Persistent HCV replication despite priming of functional CD8+ T cells by combined therapy with a vaccine and direct acting antiviral

Abstract

Exhaustion of antiviral CD8+ T cells contributes to persistence of hepatitis C virus infection. This immune response has proved difficult to restore by therapeutic vaccination, even when HCV replication is suppressed using antiviral regimens containing type I interferon (IFN). Because immunomodulatory effects of type I IFN may be a factor in poor T cell priming, we undertook therapeutic vaccination in two chronically infected chimpanzees during treatment with a direct acting antiviral (DAA) targeting the HCV NS5b polymerase protein. Immunization with genetic vaccines encoding the HCV NS3-NS5b non-structural proteins during DAA treatment resulted in a multifunctional CD8+ T cell response. However, these antiviral CD8+ T cells did not prevent persistent replication of DAA-resistant HCV variants that emerged during treatment. Most vaccine-induced CD8+ T cells targeted class I epitopes that were not conserved in the circulating virus. Exhausted intrahepatic CD8+ T cells targeting conserved epitopes did not expand after vaccination, with a notable exception. A sustained, multi-functional CD8+ T cell response against at least one intact class I epitope was detected in blood after vaccination. Persistence of HCV was not due to mutational escape of this epitope. Instead, failure to control HCV replication was likely caused by localized exhaustion in the liver, where CD8+ T cell expression of the inhibitory receptor PD-1 increased 25-fold compared with those in circulation.

Conclusion

Treatment with DAA during therapeutic vaccination provided transient control of HCV replication and a multifunctional T cell response, primarily against non-conserved class I epitopes. Exhaustion of liver-infiltrating CD8+ T cells that target conserved epitopes may not be averted when DAA therapy fails prematurely due to emergence of resistant HCV variants.

Introduction

Persistence of the hepatitis C virus (HCV) in humans and chimpanzees requires evasion of CD8+ T cell immunity(1–3). CD8+ T cells can provide transient control of virus replication during the acute phase of infection but often fail to prevent HCV persistence because of mutational escape of class I epitopes and/or exhaustion of characterized by loss of antiviral effector functions(1–3). Spontaneous reversal of CD8+ T cell exhaustion in chronic hepatitis C is rare. Exhaustion is mediated in part by expression of receptors like PD-1, TIM-3, 2B4, and CTLA-4 that delivery inhibitory signals to CD8+ T cells upon engagement of their respective ligands(4–9). Antibodies against these inhibitory receptors can restore HCV antigen-driven proliferation of CD8+ T cells in cell culture(4–6, 8, 9). Moreover, some humans(10) and chimpanzees(11) treated with anti-PD-1 antibodies displayed a sharp drop in viremia that may have been associated with recovery of T cell immunity(11).

Various approaches to therapeutic vaccination, including adjuvanted peptides(12–14) and proteins(15, 16), antigen-pulsed dendritic cells(17), and recombinant viruses(18, 19) or DNA plasmids(20), have also been assessed for restoration of T cell immunity in humans with chronic hepatitis C. Early studies were conducted without concurrent suppression of virus replication using type I IFN-based therapies(12, 13, 17, 18, 20). CD8+ T cell activity was detected in the blood of some vaccinated subjects but viremia declined modestly and transiently (usually by 1 log or less), or was unchanged when compared to pre-vaccination values(12, 13, 18). Vaccination while virus replication was suppressed with pegylated type I IFN and ribavirin (pegIFN/RVN) did not noticeably improve induction of HCV-specific cellular immune responses or the outcome of antiviral therapy(14, 15, 19). Why vaccine-induced CD8+ T cells failed to control persistent virus replication in subjects who developed a detectable response is not known.

In this study we undertook therapeutic vaccination of chronically infected chimpanzees during treatment with a direct acting antiviral (DAA) that inhibits function of the HCV polymerase protein. This approach was designed to prime CD8+ T cells while HCV antigen loads were sharply reduced, without the potential for an immunomodulatory impact of type I IFN that can interfere with development of adaptive immune responses. For vaccination we used recombinant adenoviruses (rAd), modified vaccinia virus Ankara (MVA) and a DNA plasmid encoding the HCV NS3-NS5b non-structural proteins that are dominant targets of the T cell response. Priming and boosting with these genetic vaccines elicited strong, durable T cell responses in uninfected chimpanzees(21, 22) and humans(23, 24). Importantly, T cells primed by rAd vectors and boosted with plasmid DNA expanded rapidly after HCV challenge and substantially reduced the magnitude and duration of primary acute phase viremia(22).

