Division of Gastroenterology and Hepatology, Liver Transplantation Program Hepatitis C Research Center, and Integrated Program in Immunology; University of Colorado Health Sciences Center and National Jewish Hospital, Denver, CO
Division of Gastroenterology/Hepatology, UCHSC GI Division, 4200 East Ninth Ave. #B-158, Denver, CO 80262
Lucy Golden-Mason and Cliona O'Farrelly contributed equally to this study.
Division of Gastroenterology and Hepatology, Liver Transplantation Program Hepatitis C Research Center, and Integrated Program in Immunology; University of Colorado Health Sciences Center and National Jewish Hospital, Denver, CO
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
Liver disease related to hepatitis C virus (HCV) is the single leading indication for orthotopic liver transplantation (OLT) throughout the world. HCV infection significantly impairs patient and allograft survival following liver transplantation (LT) (hazard ratio 1.30 for allograft failure after adjusting for potential confounders)1 and is unique among indications for LT in that recurrence is nearly universal. However, the spectrum of histologic injury related to HCV recurrence is highly variable, ranging from mild histologic abnormalities to allograft cirrhosis in 20 to 30% of recipients by the fifth postoperative year.2 Unfortunately, in contradistinction to other viral infections (e.g., cytomegalovirus), no tests accurately and consistently predict who will develop progressive HCV recurrence prior to LT.
Although the National Institute of Diabetes and Digestive and Kidney Diseases liver transplant database found that high viral loads correlate with diminished graft and patient survival following transplantation,3 therefore providing a rationale for preemptive antiviral treatment, much of the experience to date suggests that interferon (IFN)-based treatment in this setting is fraught with significant morbidity and lack of tolerance.4 Thus, identifying a group of variables before OLT that predicts more aggressive HCV recurrence and a therapeutic approach that targets only patients at risk is of paramount importance.
As reviewed recently, the human LT model provides a unique opportunity and theoretical framework to study HCV immunopathogenesis for a number of reasons.5 Although a number of groups have demonstrated that the presence of HCV-specific effector responses correlate with attenuated histologic severity,6, 7 considerable gaps exist in our understanding of how the immune response shapes HCV recurrence post-OLT. In particular, the role of innate immunity, as represented by natural killer (NK) and natural T (NT) cells,8 the same cell populations that are usually associated with the very early stages of an acute immune response, has not been examined. We hypothesized that the nature of the innate immune response prior to OLT would correlate with the severity of HCV recurrence following LT. In experiments aimed at elucidating the frequency, phenotype, and function of (CD)56+ lymphocytes, we examined patients with HCV who subsequently developed severe histologic recurrence or demonstrated mild histologic recurrence, as well as patients with end-stage liver disease unrelated to HCV and normal healthy controls. Our results indicate a previously unappreciated role for these innate lymphocytes and provide novel mechanistic insights into the immunopathogenesis of HCV recurrence.
The study group was comprised of healthy subjects and patients undergoing LT in Portland, OR. Informed consent was obtained in all cases and Internal Review Board approval was granted at the Portland Veterans' Administration Medical Center and the Oregon Health and Science University. The HCV-infected patients underwent OLT between November 1999 and May 2004 and typically underwent protocol liver biopsies at 6, 12, 18, and 24 months and annually thereafter. Modified hepatitis activity index and fibrosis scores were determined in all specimens9 and used to define the severity of HCV recurrence. Severe HCV recurrence was defined as the presence of grade 3 or 4 inflammation, stage ≥2 fibrosis, and/or evidence of the cholestatic variant. All patients were initially maintained on a double or triple immunosuppressive regimen including tacrolimus, steroids, and/or purine antagonists. Levels of HCV-ribonucleic acid levels in serum were determined by a branched-chain deoxyribonucleic acid signal amplification assay (Quantiplex HCV RNA 2.0 Assay; Bayer Diagnostics, Emeryville, CA). The lower limit of this assay is 0.2 and the upper limit is 120 mEq/mL (equal to 40 million copies/mL).
