3.31%–8.48%; P p .027) (figure 2D); however, in patients without active CMV infection, the frequency remained low (median, 1.11%; range, 0.02%–8.33%). Significant differences between the group with and the group

CMV, Cellular Immunity, and Septic Shock • JID 2007:196 (1 November) • 1291

Figure 2. Human cytomegalovirus (CMV)–reactive and staphylococcal enterotoxin B (SEB)–reactive Th1 cells in patients with and those without a CMV-reactive Th1 cell response. The following scenarios were analyzed: patients with active CMV infection (group 1; A and D), patients negative for pp65 antigenemia and without CMV-reactive Th1 cells (group 2; B and E), and patients negative for pp65 antigenemia but with CMV-reactive Th1 cells (group 3; C and F). In patients with active CMV infection, the frequencies of CMV-reactive and SEB-reactive Th1 cells significantly increased during the observation period. In patients negative for pp65 antigenemia and without CMV-reactive Th1 cells, the frequency of SEB-reactive Th1 cells was persistently low, except for 1 patient. In patients negative for pp65 antignemia but with CMV-reactive Th1 cells, the frequency of SEB-reactive Th1 cells at the beginning of septic shock (time 0) was significantly higher than that in group 2 (P p .012 ). During the observation period, the frequency of CMV-reactive Th1 cells increased (P p .027 ), and the frequency of SEB-reactive Th1 cells tended to increase (but without significance; P p .069). Statistical evaluation was performed using the paired t test and the Mann-Whitney U test, respectively. Differences were considered to be significant at P ⭐ .05.

without active CMV infection appeared for the first time 3 days after the onset of septic shock. CMV-reactive Th1 cells and influence on clinical signs and symptoms. Analyzing the group with and the group without CMV-reactive Th1 cells, we found 3 different scenarios (figure 2): patients with active CMV infection (pp65 antigen positive; group 1), patients negative for pp65 antigenemia and without CMV-reactive Th1 cells (group 2), and patients negative for pp65 antigenemia but with CMV-reactive Th1 cells. No patients with active CMV infection and without CMV-reactive Th1 cells were found. In patients negative for pp65 antigenemia but without CMVreactive Th1 cells, stimulation with SEB also showed a more general suppression of T cell reactivity (figure 2E). In the group with CMV-reactive cells, the frequency of SEB-reactive Th1 cells was higher at the beginning of septic shock (time 0) (P p .012) and increased further over time (highest value); however, the significance level was not achieved (P p .069). Age was significantly increased in pp65 antigen–negative patients with1292 • JID 2007:196 (1 November) • von Mu¨ller et al.

out CMV-reactive Th1 cells (median, 74 years; range, 49–79 years), compared with that in the group with CMV-reactive Th1 cells (median, 56 years; range, 44–75 years). Other clinical signs and symptoms of the severity of disease (median SOFA score, 10 vs. 10), ICU stay (median, 17 vs. 20 days), duration of mechanical ventilation (median, 16 vs. 18 days), days with catecholamine use (median, 7 vs. 8 days), and mortality rate (67% vs. 57%) were not different between pp65 antigen–negative patients with and those without CMV-reactive Th1 cells. NK cells and NK cell function. Although CD56+CD16+ lymphocytes were present in each sample (median, 23% of the lymphocytes; range, 11%–88% of the lymphocytes), cytotoxic NK cell activity was suppressed to !10% (E:T ratio, 50:1) in all except 1 patient throughout the complete observation period (figure 3). Also, application of IL-2 (1000 U/mL) in vitro did not increase cytotoxic NK cell activity. Even using purified NK cells (positive selection with immunomagnetic beads; Miltenyi Biotec), cytotoxic NK cell activity of patients with septic shock remained !10% (data not shown). The one patient with var-

Figure 3. Cytotoxic NK cell activity. NK cell function detected by the nonradioactive NK cell assay was demonstrated in healthy volunteers by use of increasing effector to target cell (E:T) ratios (6.25:1 to 50:1) (A). Cytotoxic NK cell activity at an E:T ratio of 50:1 was 110% in all probands. Cytotoxic NK cell activity was regularly monitored in patients with septic shock and was persistently suppressed to !10% in all except 1 patient (B). Gray bars show patients with active human cytomegalovirus (CMV) infection, and white bars show patients without active CMV infection. Statistical evaluation was performed using the Mann-Whitney U test. Differences were considered to be significant at P ⭐ .05.