Here, we demonstrate that genetic vaccines encoding non-structural proteins NS3-NS5b also prime a multifunctional CD8+ T cell responses in persistently infected chimpanzees during treatment with a direct acting NS5b polymerase inhibitor. The CD8+ T cells were directed predominantly against HCV epitopes that were not conserved in the circulating virus. Most intrahepatic CD8+ T cells recognizing intact epitopes did not expand in blood after vaccination. When an exceptional multifunctional CD8+ T cell response against an intact epitope was observed in blood, it did not prevent resurgent replication of a DAA-resistant HCV variant. Failure to control HCV replication was not due to mutational escape in the class I epitope, but instead to localized exhaustion in the liver associated with very high levels of PD-1 expression.

Materials and Methods

Chimpanzees

Chimpanzees (Pan troglodytes) CH5681 and 5492 were maintained under standard conditions for humane care at the Michale E. Keeling Center for Comparative Medicine and Research, M. D. Anderson Cancer Center. The experiments were reviewed and approved by the Institutional Animal Care and Use Committee of the Michale E. Keeling Center. The study was performed between 2008 and 2010 with funding from NIH grant R37 AI47367.

Persistent infection was established after challenge with the HCV J4/91 genotype 1b virus approximately 3 (5492) and 4 (5681) years before enrollment in this study. Chimpanzee 5492 expressed Patr (Pan troglodytes) class I molecules A*0901, B*0101, and B*2001 as determined by genotyping(25). Chimpanzee 5681 was positive for Patr A*0701, B*0501, and B*0901.

Vaccination and dosing with an NS5b inhibitor

An NS5b inhibitor, the nucleoside analogue 2′-C-methyl-7-deaza-adenosine (MK 0608), was synthesized from a published structure(26). It was administered orally at 2 mg/kg approximately every 24 hours for 37 days (Figure 1). The NS3 to NS5b proteins encoded by the genetic vaccines were based on genotype 1b HCV-BK strain that contained a 3 amino acid substitution in the catalytic domain of NS5b to prevent polymerase activity (designated NSmut). The vaccines were delivered according to the schedules shown in Figure 1. Briefly, a human adenovirus serotype 6 vector (Ad6-NSmut; 1X10¹¹ viral particles) was delivered by direct intramuscular injection at the same time as (5681) or 9 days after (5492) the first dose of the NS5b polymerase inhibitor. Both animals were also boosted by i.m. injection of 1×10¹¹ viral particles of a chimpanzee adenovirus serotype 63 vector (AdCh63-NSmut) as indicated in Figure 1. Chimpanzee 5681 was also boosted with 2×10⁸ plaque forming units (pfu) of a modified vaccinia virus Ankara expressing the same HCV NS3-NS5b gene cassette (MVA-NSmut; Figure 1). Chimpanzee 5492 was boosted 5 times at 2 week intervals with 5mg of plasmid DNA vaccine expressing NSmut by gene electrotransfer (GET) (Figure 1). The electrical field comprised two trains of 100 square bipolar pulses delivered every other second with each pulse lasting 2 ms/phase(27). The electrical field strength was 150 V (constant voltage mode) with current limitation set at 100 mA.

Figure 1. Timing of antiviral treatment and vaccination.

Both animals were dosed daily with MK 0608 for 37 days (gray shaded area) and vaccinated with Ad6, AdCh63, and plasmid DNA (animal 5492) or Ad6, AdCh63 and MVA (animal 5681) on the indicated study day.