Sample Preparation and Storage
Peripheral blood mononuclear cells (PBMCs) were isolated from whole blood using cellular preparation tubes (Becton-Dickinson, Franklin Lakes, NJ; anticoagulant sodium citrate). PBMCs were viably frozen in 80% fetal bovine serum (FBS; BioWhittaker, Walkersville, MD), 10% dimethylsulfoxide, and 10% Roswell Park Memorial Institute (RPMI) 1640 Media (Life Technologies, Grand Island, NY) in liquid nitrogen. Following thaw, PBMCs were cultured in medium consisting of RPMI 1640 Media supplemented with 10% fetal bovine serum (FBS), 50 μg/ml gentamicin sulfate (BioWhittaker), 5 × 10−5 M 2 ME (Sigma, St. Louis, MO), and 2 mM glutamine (Life Technologies).
Plasma preparation tubes (PPT tubes; BD Biosciences, San Jose, CA) were used to isolate plasma from whole blood, which was frozen and later thawed for viral load and genotype testing.
Flow Cytometric Analysis of Cell Surface Antigens
Four-color multiparameter flow cytometry was performed using a BD FACSCalibur instrument (BD Biosciences) compensated with single fluorochromes and analyzed using CellQuest software (BD Biosciences). Fluorochrome-labeled (fluorescein isothiocyanate/PE/PerCP/APC) monoclonal antibodies specific for CD3, CD4, CD8, CD16, CD161, CD56, CD94, CD95, and CD95L were obtained from BD Biosciences. Anti-NKG2A-PE was obtained from Immunotech (Beckman Coulter, Fullerton, CA). PBMCs (2.5 × 105) were stained for cell surface antigen expression at 4°C in the dark for 30 minutes, then washed twice in 2 mL phosphate-buffered saline containing 1% bovine serum albumin and 0.01% sodium azide (FACS Wash) and subsequently fixed in 200 μL of 1% paraformaldehyde (Sigma-Aldrich, St. Louis, MO). Isotype-matched fluorescently-labeled control antibodies were used to determine background levels of staining. Lymphocytes were identified by characteristic forward scatter and side scatter parameters and populations of interest were gated on specific patterns of staining for CD3 and CD56 within the lymphocyte population (Fig. 1A). Results are expressed as % positive of gated populations.
Antibodies for measurement of intracellular IFN-γ were supplied by BD Pharmingen. Thawed mononuclear cell suspensions were stimulated with phorbol myristate acetate (PMA; 10 ng/mL: Sigma-Aldrich) and ionomycin (1 μg/mL; Sigma-Aldrich) for 4 hours at 37°C in the presence of brefeldin A (Sigma-Aldrich). After stimulation cells were stained for surface antigens (as above), fixed for 30 minutes at 4°C in 100 μL Fix and Perm Medium A (Caltag, Burlingame, CA), permeabilized using 100 μL Fix and Perm Medium B (Caltag) and incubated with anti-cytokine monoclonal antibodies for 1 hour. Cell suspensions were then washed in phosphate-buffered saline–bovine serum albumin–sodium azide and fixed in 200 μL 1% PFA and acquired after 1 hour. Cells cultured under the same conditions in the absence of PMA and ionomycin served as controls.
Natural and interleukin-2-induced cytotoxicity by PBMCs against the target cell line K562 were assayed in 4-hour 51chromium-release assays. All PBMCs were cryopreserved immediately after preparation. Upon recovery they were cultured for 48 hours in complete RPMI medium (RPMI 1640 containing 25 mM 4-2-hydroxyethyl-1-piperazineethanesulfonic acid, 2 mM L-glutamine, 50 μg/mL streptomycin, 50 U/mL penicillin, and 10% fetal calf serum) at a density of 0.5 × 106 cells/mL in the absence or presence of 50 units/mL recombinant human interleukin-2 (Biological Resources Branch, National Cancer Institute at Frederick [NCI-Frederick], Frederick, MD), to measure natural and interleukin-2-induced cytotoxicity, respectively. The PBMCs were then incubated with 2,000 51Cr-sodium chromate-labeled target cells (MP Biochemicals, Mechelen, Belgium) at PBMC/target ratios of 1, 5, 25, and 50 in triplicate wells of 96-well plates. Specific lysis was calculated from the amounts of 51Cr released into supernatants using the formula: % specific lysis = (cpm of sample − cpm of spontaneous release) × 100/(cpm of maximum release − cpm of spontaneous release). Spontaneous release was determined by incubating the target cells in the absence of effector cells and maximum release was obtained by incubation of targets with 0.1% Triton X-100. In all experiments, cytotoxicity by an internal control sample consisting of cryopreserved PBMCs from a healthy donor, was assayed in parallel to monitor for day-to-day variation in target cell viability and specific activity of 51Cr.