iable NK cell activity did not differ from the patients without NK cell activity with respect to patient characteristics and treatment. DISCUSSION Active CMV infection is a common cause of morbidity after hemopoietic and solid organ transplantations [1]. CMV can be reactivated in critically ill patients not undergoing immunosuppressive therapy [7, 10], as recently confirmed by us by analyzing patients with septic shock [12]. Using prospective monitoring of patients with septic shock, we are now capable of investigating the natural course of human CMV reactivation and of the antiviral immune response in patients not undergoing immunosuppressive therapy. This is of outstanding importance, because models for human CMV reactivation in the absence of immunosuppressive therapy do not exist. Also, animal models are not available, because of the host restriction of human CMV. The early onset of active CMV infection during the first 2 weeks implies that CMV reactivation was stimulated at the beginning of septic shock [12]. The mechanisms of CMV reactivation are still not clear, but inflammation (SIRS), antiinflammation (CARS), drugs, and still-unknown factors associated with sepsis and septic shock are supposed to be involved [16, 24, 25]. Generally, it has been assumed that suppression of antiviral immunity predisposes for CMV reactivation [1]. However, the present study has clearly demonstrated that active CMV infections developed in patients with septic shock despite functionally active CMV-reactive Th1 cells (table 2). We suppose that lymphocyte reactivity, such as production and secretion of lymphocyte-derived proinflammatory cytokines, contributes to CMV reactivation. This has previously been shown in the mouse model of CMV reactivation [26]. Also, in the present

study, human CMV reactivation occurred in patients with reactive T cells only. After active CMV infection, the frequency of CMV-reactive Th1 cells profoundly increased (figures 1 and 2), which might have stopped active CMV infections in the absence of antiviral therapy. In this pilot study, the analysis of CMV-specific T cells was focused on the CD4+ T cell response because of the universality of the assay. Also, the more routinely used CMVspecific CD8+ T cell assays (e.g., tetramers) are HLA restricted, requiring HLA typing and exclusion of patients with less frequent HLA types (150% of the patients). However, the function of specific T helper (CD4+) cells and that of cytotoxic (CD8+) T cells are closely correlated, which has been shown previously for patients after transplantations [27, 28]. More detailed analysis of the CD8+ T cell response is planned for future studies. CMV-reactive T cells in patients with active CMV infection may have been increased because of a recovery from T cell anergy, modulation of regulatory T cells, or clonal expansion of activated T cells (which remains to be investigated in consecutive studies). Although CMV-reactive Th1 cells could not prevent CMV reactivation in patients with septic shock, the expansion of CMV-reactive Th1 cells was associated with termination of pp65 antigenemia within a few weeks without antiviral therapy. Surprisingly, the frequency of SEB-reactive Th1 cells increased in the group with active CMV infection and did not increase in the groups without active CMV infection (figures 1 and 2), although superantigen stimulation acts more generally, without restriction to CMV-reactive cells. We hypothesize that a more general lymphocyte activation could be required to stimulate CMV reactivation [29]. Thus, the more general activation of the immune system during CMV reactivation could contribute to antimicrobial immunity on the one hand and to inflammation and immunopathological phenomena

