Direct ex vivo comparison of the breadth and specificity of the T cells in the liver and peripheral blood of patients with chronic HCV infection

Abbreviation:

1 Introduction

Hepatitis C virus (HCV) infection is an important cause of chronic liver disease. Up to 80 % of HCV infections become chronic and infected individuals may develop liver disease, including cirrhosis and hepatocellular carcinoma. HCV is now the most common cause of chronic viral hepatitis in the developed world and the most common cause of cirrhosis requiring liver transplantation 1.

The immune system plays an essential role in the control and elimination of acute viral infections and there is evidence that vigorous and rapid cellular immune responses with broad specificity can control and, under certain circumstances, clear persistent viruses such as HIV 2 and HBV 3. The situation during HCV infection is less clear. T cells may contribute to control of the virus 4, 5, but as has been shown for LCMV 6, 7, they may also cause hepatic disease 4, 5. T cells from both the peripheral blood and liver of infected patients can be expanded 8, 9; however, such studies may identify only a small proportion of T cells selected on the basis of retention of their proliferative and functional capacity in vitro. In addition, studies relying on in vitro expansion of T cells do not allow direct comparison of the numbers of cells with specificity for HCV in the peripheral and liver compartments. More recently, the advent of tetrameric MHC-peptide complexes, which can be used to determine the specificity of the antigen-specific receptor of T cells in combination with other phenotypic markers 10, 11, has allowed more direct quantitation of virus-specific T cells in infected patients.

In acute HCV infection, strong peripheral T cell responses can be detected 12, 13; however, such responses are frequently not sustained 14 and are more difficult to detect during the chronic phase of the disease 8, 9. It is not clear if this is due to a down-regulation of the immune response to HCV, which could help to explain the persistence of the virus, or compartmentalization of the cells to, for example, the liver, where they may be responsible for the damage associated with the infection. Characterization of the intrahepatic T cell response is, therefore, likely to be extremely important in understanding the mechanisms resulting in chronic HCV infection and the underlying liver pathology. There have been very few studies on liver-derived cells in HCV infection using MHC tetramers, and the data generated are not clear: in one study a relatively high frequency of HCV-specific T cells was found in the liver of two HCV-infected patients 15, but in another there was no evidence of such T cells in six HCV-infected liver samples 16. We have isolated individual cells from small amounts of tissue from needle liver biopsies and from peripheral blood of patients chronically-infected with HCV and provide a direct comparison of the T cells in these two compartments in terms of their specificity and TCR gene usage.

2 Results

2.1 Investigation of the diversity of the T cell repertoire within the liver

T cells from liver biopsies and paired peripheral blood cells from six HCV RNA⁺ patients were enriched using anti-CD3-labeled magnetic beads. RNA isolated from these cells was analyzed for variability in the size of the Vβ-Cβ region. Such variation arises from differences in D and J region usage and junctional diversity arising during recombination of the Vβ gene. The Vβ profiles for each of the patients are shown in Fig. 1. For each patient, the data for two independent PCR from liver cDNA and one PCR from peripheral blood cDNA are shown. In all but one of the patients both samples show variable Vβ profiles, comparable with that found in the peripheral blood of the same patients. In the remaining patient (75) one sample had only one peak and the other had limited variability in terms of numbers of peaks. Such variability is likely to be due to sampling errors as a result of sampling from a single liver site.

2.2 Quantitation of HCV-specific intrahepatic CD8⁺ T cells

Cells were dissociated from the liver biopsies of six HCV-infected, HLA-A2 patients and also prepared from paired peripheral blood samples. Data relating to these patients at the time of biopsy are given in Table 1. The cells were stained with a pool of HCV-specific HLA-A2 tetramers (Table 2) and markers for CD8 and CD69. The results are shown in Fig. 2. The quadrants were set based on the results obtained with 2 HCV-negative biopsies and one biopsy from an HCV⁺ patient who was not of the HLA-A2 genotype. In three of the patients (250, 261 and 344) a clear population of tetramer⁺ cells can be seen in the liver CD8⁺ populations. In a fourth patient (263) a small number of tetramer⁺ cells were seen in the liver; these did not form a clear cluster of cells, but were considered to be positive since they lie within the region of the graph where tetramer⁺ cells are found. The remaining two livers (251 and 87) appear to be negative; while some events did fall within the upper two quadrants, the majority of these were outside the area expected for tetramer⁺ cells. In contrast, the peripheral blood samples had much lower numbers and percentages of tetramer⁺ cells. Only one patient (261) showed a clear population of tetramer⁺ cells in the periphery.