Results are expressed as median (range). Nonparametric Mann Whitney U test was used to compare differences between study groups. Spearman's test was used for correlation analyses. Significance was defined as a P value of less than 0.05. The JMP 6.0 (SAS Institute, Inc., Cary, NC) statistical package was used.
The study was comprised of four groups: patients with mild HCV recurrence (M1-M9), severe HCV recurrence (S1-10), patients with non-HCV related liver failure (n = 10), and normal healthy subjects (n = 10). The demographic and virologic features of the 19 patients who underwent LT from deceased donors for HCV-related liver failure and from whom PBMCs were collected prior to transplantation are shown in Table 1. All patients received tacrolimus-based basal immunosuppression.
Table 1. Demographic and Clinical Features of HCV-Infected Liver Transplant Recipients
1 month post
2 months post
Abbreviations: na, not available; ABP, atypical band pattern; ND, not detectable.
Protocol liver biopsies were performed at 6, 12, 18, and 24 months and annually thereafter; histology was used to define HCV severity. Like many transplant programs, the strategy we use with regard to initiation of antiviral therapy is to treat when clinically significant evidence of recurrence exists (grade 3 or 4 inflammation, stage ≥2 fibrosis and/or there is evidence of the cholestatic variant).10 Patient S4 had undetectable HCV ribonucleic acid by branched-chain deoxyribonucleic acid testing pre-OLT, but by the first month had >7,692,316 IU/mL. All 10 patients with severe HCV were treated with antiviral therapy, but none in the mild group. Patients S2, S8, and S10 had cholestatic HCV recurrence. A total of 6 of the severe HCV patients are deceased (60%), on the average dying 844 days after their transplantation. All patients in the mild HCV group are alive. Table 2 shows the demographic characteristics of the non-HCV controls, including those with advanced liver disease (pre-OLT) and healthy subjects.
Table 2. Demographic Features of HCV-Negative Controls (± Liver Disease)
Pretransplantation CD56+ Levels Correlate With Severity of HCV Recurrence
Multiparameter flow cytometric analysis was used to determine baseline conventional T (CD56−CD3+), NK (CD56+CD3−), and CD56+ NT cells. CD56+ lymphocytes, NK, and NT cells comprise the innate lymphocyte populations that can recognize conserved structures and signal viral invasion, thus providing an important first line of defense against viral infection.11, 12 We hypothesized that the level and function of these innate lymphocytes pretransplantation might predict the outcome of HCV reinfection and histologic recurrence.
As shown in Figure 1, the frequency in total CD56+ lymphocytes in peripheral blood prior to OLT was significantly lower in patients who subsequently developed severe HCV recurrence relative to those patients who developed mild histologic recurrence, as well as non-HCV liver disease controls. The same pattern was demonstrable when the frequency of NK cells (CD56+CD3−) cells was analyzed (Fig. 1, mid-panel). Moreover, the frequency of NT (CD56+CD3+) cells was significantly lower in HCV-positive patients as compared to normal controls, and the median frequency of NT cells was higher in patients who later demonstrated only mild histologic HCV recurrence (vs. severe HCV).
Next, we examined whether there were differences in phenotypic and functional features among the 3 transplant groups. The phenotypic markers included CD16, CD161 (both activating receptors), and NKG2A (inhibitory receptor). Compared to the normal healthy controls, expression of CD16 on NK cells was significantly downregulated in the HCV patients but not different between mild and severe (Fig. 2A); these data suggest that HCV infection per se impairs antibody-dependent cell cytotoxicity of NK cells. It has been previously shown that the subset of human NK cells that express CD8 are more cytotoxic and more likely to survive after target cell lysis, making them “serial killers.”13 We found that these CD8+CD56+ lymphocytes were depleted in HCV infection although there were no differences between the HCV groups (Fig. 2B). No differences were observed with respect to NK cell expression of CD161, NKG2A, or total CD94 (data not shown).
