HCV is a hepatotropic virus w ith up to 80% o f those infected developing chronic disease. It is therefore a leading cause o f liver related m orbidity and m ortality worldwide.
U nderstanding o f the pathogenesis o f HCV infection is lim ited by the lack o f sm all animal m odels or tissue culture system s that support viral replication. V iral clearance is dependent upon a strong, sustained and m ulti-specific response by C D 4+ T helper cells that is
supported by a sim ilar response by C D 8+ cytotoxic T lym phocytes, in the presence o f a Th- 1 like cytokine response. C hronic infection is thought to result from inadequacies in the strength, specificity and duration o f the response by T helper cells and C T L ’s.
T helper cells and C T L ’s are activated upon presentation o f viral peptide bound to MHC class II and I alleles respectively. Class I and II alleles are inherited as a haplotype as a single M endelian trait. A lthough class I and II alleles are highly polym orphic their scope for foreign peptide binding is genetically restricted.
T he outcom e in HCV m ay be due to host factors such as gender, age and alcohol consum ption and viral factors such as genotype and inoculating dose. However,
epidem iological studies on HCV are confounded by the fact that m ost patients are anicteric during acute infection and therefore the duration o f infection is ill defined. In Ireland, the cohort o f R hesus negative w om en that received anti-D Ig in 1977 represents a unique study population for the exam ination o f the im m unogenetics o f HCV. They are a single gender, single ethnic group, infected w ith a single viral genotype over a known period o f time. In our exam ination o f the class II locus, w e show a significant association betw een D R B l *0101 and DQB1*0501 and viral clearance. W orldw ide, several studies have dem onstrated associations betw een class II alleles and viral outcom e, but the European studies are m ost consistent in their D Q B l *0301 association. H ow ever, this association is m ade through linkage disequilibrium w ith D R B l alleles; D R B 1 * 1 1 in the French and D R B l *04 in the English populations respectively. W e did show an association with D R B l*0401, but com parisons betw een studies show different allele frequencies, to be expected in different ethnic groups, but also varying viral genotypes. A second study o f D R B l alleles in tw o Irish anti-D cohorts dem onstrated an association betw een D R B l *0101 and viral clearance, supporting our work. However, this study w as limited by the num ber o f alleles exam ined and by the m ixed viral genotype.
Subsequently, we show that class I alleles, A*03, B*07, B*27 and Cw*01 are associated w ith viral clearance and B*08 and B*18 are associated w ith chronic infection. Com bining data, the haplotypes A *03-B *07-D R B 1*15-D Q B 1*0602 and A *02-B *27-D RB 1*0101- D Q B l *0501 are associated w ith viral clearance. T he haplotype A *01-B *08-C w *07-
D R B l * 03011-D Q Bl *0201 is associated with chronic infection in this population. W hereas the m ajority o f other studies have failed to dem onstrate a class I association, our positive findings are m ost likely due to the unique single viral genotype, single ethnic cohort that removes confounding variables that im pact upon the host im m une response. The association o f both class 1 and II alleles supports the hypothesis that viral clearance is dependent on a strong and m ultispecific response from both C D 4+ and C D 8+ T cells.
M ultiple logistic regression analysis reveals that HLA A*03, B*27, D R B 1*0101, *0401 and *15 are significantly associated w ith viral clearance w hen the confounding effects o f other alleles are taken into account but the frequency o f som e o f these alleles is small and o f doubtful biological significance. However, over 75% o f subjects w ho clear the virus carry either an A*03, D R B l *0101 or -*0401 allele com pared with only 37% o f those with chronic infection. The significance o f this lies in the potential to identify a com m on viral epitope that binds these com m on alleles and results in appropriate im m une response to clear HCV.
T N F a is a cytokine w ith both im m unom odulatory, pro-inflam m atory and apoptotic functions. In HCV, elevated T N F a levels in a Th-1 like cytokine response has been shown in viral clearance and chronic infection. T he gene for T N F a lies w ithin the class III locus o f the MHC and contains both m icrosatellite and gene prom oter polym orphism s that are inherited in linkage w ith class I and II alleles. The secretion o f T N F a m ay be the outcom e o f class II restricted T cell activation. W ithin this population TN F a6b5, linked to
D R B 1 *0101 -D Q B 1 *0501, and TN F a 11 b4, linked to D R B 1 * 1501 -D Q B 1 *0601, are associated with viral clearance and TN F a2b3-TN F 308.A, linked to D R B l *03011,
DQB 1 *0201, is associated w ith chronic infection. No TN F polym orphism is independently associated w ith viral outcom e by m ultiple logistic regression analysis. H ow ever, the inheritance o f TN F polym orphism s in a class II restricted pattern m ay have functional significance as the strength o f association betw een outcom e and haplotypes is strengthened w hen the TN F loci are included.