ELISpot assay

Peripheral blood and intrahepatic mononuclear cells were tested in an interferon-γ ELISpot assay for reactivity to 2 sets of overlapping peptides spanning the entire NS3-NS5b coding region. One set, derived from the HCV-BK genotype 1b strain of HCV was matched with the NSmut vaccine sequence. The second set of overlapping peptides was matched with the HCV-J4/91 virus used to establish persistent infection. Briefly, 2×10⁵ chimpanzee PBMC and pooled synthetic HCV-BK or HCV-J4 peptides (1 μM of each peptide) were added to duplicate wells of a 96-well microtiter plate pre-coated with antibodies to human interferon-γ (U-Cytech, Utrecht, Netherlands). After 36 hours of incubation ELISpot plates were developed according to the manufacturer’s instructions and spot-forming cells (SFC) were quantified. Background spot formation was determined in wells that received DMSO (the peptide diluent) but not HCV peptides. Background was consistently less than 10 SFC/well. Responses were considered positive when an average of 10 SFC over background were detected in the duplicate wells. CD8+ T cell lines derived from blood after vaccination were used to map dominant class I epitopes as previously described{Kowalski, 1996 #460;Erickson, 2001 #222}.

Tetramer and intracellular cytokine staining assays

To analyze cytokine production by HCV-specific T cells, cryopreserved PBMCs were thawed and rested in media (RPMI 1640 and 10% FCS) containing 5 U/ml Benzonase (EMD) at 37°C for 10 hr. Cells were collected and resuspended in media at 10⁶/ml. Cells (10⁶) were incubated in the presence or absence of peptides (1 μg/ml) at 37°C for 16 hr. GolgiPlug (BD Bioscience) was added after the first hour. At the end of incubation, cells were washed with FACS buffer (PBS, 2.5% FCS and 2% NaN₃) and stained with antibodies recognizing cell surface markers at 4°C for 20 min. After washing with FACS buffer, cells were stained with LIVE/DEAD blue fluorescent reactive dye (Invitrogen) at 4°C for 20 min. Cells were then washed twice with FACS buffer and permeabilized using Cytofix/Cytoperm (BD Bioscience). Intracellular staining for cytokines was carried out in Cytoperm buffer at 4°C for 30 min. Then cells were washed twice with Cytoperm buffer and fixed in 1% paraformaldehyde until analysis. Class I tetramer staining was performed on cryopreserved PBMCs. Tetramers were functional as assessed by staining of CD8+ T cell lines established from the blood and/or liver. PBMC were rested for 2 hours after thawing, washed, and incubated with class I tetramers at 4°C for 30 min. After washing with FACS buffer, cells were stained with antibodies against surface markers and LIVE/DEAD blue fluorescent reactive dye sequentially as mentioned above, and fixed in 1% paraformaldehyde.

Antibodies

Fluorphore-conjugated monoclonal antibodies from BD Bioscience (CD3-V500, CD8-V500, CD3-Alexa 700 and TNF-α-PE-Cy7), Biolegend (CD14-FITC, CD16-FITC, CD19-FITC, CD14-PerCP/Cy5.5, CD16-PerCP, CD19-PerCP, IFN-γ-Pacific Blue, PD-1-PerCP/Cy5.5, IL-2-APC and CD4-Alexa 700) and eBioscience (CD127-V450) and Invitrogen (LIVE/DEAD blue fluorescent reactive dye and CD8-Qdot 605) were used in these studies.

Flow cytometry

Cells were analyzed by flow cytometry using an LSRII instrument (BD Immunocytometry). Data analysis and presentation of distributions was performed using FlowJo (v.9.2, TreeStar). Dead cells and CD14⁺/CD16⁺/CD19⁺ cells were excluded from the analysis. For cytokine production analysis, between 500,000–1,500,000 events were acquired. A response was considered positive when the number of peptide-stimulated cells that produced a cytokine was more than twice that of cells not stimulated with peptide, and the value after background subtraction was at least 0.003%. For tetramer analysis, between 1,000,000–2,000,000 events were acquired. Background for tetramer staining was less than 0.001%.

Characterization of intrahepatic CD8+ T cells

To identify HCV-specific CD8+ T cells in liver before vaccination, mononuclear cells were isolated from homogenized liver tissue that was obtained via transcutaneous biopsy. CD8+ T cells positively isolated with paramagnetic beads were co-cultured with irradiated (5000R) autologous PBMC pulsed with HCV peptides. After 2–3 weeks of culture, cells were expanded with anti-CD3 antibodies and irradiated human PBMC. Three to 4 weeks later, cells were tested for HCV specificity by IFN-γ ELISpot(28). Cell lines derived from these cultures were used to map cognate class I HCV epitopes as described.