We also examined the expression of Fas ligand (FasL), previously demonstrated to be a primary mechanism of tumor necrosis factor–dependent hepatotoxicity after activation of innate lymphocytes.14 The NK cells from both HCV-positive groups overexpressed FasL relative to the HCV-negative liver disease control group. Moreover, FasL expression on NK cells pretransplantation was highest in those subjects who subsequently developed severe HCV recurrence (Fig. 3A). FasL was also upregulated on CD56+CD3+ NT and T cells in the HCV group prior to LT compared to the non-HCV LT patient group (Fig. 3B and C). Of interest, increased FasL expression has been previously shown to induce T cell-dependent hepatic inflammation.15 Fas (CD95) was similarly expressed on NK, NT, and NT cells in all study groups (data not shown).
NK Cells and Viral Level
Because hepatitis C viral level previously has been associated with increased risk of developing severe HCV recurrence, we determined whether lower NK levels might be explained by higher viral level. As shown in Figure 4, there were no statistically significant relationships between NK levels and viral level at baseline or 1 month posttransplantation, 2 months posttransplantation, or a serial increase in viral level (data not shown). Thus, the association between NK level and severity of HCV recurrence appears to be independent of viral level.
Functional Analysis of NK cells
Previous studies in immunocompetent individuals have suggested that chronic HCV persistence may be associated with defective NK or NT function.16–19 By secreting cytokines and killing infected targets, both NK and NT cell populations can provide an immediate response, making them poised for early defense. Therefore, diminished function prior to OLT might be hypothesized to be a risk for development of severe HCV recurrence.
Percent specific lysis was assessed for natural cytotoxicity (K562) and lymphokine-activated killing, and as shown in Figure 5, both were diminished in HCV-infected patients as compared to non-HCV controls (including patients with and without liver disease). There were no significant cytotoxic differences among the HCV severity groups, but the numbers of patients were relatively small; HCV per se was associated with significant diminution in natural cytotoxicity and lymphokine-activated killing. Next, we determined whether there was an association between NK frequency and cytolytic activity. Figure 6 demonstrates that in HCV-uninfected liver disease controls, there is a trend toward statistically significant correlation between NK cell frequency and natural cytotoxicity (at an E:T ratio of 50:1). In contrast, HCV infection was characterized by dissociation between NK cell frequency and cytolysis (Fig. 6B). Moreover, NK cell frequency was not associated with lymphokine-activated killing (data not shown).
Experiments designed to measure intracellular cytokine production following short-course PMA stimulation are shown in Figure 7. Comparison of the patients with HCV patients to normal controls showed a reduced frequency of IFN-γ-producing NT cells, however this reduction was also observed in non-HCV-related liver disease.
Pretransplantation T Cell Frequencies and Phenotype Also Correlate With Severity of HCV Recurrence
As an important comparator group to the CD56+ populations, we examined the frequency and phenotype of CD3+ T cells. Unlike NK cells, T cells were not depleted in HCV-infected patients; indeed, the circulating frequency of T cells was greater in the HCV-infected patients awaiting OLT as compared to HCV-uninfected liver transplantation patients (Fig. 8A). Among the patients who subsequently developed severe HCV recurrence, a significantly lower percentage expressed CD4+ (median % for mild vs. severe) and a significantly higher percentage were CD8+ (Fig. 8B) or “double positive,” i.e., expressed both CD4 and CD8 (Fig. 8B). Although relatively rare, the double-positive population has recently been shown to constitute a distinct T-cell population exhibiting typical CD4 and CD8 T-cell functions with a predominant T helper 1/T cytotoxic 1 (Th1/Tc1) profile in chronic HCV.20
NT cells coexpress CD56 and CD3 (Fig. 1A) and can be activated through either receptor. We found that NT cells that coexpressed CD4/CD8 or expressed CD8 alone were more frequent in patients who subsequently developed severe recurrence. In contrast, NT cells that expressed only CD4 appeared to be depleted in HCV infection (relative to non-HCV liver controls) (Fig. 9A). A significantly higher percentage of NT cells in both HCV groups expressed the inhibitory receptor NKG2A relative to HCV-negative controls with or without liver disease (Fig. 9B). No differences were observed with respect to expression of CD16, CD161, or CD94 on NT cells between the study groups (data not shown).
