Antagonistic expression of hepatitis C virus and alpha interferon in lymphoid cells during persistent occult infection

Abstract

Detection of residual HCV in individuals with SVR after treatment of CHC can be significantly heightened by analyzing ex vivo mitogen-activated T and B lymphocytes and applying sensitive nucleic acid amplification assays. However, it remained unknown if synergistic activation of lymphocytes and monocytes would further augment HCV detection, if viral replication becomes universally upregulated in treated cells, and if examining sequential sera and lymphoid cells would improve detection of occult infection. Using paired sera and lymphoid cells collected 1 year apart from 17 individuals with normal liver enzymes for up to 72 months after SVR, it was found that simultaneous activation of lymphocytes and monocytes enhanced identification of silent HCV infection and revealed that in some cases monocytes were the principal immune cell type where HCV persisted. Testing of serial samples further increased detection of occult infection. Ultimately, by combining the above two approaches, all individuals with SVR were found to be silent carriers of HCV. Clonal sequencing revealed HCV variations in sera and lymphoid cells and evolution of viral genomes confirming ongoing virus replication. Surprisingly, similar to those with CHC, naive lymphoid cells from some individuals carried approximately 10(3) HCV copies/microg total RNA. HCV loads in naive lymphoid cells predetermined the outcome of ex vivo stimulation with respect to upregulation or inhibition of HCV replication. HCV RNA levels in occult infection were inversely proportional to the expression of IFNalpha and IFN-inducible MxA, but not to IFNgamma or tumour necrosis factor alpha in naive and mitogen-treated lymphoid cells.

Figure 1

Detection of hepatitis C virus (HCV) RNA in paired serum and peripheral blood mononuclear cells (PBMC) samples collected a year apart from an individual with persistent occult HCV infection continuing after apparent complete recovery from hepatitis C. Samples from Patient 17 were obtained at 36 and 48 months (mo) after sustained virological response. RNA extracted from 250 μL serum or ∼2 × 10⁶ naive (mitogen‐untreated) PBMC was amplified by nested reverse transcription‐polymerase chain reaction using 5′ untranslated region (5′‐UTR)‐ or E2‐specific primers. The specificity of the amplicons was verified by nucleic acid hybridization. RNA extracted from an equivalent amount of serum and the same number of naive PBMC from a patient with chronic hepatitis C (CHC‐1) were used as a positive control. Contamination controls included water added instead of cDNA and amplified by direct (DW) and nested (NW) reactions and mock (M) treated as test RNA. Positive signals showed the expected 346‐bp (direct) and 244‐bp (nested) 5′‐UTR fragments, as well as the 476‐bp (direct) and 443‐bp (semi‐nested) E2 amplicons. Numbers under the panel represent relative densitometric units given by hybridization signals.

Figure 2

Hepatitis C virus (HCV) sequences identified in sera and lymphoid cells from individuals with occult HCV infection continuing after a sustained virological response. HCV 5′ untranslated region (5′‐UTR) sequences amplified from serum and peripheral blood mononuclear cells obtained at the second collection from Case 19 and from both collections from Case 20 were cloned, sequenced bidirectionally, and the sequences aligned with the prototype HCV genotype 1b sequence, HCV‐1J [27]. Sequences of 6 to 8 clones per test sample are shown. Nucleotides in the cloned sequences identical with the HCV reference (top line) are shown as dots and differences are identified by letters.

Figure 3

Upregulation of hepatitis C virus (HCV) RNA expression in lymphoid cells from individuals with occult infection following their ex vivo treatment with mitogen‐cytokine cocktails. RNA extracted from lymphoid cells which had been either untreated (UT) or cultured in the presence of C5 or C5L were analyzed for HCV RNA positive (a) and negative (b) strands by strand‐specific reverse transcription‐polymerase chain reaction/nucleic acid hybridization. Peripheral blood mononuclear cells samples from a patient with chronic hepatitis C (CHC‐2) treated under the same conditions as test cell samples were used as positive controls. Contamination controls were as described in the legends to Fig. 1. Numbers under the panel represent relative densitometric units given by hybridization signals.

Figure 4

Inhibition of hepatitis C virus (HCV) replication in lymphoid cells from patients with occult HCV infection and high HCV RNA load after treatment with mitogens. Peripheral blood mononuclear cells with HCV RNA load of ∼10³ vge/μg RNA were left untreated (UT) or cultured for 72 h in the presence of either C5 or C5L. RNA was extracted and evaluated for HCV RNA positive (a) and negative (b) strands or glyceraldehyde‐3‐phosphate dehydrogenase (c) by reverse transcription‐polymerase chain reaction/nucleic acid hybridization assays. Contamination controls were as described in the legend to Fig. 1. Numbers under the panel represent relative densitometric units of hybridization signals.

Figure 5

Expression of interferon 5 α (IFN5α) and MxA in naive and mitogen‐treated lymphoid cells which carried initially either low or high loads of hepatitis C virus (HCV). Lymphoid cells from two healthy persons, individuals with occult HCV infection carrying undetectable or low levels of HCV RNA in these cells (Cases 13 and 6), individuals with occult HCV infection and high HCV RNA load in the cells (Cases 18 and 19), and a patient with chronic hepatitis C (CHC‐2) were left untreated (UT) or treated with C5L as described in Materials and methods. RNA was examined for IFN5α, MxA and glyceraldehyde‐3‐phosphate dehydrogenase (GAPDH) by specific reverse transcription‐polymerase chain reaction/nucleic acid hybridization assays. Ten‐fold serial dilutions of rIFNα and MxA (rMxA) gene fragments were used for enumeration of the test samples. Recombinant GAPDH gene fragment (rGAPDH) was amplified in parallel and used as a loading control. Contamination controls were described in legends to Fig. 1. Numbers under the panel represent relative densitometric units given by hybridization signals.