Tetramer analysis

Peripheral blood mononuclear cells (PBMC) were incubated with Patr class I tetramers for 30 minutes at 4°C. After co-staining with antibodies to CD8, CD3, CD127, and PD-1 for 20 minutes at 4°C, cells were fixed and analyzed by flow cytometry. Live and dead cells were discriminated with an amine-reactive viability dye. Cells expressing CD4, CD14, CD16, and CD19 were gated out of the analysis. Intrahepatic mononuclear cells from chimpanzee 5681 were isolated from needle biopsy specimens. They were incubated immediately (i.e. without ex vivo culture) with the Patr class I NS3₁₅₆₅ tetramer and co-stained with antibodies to cell surface antigens as described above for PBMC. HCV-specific CD8+ T cells were then visualized with a Becton Dickinson LSR II flow cytometer.

Virus Sequencing

RNA extracted from chimpanzee serum (100 μl) was reverse transcribed using random hexamers. cDNA was amplified with nested sets of PCR primers that flanked the Patr class I epitopes. PCR primers are described in supplementary Figure 1 and supplementary Table 1. Nested-PCR products containing HCV inserts were cloned and sequenced by the Laboratory for Genomics & Bioinformatics (University of Oklahoma Health Sciences Center).

Results

Therapeutic vaccination was undertaken in two chimpanzees approximately 3 (5492) or 4 (5681) years after chronic infection was established with the genotype 1b HCV-J4 virus. Chimpanzee 5492 was first treated with the MK 0608 NS5b inhibitor at study day 0 and vaccinated 9 days later with the Ad6-NSmut vector (Figure 1). Viremia was below the limit of detection between treatment study days 2 and 15 except for a small rebound at day 7 (Figure 2A). The chimpanzee was boosted with AdCh63-NSmut at day 22, after breakthrough replication of an HCV variant carrying the NS5b S282T mutation (data not shown) that confers resistance to MK 0608 antiviral(29). HCV-specific T cell responses against the HCV-J4 and NSmut peptide pools were detected at day 36 despite resurgent HCV replication. Five immunizations at 2 week intervals with a plasmid DNA vaccine resulted in a further gradual increase in HCV-specific T cell frequencies (Figure 2A). T cell frequencies against the NSmut and HCV-J4 pools were quite stable after the last boost, with no sharp contraction of the response over 4 months of follow-up (Figure 2A). The response was somewhat stronger against the vaccine (NSmut) peptide pools (Figure 2A), but two pools (NS3-2 and NS5b-2) accounted for most of the response against HCV-J4 and NSmut (Figure 2A).

Figure 2. Patterns of HCV replication and T cell immunity after treatment with MK 0608 and therapeutic vaccination.

Viremia is indicated by the solid black line and treatment with MK 0608 is represented by the gray shaded zone. Bar graphs represent the magnitude of the IFN-γ ELISpot response when PBMC were stimulated with peptides matched to the HCV-J4 challenge virus (hatched bars) or the NSmut vaccine sequence (solid bars) for chimpanzees 5492 (2A) and 5681 (2B). Pie charts show the breadth of the ELISpot response against individual HCV-J4 and NSmut peptide pools at the peak of the response (day 92 for animal 5492 and day 21 for animal 5681).

HCV-specific T cells also expanded in animal 5681 using a different treatment regimen (Figure 1). The Ad6-NSmut vector and first dose of NS5b inhibitor were co-delivered at study day 0 when viremia was approximately 10⁶ IU/ml plasma (Figure 2B). Viremia was reduced approximately 1000-fold at day 14 when the MVA-NSmut boost was delivered. T cell activity against the HCV-J4 and NSmut peptide pools peaked at day 21, after a substantial rebound in virus replication by a variant carrying the S282T NS5b resistance mutation (Figure 2B). T cell activity declined sharply after day 21 and was not boosted by further vaccination with MVA-NSmut or AdCh63-NSmut vectors (Figure 2B). All 6 HCV-BK and J4 nonstructural peptide pools were recognized at day 21 in approximately equal proportion. The response to one NS3 pool (NS3-2) accounted for approximately half of the activity against the HCV-J4 and NSmut sequences (Figure 2B).