This is the first study to characterize the potential role of CD56+ lymphocytes in patients undergoing LT for HCV-related liver failure. The rationale for studying these lymphocytes is that they are believed to play important roles in the innate immune response to viral infections by production of IFN-γ and/or the recognition of virally-infected cells.8 These functions are an important aspect of the immediate response of the host to the virus and are believed to control the initial infection. Accordingly, injection drug users who remain human immunodeficiency virus-1 uninfected despite many years of high-risk exposure demonstrate significantly augmented NK cell lytic activities and cytokine secretion when compared to human immunodeficiency virus-1 infected injection drug users.21
Our results can be summarized as follows: 1) total CD56+, NK, and NT cells are depleted in the peripheral blood of patients with HCV awaiting LT as compared to patients with other causes of liver failure and healthy controls; 2) NK cells from HCV-positive patients overexpress FasL relative to the HCV-negative liver disease control group and are highest in those subjects who subsequently developed severe HCV recurrence; 3) cytotoxicity is decreased in HCV-positive patients as compared to HCV-negative liver disease and healthy controls; 4) in contrast, the circulating frequency of T cells is greater in the HCV-infected patients, including a double-positive (CD4/CD8) subset more frequent in patients who developed severe recurrence; 5) pre-OLT, patients with chronic HCV demonstrate lower production of IFN-γ by NT cells (but not NK or T cells); and 6) NT cells differentially expressed phenotypic markers associated with HCV infection and severity post-OLT.
The finding that NK and NT levels, phenotype, and function before OLT are associated with divergent outcomes of HCV recurrence is noteworthy because this topic has not been studied previously. NK and NT cells recognize virally-infected cells before major histocompatibility complex class I expression is upregulated, a particularly important property for the control of infections of certain cell types, e.g., hepatocytes, that normally express little or no major histocompatibility complex class I. Moreover, class I antigen recognition required for T cell stimulation is limited by the fact donor and recipients are frequently human leukocyte antigen mismatched in LT.22
Our data indicate that HCV-infected subjects have fewer circulating NK cells as well as total CD56+ cells compared to HCV-uninfected controls, consistent with results published recently by Meier et al.23 The fact that higher levels of CD56+ lymphocytes is protective, i.e., associated with milder HCV recurrence, corroborates the original hypothesis of our study. NK cells mediate the lysis of virus-infected cells via natural cytotoxicity and antibody-dependent cellular cytotoxicity and are controlled by positive and negative cytolytic signals. We found that HCV infection per se results in impairment of cytolytic ability; these results contrast with a recent study that used NK-enriched populations,24 but the study cohorts were different (ours was comprised entirely of patients with advanced liver disease). We examined both natural and lymphokine-activated killing, and found dissociation between NK frequency and K562 cytolytic function in HCV that was not readily demonstrable in the HCV-uninfected patients. This suggests that the functional defect in HCV infection may be even more disproportionately profound than appreciated simply by enumeration of NK frequency. Taken together, our data demonstrating decreased CD16 expression, decreased circulating frequency of CD8+CD56+ lymphocytes, and increased FasL on CD56+ lymphocytes depict a model of global dysfunction in HCV infection with impaired antibody-dependent cellular cytotoxicity and natural cytotoxicity as well as enhanced apoptosis.
There is consensus that CD56+ cells are regulated by a fine balance between positive and negative signals, as reflected by the higher expression of the inhibitory receptor NKG2A on NT cells of patients with HCV (Fig. 9B) in the current study. Future studies will focus on understanding how these components are functionally orchestrated to mediate or protect against HCV-related allograft injury, including the role of recipient/donor immune interactions.
We thank John Ham and Sharlene Winters for their excellent care of the patients and the kind provision of follow-up data. We are also grateful to the patients for their willingness to participate in this study.