In this population we have shown M HC haplotypes extending across class I and II loci that are associated w ith both viral clearance and chronic infection. W hether it is these alleles or other loci carried in linkage disequilibrium , acting independently or in synergy, that control the host im m une response to HCV rem ains undefined. H ow ever, this study dem onstrates that in this unique population, viral clearance is dependent on both class I and II alleles and supports the notion that a com bined appropriate response from both C D 4+ and CD 8+ T cells, w ith a Th-1 like cytokine profile results in HCV clearance.
On the other hand, viral persistence may be as a result o f viral escape m utants, T cell dysfunction or T ccll exhaustion (N cum ann-H aefelin et al, 2005, R eherm ann et al 2005, B ow en et al, 2005). How the interaction o f the HLA allele, the viral epitope and the ensuing T cell response results in either HCV clearance or persistence is the fundam ental question. As previously discussed, other studies have exam ined the evolution o f viral escape in acute H CV infection (T ester et al, 2005, C ox et al, 2005). In this unique cohort in an effort to answ er this question one part w ould be to identify the viral epitopes that bind w ith affinity to alleles associated w ith clearance, that is H LA -B*27, A *03, DRB1*0101 and
D R B l *0401 alleles. O nce identified these epitopes can then be exam ined for escape m utations and utilised to define the response o f both T helper cells and C T L ’s, at best ex- vivo, in subjects w ith both viral persistence and clearance. To this end, in collaboration, C T L ’s from H LA B *27+ subjects w ere exam ined for peptide-specific responses when tested against overlapping peptides (IS m ers overlapping by 1 la a ) derived from genotype la and lb strains (N eum ann-H aefelin et al, subm itted for publication). Five B*27 restricted epitopes had been previously identified in a HLA B*27+ subject w ith acute resolving HCV. Five o f 6 subjects that cleared HCV responded to only one o f the known HLA B*27- restricted epitopes, NS5 2841. hi chronically infected B*27+ subjects, the C TL response w as again lim ited to the NS5 2841 epitope but only occurred in 3 o f 8 subjects exam ined. Sequence analysis o f this epitope revealed variations that occurred significantly m ore often in B*27 + com pared with B*27- subjects, suggesting HLA -B*27 m ediated selective pressure against this region. A lso NS5 2841 variants resulted in reduced IFNy production com pared to w ild type, supporting the biological significance o f variants at this position. This study provides an im m unological basis for the protective role o f HLA B*27 in HCV infection and suggests that the em ergence o f escape m utations w ithin a single key dom inant epitope is central to viral persistence. R eplication o f this study to identify epitopes
restricted by other outcom e-associated alleles is needed to validate these findings and strengthen their significance as the population frequency o f H LA B *27 com pared w ith the o ther alleles described is low.
As discussed a broad and m ulti-specific C TL response to HCV m ust occur to clear HCV but also a sim ilar T helper response is required to support this. HCV antigen-driven proliferation in individuals w ho develop persistent infection are usually w eak or absent w hen com pared w ith spontaneously resolving infections. T his study has dem onstrated that w hile the class I alleles have the strongest independent association w ith viral clearance, the class II alleles are also significantly associated. W hat o f the im m une response in the 24 HLA D R B l *0101+ individuals in this study that cleared HCV com pared w ith the 4 HLA D R B l*0101+ subjects that also carry B*27? Is it possible in these individuals that the
effector m echanism s o f the T helper cells is the m ore critical factor in viral elim ination com pared w ith the C TL response o f the B*27+ individuals? T hat is, is there a difference betw een the type o f im m une response generated against HCV epitopes dependent upon HLA restriction. M any studies have exam ined HLA restricted C TL responses and considerably less the HLA restricted CD 4+ T cell responses to HCV epitopes, none in parallel. T herefore to clarify w hether the driving force o f the im m une response to HCV is HLA dependent requires an exam ination o f the CTL and C D 4+ responses to H LA restricted epitopes in individuals w ho possess the HLA-B*27, A*03, DRB1*0101 and D R B 1*040I alleles and com pare these responses to those who lack one or all.
C ertainly, m uch ground has been achieved in understanding the im m une response to HCV since its identit'ication in 1992. However, in order to achieve goals such as a CTL- based vaccine m ore definitive answ ers are required and hopefully som e m ay be provided by studies such as that proposed.