Virus-specific CD8+ T cells expanded from the liver of both persistently infected chimpanzees before vaccination and DAA treatment were used to map escaped and intact class I epitopes in the HCV non-structural proteins. Chimpanzee 5492 had intrahepatic CD8+ T cells against 8 different non-structural epitopes after 3 years of chronic infection (Table 1A). Four of the epitopes were intact in the circulating virus and conserved in the HCV-BK NSmut vaccine sequence (Table 1A). The other 4 epitopes developed escape mutations (Table 1A) that impaired recognition of the circulating virus (supplementary Figure 2). CD8+ T cell lines expanded from the liver of chimpanzee 5681 before vaccination targeted 6 discrete epitopes in NS3, NS4, and NS5 (Table 1B). Three of the epitopes were intact in the circulating virus and conserved in the vaccine (Table 1B). The other 3 epitopes acquired escape mutations (Table 1A and supplementary Figure 2). CD8+ T cells that recognized epitopes unique to the HCV-BK NS-mut vaccine sequence were detected after immunization of the animals. Two CD8+ T cell lines established from the blood of chimpanzee 5492 after vaccination recognized HCV-BK vaccine-specific sequences (Table 1A) and were poorly cross-reactive for the circulating HCV-J4 virus (supplementary Figure 3). Chimpanzee 5681 recognized 3 vaccine-specific class I epitopes after immunization (Table 1B and supplementary Figure 3).

Table 1.

Class I restricted epitopes recognized by CD8+ T cells from animals 5492 (A) and 5681 (B). Amino acid sequence (single letter code) of each epitope in the HCV-J4 challenge virus (inoculum), the HCV-BK NSmut vaccine (vaccine), and circulating virus at the time of treatment with the vaccine and NS5B inhibitor are shown. Dashes indicate identity with the inoculum sequence. Escaped and intact HCV-J4 epitopes were mapped using CD8+ T cell lines established from liver before treatment. Vaccine epitopes were mapped using CD8+ T cells generated from blood after immunization.

A. 5492		Amino Acid Sequence
Epitopes		Inoculum (HCV-J4)	Vaccine (HCV-BK)	Circulating
intact	NS31349	ATPPGSVTV	---------	---------
	NS31578	KQAGDNFPYLV	-----------	-----------
	NS41943	RVTQILSSLTI	-----------	-----------
	NS52054	NTWHGTFPI	---------	---------
escaped	NS31560	SVFTGLTHI	---------	------S--
	NS41779	STLPGNPAI	---------	------S--
	NS51992	KTWLQSKLL	---------	-I-------
	NS52423	YTWTGALI	--------	-------V
vaccine	NS31419	RGLDVSVIPPI	---------T-	-----------
	NS31444	FTGDFDSVI	Y--------	---------

B. 5681		Amino Acid Sequence
Epitopes		Inoculum (HCV-J4)	Vaccine (HCV-BK)	Circulating
intact	NS3 1377	GKAIPIEA	--------	--------
	NS31565	LTHIDAHFL	---------	---------
	NS41728	FKQKALGL	--------	--------
	NS51991	FKTWLQSKL	---------	---------
escaped	NS5 2804	LTRDPTTPL	---------	--------I
	NS52847	THFFSILL	--------	-------I
vaccine	NS3 1101	YTNVDLDLV	-----Q---	---------
	NS41801	LTTQNTLLF	----S----	---------
	NS52279	RKFPSALPI	K---A-M--	---------