CMV, Cellular Immunity, and Septic Shock • JID 2007:196 (1 November) • 1293

with pulmonary dysfunction on the other [8], as observed clinically in patients with septic shock [12, 30]. The role played by CMV-reactive Th1 cell function was also investigated in subgroups of patients negative for pp65 antigenemia. In pp65 antigen–negative patients, we also clearly demonstrated that CMV anergy was associated with a more global suppression of T cell reactivity after superantigen stimulation (SEB). Unresponsiveness to CMV antigen was associated with an increased age, which is a well-known phenomenon due to the senescence of the immune system [31]. Interestingly, 2 HSV reactivations were found in the group with CMV-reactive Th1 cells, giving an additional hint to our hypothesis that T cell reactivity contributes to the reactivation of various herpesviruses [12]. However, the clinical signs and symptoms were not different between pp65 antigen–negative patients with and those without CMV-specific Th1 cells by SOFA score, hospitalization, mechanical ventilation, and catecholamine use, indicating that increased morbidity was correlated with active CMV infection and not with T cell reactivity alone. In patients with T cell deficiency due to T cell–depleted hemopoietic stem cell transplantation, NK cells may at least partially substitute for the absence of adaptive T cell immunity [3, 5]. However, in our patients with septic shock, the termination of active CMV infection occurred despite the long-lasting suppression of cytotoxic NK cell activity, and even use of IL-2 application in vitro could not restore the cytotoxic NK cell activity [21]. Suppression of cytotoxic NK cell activity (!10%) was found despite the presence of CD56+CD16+ cells. Thus, we suggest that the absence of cytotoxic NK cell activity mostly occurred because of NK cell anergy and not NK cell depletion. NK cell function can be suppressed by IL-6 [32, 33], an abundant cytokine in sepsis and SIRS [15]; however, in the present study, cytokine concentrations were not determined. Therefore, the role played by IL-6 in NK cell function in patients with septic shock remains speculative. In the present study, we have succeeded for the first time in monitoring CMV reactivation and antiviral immunity in patients with septic shock who were not undergoing immunosuppressive therapy. In future studies, this natural model of CMV reactivation may help to investigate the mechanisms of CMV reactivation and the interactions between active CMV infection, immunity, and CMV-associated morbidity in patients not undergoing immunosuppressive therapy. We conclude that reactivation of CMV develops despite the presence of CMV-reactive Th1 cells in patients with septic shock. CMV reactivation in patients with septic shock may occur because of a proinflammatory immune response with T cell reactivity and, at least in part, NK cell dysfunction. Adaptive T cell immunity, but not innate NK cell immunity, was activated in patients with CMV reactivation. Therefore, adaptive T cell immunity, more likely than NK cell immunity, may contribute 1294 • JID 2007:196 (1 November) • von Mu¨ller et al.

to the termination of active CMV infection in the absence of antiviral therapy in patients with septic shock.

References 1. Pass RF. Cytomegalovirus. In: Knipe DM, Howley PM, eds. Fields virology. Philadelphia: Lippincott Williams & Wilkins, 2001:2675–705. 2. Asanuma H, Sharp M, Maecker HT, Maino VC, Arvin AM. Frequencies of memory T cells specific for varicella-zoster virus, herpes simplex virus, and cytomegalovirus by intracellular detection of cytokine expression. J Infect Dis 2000; 181:859–66. 3. Hertenstein B, Hampl W, Bunjes D, et al. In vivo/ex vivo T cell depletion for GVHD prophylaxis influences onset and course of active cytomegalovirus infection and disease after BMT. Bone Marrow Transplant 1995; 15:387–93. 4. Cerboni C, Mousavi-Jazi M, Linde A, et al. Human cytomegalovirus strain-dependent changes in NK cell recognition of infected fibroblasts. J Immunol 2000; 164:4775–82. 5. Cook M, Briggs D, Craddock C, et al. Donor KIR genotype has a major influence on the rate of cytomegalovirus reactivation following T-cell replete stem cell transplantation. Blood 2006; 107:1230–2. 6. Boeckh M, Boivin G. Quantitation of cytomegalovirus: methodologic aspects and clinical applications. Clin Microbiol Rev 1998; 11:533–54. 7. Kutza AST, Muhl E, Hackstein H, Kirchner H, Bein G. High incidence of active cytomegalovirus infection among septic patients. Clin Infect Dis 1998; 26:1076–82. 8. Barry SM, Johnson MA, Janossy G. Cytopathology or immunopathology? The puzzle of cytomegalovirus pneumonitis revisited. Bone Marrow Transplant 2000; 26:591–7. 9. Curtsinger LJ, Cheadle WG, Hershman MJ, Cost K, Polk HC Jr. Association of cytomegalovirus infection with increased morbidity is independent of transfusion. Am J Surg 1989; 158:606–10. 10. Heininger A, Jahn G, Engel C, Notheisen T, Unertl K, Hamprecht K. Human cytomegalovirus infections in nonimmunosuppressed critically ill patients. Crit Care Med 2001; 29:541–7. 11. Cook CH, Martin LC, Yenchar JK, et al. Occult herpes family viral infections are endemic in critically ill surgical patients. Crit Care Med 2003; 31:1923–9. 12. von Mu¨ller L, Klemm A, Weiss M, et al. Active cytomegalovirus infection in patients with septic shock. Emerg Infect Dis 2006; 12: 1517–22. 13. American College of Chest Physicians/Society of Critical Care Medicine consensus conference: definitions for sepsis and organ failure guidelines for the use of innovative therapies in sepsis. Crit Care Med 1992; 20: 864–74. 14. Riedemann NC, Guo RF, Ward PA. The enigma of sepsis. J Clin Invest 2003; 112:460–7. 15. Levy MM, Fink MP, Marshall JC, et al. 2001 SCCM/ESICM/ACCP/ ATS/SIS International Sepsis Definitions Conference. Crit Care Med 2003; 31:1250–6. 16. Bone RC. Sir Isaac Newton, sepsis, SIRS, and CARS. Crit Care Med 1996; 24:1125–8. 17. Hotchkiss RS, Karl IE. The pathophysiology and treatment of sepsis. N Engl J Med 2003; 348:138–50. 18. von Mu¨ller L, Hampl W, Hinz J, et al. High variability between results of different in-house tests for cytomegalovirus (CMV) monitoring and a standardized quantitative plasma CMV PCR assay. J Clin Microbiol 2002; 40:2285–7. 19. Waner JL, Budnick JE. Blastogenic response of human lymphocytes to human cytomegalovirus. Clin Exp Immunol 1977; 30:44–9. 20. Sester M, Gartner BC, Sester U, Girndt M, Mueller-Lantzsch N, Kohler H. Is the cytomegalovirus serologic status always accurate? A comparative analysis of humoral and cellular immunity. Transplantation 2003; 76:1229–30. 21. Puente J, Carvajal T, Parra S, et al. In vitro studies of natural killer