The tetramer and CD69 data for liver and paired peripheral blood-derived T cells are summarized in Table 1. In general, the proportion of peripheral CD8⁺ cells which are tetramer⁺ is very low, even in those patients in whom there is a significant proportion of tetramer⁺ cells in the liver. There was a significant difference between the percentage of tetramer⁺ cells in the liver-derived and peripheral populations (p < 0.05, Wilcoxon matched pairs test). In the one patient with a clear peripheral population of tetramer⁺ cells the corresponding frequency in liver-derived cells was approximately 16-fold higher. Of a further eight HLA-A2 chronic HCV RNA⁺ patients from whom only peripheral blood-derived T cells were tested (data not shown), only one had a significant population of tetramer⁺ cells (0.04 %).

Table 1. Summary of staining of paired liver and peripheral blood CD8⁺ cells with HCV-specific tetramers and CD69^a)

Patient	Age	Sex	ALT	Knodell	Liver cells			Peripheral blood cells
			at time	score	CD8+	CD8+	tetramer+	CD8+	CD8+	tetramer+
			of biopsy	of biopsy	tetramer+	CD69+	CD69+	tetramer+	CD69+	CD69+
250	39	M	64	3	4.00 %	56.3 %	87.5 %	0.04 %	2.7 %	50.0 %
251	34	F	19	1	0.06 %	57.3 %	66.7 %	0.02 %	0.8 %	50.0 %
261	47	M	332	7	2.10 %	36.3 %	97.1 %	0.13 %	1.4 %	15.4 %
263	47	F	108	3	0.36 %	65.2 %	83.3 %	0.02 %	0.4 %	0.0 %
87	35	F	55	3	0.20 %	58.6 %	100.0 %	0.01 %	0.4 %	0.0 %
344	43	M	148	1	1.27 %	55.8 %	97.6 %	0.03 %	1.6 %	50.0 %

• a) Percentages of CD8⁺, tetramer⁺ which were considered to be significant are indicated in bold-type.

Table 2. Sequences and derivation of peptides of the HCV-specific HLA-A2 tetramers.

Tetramer	Peptide sequence	Derivation of peptide
NS3 1073	CINGVCWTV	HCV NS3 1073 – 1081
NS3 1406	KLVALGINAV	HCV NS3 1406 – 1415
NS4B 1807	LLFNILGGWV	HCV NS4B 1807 – 1816
NS5 2594	ALYDVVTKL	HCV NS5 2594 – 2602

A large proportion of the total liver CD8⁺ population was CD69⁺ with a CD8⁺/CD69⁺ to CD8⁺/CD69^– ratio of approximately 2 : 1. Again, this is in contrast with the peripheral blood where very few CD8⁺ cells were CD69⁺. The tetramer⁺ liver cells had a raised proportion of CD69⁺ cells compared with the total liver CD8⁺ populatijon (p < 0.05). In the patient (261) with a clear population of tetramer⁺ peripheral CD8 cells, only 15 % co-expressed CD69 as compared with the same patient's liver-derived tetramer⁺ cells of which 88 % co-expressed CD69.

3 Discussion

Our data provide a direct comparison of the breadth and specificity of the intrahepatic and peripheral T cells in patients with chronic HCV infection. It has been suggested that an immune response with broad specificity is likely to be most effective in viral clearance. Therefore, we firstly investigated the variability of the TCR Vβ usage in CD3⁺ cells from the liver, in comparison with peripheral blood from the same patients, by RT-PCR. The Vβ usage in the liver-derived CD3⁺ cells was comparable to that found in the periphery in the same patients, in spite of the potential for the Vβ profiles of the liver-derived cells to appear clonal due to sampling from a single liver site. Our results are unlikely to be due to contamination with peripheral blood PBMC since other characteristics of the liver cells contrast markedly with those of peripheral cells from the same patients. These data are in broad agreement with previous studies 17, 18 which showed that while there may be some biases in TCR usage at a detailed level, the overall profile is not clonal. The characteristics of the liver-derived T cells are those of classical T cells, with expression of α β TCR and no obvious bias towards particular Vβ usages. We cannot rule out that non-classical T cells, such as the NKT cells ob-served in normal 19 and HCV-infected liver 20 are also present, but since such cells tend to have restricted Vβ as well as Vα usage, they are likely to be in the minority. Direct studies of the Vβ usage of tetramer⁺ cells in the periphery by antibody staining and flow cytometric analysis has, similarly, not revealed large clonal expansions in this antigen-specific population (Klenerman et al., unpublished data).