Class I tetramers were used to visualize expansion of CD8+ T cells targeting escaped, intact, and vaccine-specific epitopes after vaccination. CD8+ T cells targeting the escaped epitopes increased in frequency by day 36 and remained stable in blood through day 387, almost 10 months after the last DNA vaccine boost (Figure 3A). Tetramers were available for CD8+ T cells targeting 3 of the 4 intact epitopes. Two of these populations were not detected in blood through day 92 when the last DNA vaccine boost was administered (Figure 3A). Somewhat delayed expansion of NS5₂₀₅₄-specific T cells was observed by day 92 despite persistent replication of HCV but this response was not sustained (Figure 3A). The two vaccine-specific CD8+ T cell populations (targeting epitopes NS3₁₄₁₉ and NS3₁₄₄₄; Table 1A) were visible by tetramer staining at day 36, increased sharply with subsequent DNA vaccine boosting, and remained stably present in blood through study day 387 (Figure 3A). Similar expansion of CD8+ T cells in the blood of animal 5681 was confirmed by tetramer analysis. CD8+ T cells targeting escaped epitope NS5₂₈₄₇ were present in blood after immunization (Figure 3B), as were CD8+ T cells targeting three vaccine specific epitopes (data not shown). Three of the 4 CD8+ T cells targeting intact epitopes that were present in liver before vaccination could be tracked in blood after vaccination using class I tetramers. Expansion of CD8+ T cells targeting 2 of the epitopes (NS3₁₃₇₇ and NS4₁₇₂₈) was very weak and/or transient (Figure 3B). An exceptional response was detected against the intact epitope NS3₁₅₆₅. CD8+ T cell targeting this epitope were present at high frequency at day 21, after HCV replication had substantially rebounded (Figure 3B). Frequencies declined with time, but the CD8+ T cells were still visualized in blood almost 3 months after the last boost with AdCh63 vector.

Figure 3. Visualization of CD8+ T cells that expanded in blood after vaccination.

Frequency of individual circulating HCV-specific CD8+ T cells that target escaped, intact, or vaccine-specific class I epitopes as described in Table 1. (3A) T cells were visualized in PBMC with class I tetramers 26 days before (-26) and 36, 92, 141, and 387 after vaccination of chimpanzee 5492. (3B) A similar tetramer analysis was carried out for animal 5681 49 days before (−49) and 21, 49, and 122 days after vaccination. Tetramers used to detect CD8+ T cells targeting escaped and intact epitopes carried peptides matched to the HVC-J4 inoculum sequence. Vaccine-specific tetramers carried epitopes matched with NSmut sequence.

Sequencing of the persistent virus from both animals revealed that none of the intact class I epitopes developed escape mutations after vaccination (Figure 4). A late day 84 change at position 9 of the NS3₁₃₇₇ epitope (A1384V) was observed in animal 5681 (Figure 4), but it did not alter CD8+ T cell recognition (supplementary Figure 4). The absence of escape was most notable for epitope NS3₁₅₆₅ that was a dominant target of functional, circulating CD8+ T cells in persistently infected chimpanzee 5681 (Figure 4).

Figure 4. Changes in the sequence of intact class I epitopes in circulating HCV-J4 virus after vaccination.

HCV RNA isolated from plasma at the indicated study day was reverse transcribed and clonal sequencing was performed as described in the Methods and Materials section.

We next assessed the function of circulating CD8+ T cells targeting escaped, intact, and vaccine-specific epitopes. Representative CD8+ T cell responses from animal 5492 that recognized vaccine-specific and escaped epitopes had a similar functional profile. They co-produced IFN-γ and TNF-α after antigen stimulation and approximately half of them also secreted IL-2 (Figure 5). Expression of CD107a, a marker of cytotoxic degranulation, was also present on almost all HCV-specific CD8+ T cells that produced IFN-γ (Figure 5). Analysis of most CD8+ T cells targeting intact epitopes was not possible because they were absent from blood or present at low frequency. A small subset of NS5₂₀₅₄-specific CD8+ T cells from chimpanzee 5492 appeared to co-produce TNF-α and IL-2 after antigen stimulation (Figure 5). NS3₁₅₆₅- specific CD8+ T cells present in the blood of animal 5681 were multi-functional. They co-produced IFN-γ, TNF-α and IL-2 and expressed CD107a after antigen stimulation (Figure 5).

Figure 5. Function of circulating representative CD8+ T cells targeting escaped, vaccine, and intact epitopes from two vaccinate chimpanzees.

CD8+ T cells in circulation at or near the peak of the vaccine-induced CD8+ T cell response were assessed by intracellular staining for production of TNF-α and IFN-γ (top row), IL-2 and TNF-α (middle row) or CD107a and IFN-γ (bottom row).