22.

23.

24.

25.

26.

cell activity in septic shock patients: response to a challenge with alphainterferon and interleukin-2. Int J Clin Pharmacol Ther Toxicol 1993; 31:271–5. Piriou L, Chilmonczyk S, Genetet N, Albina E. Design of a flow cytometric assay for the determination of natural killer and cytotoxic Tlymphocyte activity in human and in different animal species. Cytometry 2000; 41:289–97. Vincent J-L, Moreno R, Takala J, et al. The SOFA (sepsis-related organ failure assessment) score to describe organ dysfunction/failure. Working Group on Sepsis-Related Problems of the European Society of Intensive Care Medicine. Intensive Care Med 1996; 22:707–10. Reinke P, Pro¨sch S, Kern F, Volk H-D. Mechanisms of human cytomegalovirus (HCMV) (re)activation and its impact on organ transplant patients. Transpl Infect Dis 1999; 1:157–64. Soderberg-Naucler C, Streblow DN, Fish KN, lan-Yorke J, Smith PP, Nelson JA. Reactivation of latent human cytomegalovirus in CD14+ monocytes is differentiation dependent. J Virol 2001; 75:7543–54. Cook CH, Trgovcich J, Zimmerman PD, Zhang Y, Sedmak DD. Lipopolysaccharide, tumor necrosis factor alpha, or interleukin-1beta triggers reactivation of latent cytomegalovirus in immunocompetent mice. J Virol 2006; 80:9151–8.

27. Reusser P, Attenhofer R, Hebart H, Helg C, Chapuis B, Einsele H. Cytomegalovirus-specific T-cell immunity in recipients of autologous peripheral blood stem cell or bone marrow transplants. Blood 1997; 89:3873–9. 28. Einsele H, Roosnek E, Rufer N, et al. Infusion of cytomegalovirus (CMV)-specific T cells for the treatment of CMV infection not responding to antiviral chemotherapy. Blood 2002; 99:3916–22. 29. Bernasconi NL, Traggiai E, Lanzavecchia A. Maintenance of serological memory by polyclonal activation of human memory B cells. Science 2002; 298:2199–202. 30. Cook CH, Zhang Y, Sedmak DD, Martin LC, Jewell S, Ferguson RM. Pulmonary cytomegalovirus reactivation causes pathology in immunocompetent mice. Crit Care Med 2006; 34:842–9. 31. Sester M, Sester U, Alarcon SS, et al. Age-related decrease in adenovirus-specific T cell responses. J Infect Dis 2002; 185:1379–87. 32. Riedemann NC, Guo RF, Ward PA. Novel strategies for the treatment of sepsis. Nat Med 2003; 9:517–24. 33. Vredevoe DL, Widawski M, Fonarow GC, Hamilton M, Martinez-Maza O, Gage JR. Interleukin-6 (IL-6) expression and natural killer (NK) cell dysfunction and anergy in heart failure. Am J Cardiol 2004; 93: 1007–11.

CMV, Cellular Immunity, and Septic Shock • JID 2007:196 (1 November) • 1295