Analysis of the Vβ usage of T cells in the liver gives some idea of the breadth of the T cell population. However, direct analysis of the fine specificity of CD8⁺ cells has become possible with the development of tetrameric MHC class I molecules refolded with specific peptides in their peptide-binding grooves 10, 11. Class II tetramersare being developed 21 but such reagents displaying HCV-derived peptides are not yet available. We found clear populations of cells binding HCV-specific tetramers in three out ofsix of the HCV-infected HLA-matched livers analyzed consisting of a discrete cluster of stained cells, separated from the bulk of the CD8 cells. In each of these livers, the proportion of tetramer⁺ cells was greater than 1 % of the CD8⁺ cells, and in one liver it was as high as 4 %. The results described here are in agreement with the findings of He et al. 15 who found 1 % and 2.5 % positive cells in two HCV-infected liver samples. In contrast with our results and those of He et al. 15, Valiante et al. 16 found no tetramer⁺ cells in six HCV-infected livers, using the same tetramers used in this study. However, the latter study used cells derived from cirrhotic liver explants. The discrepancy in tetramer data between that study and our own raises the possibillity that the nature of the intrehepatic T cell population may change over time as liver disease progresses.

The tetramer⁺ population measured here is likely to be an underestimate of the total population of intrahepatic HCV-specific cells in the patient, since the cellular response to HCV is thought to be broad 8 and yet we measured the response to only four epitopes of HCV in the context of a single HLA haplotype. In addition, it is possible that HCV-specific T cells which do recognize the epitopes represented by the tetramers used in this study are missed in some patients because the viral variants in those patients have minor differences in sequence within these epitopes 22, 23 so that the fine specificity of the TCR does not perfectly match the tetramers. However, use of the same set of tetramers in the two compartments allows a direct comparison of liver and peripheral blood in the same patients. Such a comparison indicated that HCV-specific cells are significantly less frequent (⩽ 0.13 %) in the periphery. While the tetramer⁺ cells in the periphery were generally not activated, a large proportion (74 – 93 %) of the tetramer⁺ CD8 cells in the liver were stained with CD69,a short-term marker of activation, suggesting that these cells had been recently activated, perhaps in the efferent lymph nodes or in the liver itself.

The broad TCR usage of the T cells in the HCV-infected livers suggests that the cells do not arise from a clonal expansion of a small number of progenitor cells within the liver, but either represent a set of peripheral blood cells which accumulate nonspecifically in the liver, or a set of broadly reactive T cells which have expanded elsewhere and home to the liver. The former possibility is unlikely because the cells in the liver have very different characteristics from those of the periphery both in terms of their specificity and their activation state. Their expression of a short-term marker of activation suggests they have been recently activated. It has been shown that activated cells, particularly CD8⁺ cells, accumulate in the liver where they die by apoptosis 24; however, such a mechanism probably does not apply to cells which have been activated in lymph nodes close to the liver and selected for their liver-homing properties 25, as is likely to be the case with the HCV-specific liver-derived T cells described here. This homing of the cells to the liver may explain the disappearance of the cells from the periphery following the acute stage of infec-tion 14.

Once in the liver, during the chronic phase of disease these activated cells may play a role in keeping viral replication under control. However, in doing so they may damage the liver 6, 7 leading to the progressive inflammatory and fibrosing disease associated with HCV infection 4, 5. It is possible that in end-stage cirrhotic liver disease, the HCV-specific T cells may disappear from the liver as well as the periphery 16. Further studies are needed to address these possibilities in more detail, in particular to measure the capacity of these cells for a functional response to HCV antigens. Correlation of such an analysis with later progression to severe disease and response to treatment may be important in increasing our understanding of the role of the immune response in HCV-related liver disease.

4 Materials and methods