Finally, PD-1 and CD127 expression was compared on circulating CD8+ T cells from animal 5492 at day 92 when the last dose of DNA vaccine was administered and at day 387 (Figure 6A). CD8+ T cells targeting vaccine (NSmut)-specific epitopes expressed more CD127 and less PD-1 at day 347, a transition that is expected as activated vaccine-primed effector cells transition to immunological memory. All CD8+ T cells targeting escaped HCV-J4 epitopes underwent a similar change in PD-1 and CD127 phenotype after vaccination. A similar transition to lower PD-1 and higher CD127 expression was also observed for CD8+ T cells from animal 5681 that recognized escaped or vaccine-specific epitopes. The frequency of CD8+ T cells against intact epitopes in chimpanzees 5492 (NS5₂₀₅₄) and 5681 (NS3₁₃₇₇) was too low for phenotypic analysis. CD8+ T cells from animal 5681 targeting intact epitope NS3₁₅₆₅ also contracted after vaccination but remained detectable in blood at later time points. CD127 expression was stable between days 21 and 122 but PD-1 levels decreased substantially (Figure 6B). A difference in PD-1 and CD127 expression was observed between blood and liver populations at day 122. Few NS3₁₅₆₅-specific intrahepatic CD8+ T cells expressed CD127 when compared with circulating populations (Figure 6B). Expression of PD-1 on intrahepatic NS3₁₅₆₅-specific CD8+ T cells (MFI 15597) was approximately 25 fold higher than on circulating populations (MFI 655) at this time point. In summary, these results indicate that CD8+ T cells targeting escaped epitopes are responsive to HCV antigens delivered by vaccination. CD8+ T cells targeting intact epitopes are either not responsive to vaccination or, less commonly, migrate to liver where they acquire very high levels of PD-1 associated with functional exhaustion.

Figure 6. Changes in expression of (A) CD127 and (B) PD-1 after vaccination.

(A). PBMC from chimpanzee 5492 were co-stained with the indicated class I tetramer to visualize changes in CD127 and PD-1 expression between the day 92 (date of last DNA vaccine boost) and day 387. Changes were measured for CD8+ T cells targeting vaccine and escaped class I epitopes. (B). Expression of CD127 and PD-1 was monitored on CD8+ T cells targeting the intact NS3₁₅₆₅ epitope in blood of chimpanzee 5861 at study days 21 and 122, and in liver at day 122.

Discussion

NSmut genetic vaccines induce strong, durable T cell immunity in HCV-naïve chimpanzees(21, 22) and humans(23, 24). In vaccinated chimpanzees, HCV challenge rapidly recalled T cell responses that appeared to reduce the magnitude and duration of acute phase viremia(22). We therefore assessed immunogenicity of similar genetic vaccines encoding non-structural HCV proteins in two chimpanzees with persistent HCV infection, in combination with DAA treatment. CD8+ T cells with antiviral effector functions were induced in both animals, but could not contain break-through replication of a virus with resistance to the NS5b direct acting antiviral.

CD8+ T cells are primed in humans by therapeutic HCV vaccination but they have only limited or no impact on persistent virus replication(12, 13, 17, 18, 20). Use of the chimpanzee model allowed us to define mechanisms of CD8+ T cell failure after therapeutic vaccination. Detailed mapping of targeted class I epitopes demonstrated that most CD8+ T cells expanded by vaccination did not recognize the circulating persistent virus. Approximately half of the CD8+ T cells detected in blood targeted vaccine-specific epitopes, even though the N3-NS5b proteins encoded by the NSmut vaccine and HCV-J4 challenge virus differed at only 77 of 2000 amino acids. All other HCV-specific CD8+ T cells that expanded in blood were present in the liver before vaccination. The presence of escape mutations in class I epitopes of the circulating virus was strongly predictive of a response to vaccination. CD8+ T cells targeting escaped or vaccine-specific epitopes had effector functions and underwent expansion and contraction after re-exposure to antigen despite persistent replication of the virus. Over time they also transitioned from effector to memory status as revealed by a phenotypic shift in expression of PD-1 and CD127. Earlier studies in HCV infected humans demonstrated lower levels of PD-1(7, 30) and higher CD127(31) on CD8+ T cells targeting escaped versus intact epitopes(7, 30). They also more efficiently suppressed HCV replication in a cell culture model(32). Use of the vaccine in this study provides direct experimental evidence that they remain responsive to antigen stimulation, activate to become effector cells, and transition towards a memory phenotype despite ongoing replication of the virus.

As expected, CD8+ T cell population cells targeting intact epitopes were present in liver before therapeutic vaccination(1–3). Most did not expand after vaccination or were present transiently at very low frequency. This is consistent with a state of profound exhaustion characterized by expression of multiple co-inhibitory receptors, an absence of effector functions, and a predisposition to programmed cell death in the liver(1–3). Moreover, we observed no apparent broadening of the response to new intact epitopes beyond those already targeted by CD8+ T cells that pre-existed in the persistently infected liver. This indicates a generally effective block on T cell priming against intact epitopes of the circulating virus. The one exception was a response against intact epitope NS3_1565. These CD8+ T cells did not result in virus control, nor did they select for escape mutations in the NS3₁₅₆₅ epitope after vaccination. Importantly, we were able to establish that these functional CD8+ T cells are present in the persistently infected liver before and after vaccination, and so failure to migrate to the site of virus replication does not explain the absence of an antiviral effect. It was not possible to assess the effector function of intrahepatic NS3₁₅₆₅-specific CD8+ T cells because of limited liver sample size. Unlike CD8+ T cells in blood, however, those that infiltrated liver had an exhausted phenotype characterized by very high levels of PD-1 and low CD127. This split response in liver versus blood, although uncommon, suggests that high antigen load does not always lead to systemic CD8+ T cell exhaustion. In this one instance, CD8+ T cells probably failed after they infiltrated the persistently infected liver. Notably, NS3₁₅₆₅-specific CD8+ T cells expressed very high levels of PD-1 when compared with those in circulation (Figure 6B). It is possible that engagement of PD-L1 ligand, which is expressed at high levels in the liver of chimpanzees with chronic hepatitis C{Park, 2012 #687}, resulted in failure of effector activity at the site of virus replication.

This study does have limitations. The vaccine regimen differed for the two chimpanzees, so we cannot draw conclusions about the relative effectiveness of boosting strategies with plasmid DNA versus viral vectors. Nonetheless, common mechanisms of CD8+ T cell failure were observed in both animals, including an inability to recognize epitopes in the circulating virus and (less commonly) apparent exhaustion after migration from blood to liver. Also, MK0608 did not suppress HCV-J4 replication through the entire cycle of vaccine priming and boosting. Rapid emergence of drug-resistant HCV variants in both animals was not expected as MK608 suppressed replication of other genotype 1 HCV strains in chimpanzees through 37 days of treatment(26). It is likely that the HCV-J4 challenge virus contained pre-existing variants with the NS5b S282T substitution known to confer resistance to this nucleoside analogue. Because antiviral resistance developed during vaccination, we cannot be certain that CD8+ T cells targeting intact epitopes are permanently unresponsive to antigen. One recent study of humans subjects cured with DAA suggests that some of CD8+ T cells become responsive to antigen stimulation in cell culture(33). However, we observed that rechallenge of a chimpanzee with HCV 2 years after DAA cure led to expansion of intrahepatic CD8+ T cells targeting escaped but not intact epitopes(34). This observation, combined with the failure to restore CD8+ T cell activity through vaccination in this study, might indicate that most populations targeting intact epitopes are not readily rescued from exhaustion. In summary, vaccine antigen delivery, processing, and class I presentation was effective in individuals with chronic hepatitis C because CD8+ T cells targeting vaccine-specific and escaped class I epitopes expanded readily in blood. Furthermore, our observations also provide a likely explanation for the absence of a virologic response in chronically infected humans with vaccine-primed CD8+ T cell responses. Failure of the vaccine to promote recovery of most CD8+ T cell populations that target intact epitopes, even when virus replication was transiently controlled with a DAA that does not have the immunodulatory effects of type I IFN, remains an area of concern. The vaccine approach employed here does substantially suppress HCV replication during the acute phase of infection in chimpanzees and may prevent persistent infection. Further study is required to determine if this or similar preventive vaccines will be sufficient to restore protective CD8+ T cell immunity in humans who cured with DAA but remain at risk for reinfection with HCV.

Abbreviations in this paper

B-LCL

B lymphoblastoid cell lines

DAA

direct acting antiviral

HCV

hepatitis C virus

IFN

interferon

Patr

Pan troglodytes

PBMC

peripheral blood mononuclear cell

PD-1

programmed cell death 1

MFI

mean fluorescence intensity

SFC

spot forming cells