Pathogen-Associated Molecular Pattern Recognition of Hepatitis C Virus Transmitted/Founder Variants by RIG-I Is Dependent on U-Core Length

Abstract

Despite the introduction of direct-acting antiviral (DAA) drugs against hepatitis C virus (HCV), infection remains a major public health concern because DAA therapeutics do not prevent reinfection and patients can still progress to chronic liver disease. Chronic HCV infection is supported by a variety of viral immune evasion strategies, but, remarkably, 20% to 30% of acute infections spontaneously clear prior to development of adaptive immune responses, thus implicating innate immunity in resolving acute HCV infection. However, the virus-host interactions regulating acute infection are unknown. Transmission of HCV involves one or a few transmitted/founder (T/F) variants. In infected hepatocytes, the retinoic acid-inducible gene I (RIG-I) protein recognizes 5' triphosphate (5'ppp) of the HCV RNA and a pathogen-associated molecular pattern (PAMP) motif located within the 3' untranslated region consisting of poly-U/UC. PAMP binding activates RIG-I to induce innate immune signaling and type 1 interferon antiviral defenses. HCV poly-U/UC sequences can differ in length and complexity, suggesting that PAMP diversity in T/F genomes could regulate innate immune control of acute HCV infection. Using 14 unique poly-U/UC sequences from HCV T/F genomes recovered from acute-infection patients, we tested whether RIG-I recognition and innate immune activation correlate with PAMP sequence characteristics. We show that T/F variants are recognized by RIG-I in a manner dependent on length of the U-core motif of the poly-U/UC PAMP and are recognized by RIG-I to induce innate immune responses that restrict acute infection. PAMP recognition of T/F HCV variants by RIG-I may therefore impart innate immune signaling and HCV restriction to impact acute-phase-to-chronic-phase transition.

Importance: Recognition of nonself molecular patterns such as those seen with viral nucleic acids is an essential step in triggering the immune response to virus infection. Innate immunity is induced by hepatitis C virus infection through the recognition of viral RNA by the cellular RIG-I protein, where RIG-I recognizes a poly-uridine/cytosine motif in the viral genome. Variation within this motif may provide an immune evasion strategy for transmitted/founder viruses during acute infection. Using 14 unique poly-U/UC sequences from HCV T/F genomes recovered from acutely infected HCV patients, we demonstrate that RIG-I binding and activation of innate immunity depend primarily on the length of the uridine core within this motif. T/F variants found in acute infection contained longer U cores within the motif and could activate RIG-I and induce innate immune signaling sufficient to restrict viral infection. Thus, recognition of T/F variants by RIG-I could significantly impact the transition from acute to chronic infection.

FIG 1

RIG-I binds T/F pU/UC RNA. (A) T/F pU/UC RNA on denaturing agarose gel. (B) Electrophoretic mobility shift assays using 6 pmol RNA incubated with recombinant RIG-I protein. (C) Densitometric analysis was performed to measure the relative quantities of RNA from the shifted region and the unshifted region; data are plotted as percent RNA shifted. (D) Effective millimolar concentrations of RIG-I at which 10%, 50%, or 90% of RNA was shifted in the gel. An asterisk (*) denotes an in vivo T/F pU/UC sequence; “-V” denotes an in vitro U-core variant.

FIG 2

Differential ATPase activity of RIG-I bound to T/F PAMP RNA. Graphs show the ATPase activity of 5 pmol purified RIG-I protein incubated with increasing amounts of RNA. Levels of free phosphate released upon incubation with Con1 pU/UC RNA (blue circles), X RNA (red squares), and individual T/F pU/UC RNA (black triangles) are shown as averages of the results of three or more experiments ± standard deviations (SD). EC₅₀ and EC₉₀ values for each RNA were calculated and are shown in the adjacent table.

FIG 3

Limited trypsin proteolysis of 30 pmol purified RIG-I protein with increasing amounts (1 to 5 pmol) of RNA from T/F pU/UC regions. Following 15 min of protein digestion, products were run on a 12% SDS-PAGE gel and stained with colloidal Coomassie blue. (A) RIG-I (30 pmol) was incubated with increasing amounts of RNA (1 to 5 pmol). FL, full length. (B) RIG-I (30 pmol) was incubated with 3 pmol RNA for densitometric analysis of the quantity of RIG-I RD and CARD-containing helicase (heli-card) accumulation. The results of densitometric analysis of RD accumulation as a percentage of accumulation following incubation with Con1 pU/UC RNA are shown in the table at the bottom of the figure.

FIG 4

T/F HCV pU/UC sequences differentially activate RIG-I signaling. Huh7 cells (A and C) or Huh7.5 cells (B and D) were transfected with purified pU/UC RNA from the indicated T/F genomes, JFH1, or X RNA, and cells were harvested 18 h later for immunoblot analysis (A and B) or RT-PCR assay (C and D).

FIG 5

The innate immune response to pU/UC motifs restricts HCV replication. (A) Huh7 cells were transfected with 15 pmol RNA from the indicated T/F genome pU/UC regions or the X region or were mock transfected in duplicate. At 18 h later, the cells were infected with HCV JFH1 (MOI = 1). Cells were harvested 48 h postinfection, and viral RNA was quantified by a RT-quantitative (qPCR) assay. Data are shown as means of the results of two independent experiments (± SD). (B) Quantification of infectious virus from supernatants. Supernatant from infected cells was isolated and infectious virus measured by focus-forming-unit (ffu) assay.

FIG 6

RIG-I stimulatory activity of pU/UC sequences correlates with U composition and the length of the U core. Each T/F pU/UC RNA was assigned an activity rank based on the level of RIG-I activation measured in biochemical assays and the innate immune response elicited in vitro. Two-tailed Spearman nonparametric correlation was then calculated for all RNAs with an activity score based on the variable characteristics. α = 0.0083.

FIG 7

Diagram (1) representing the HCV genome 3′ UTR. Determinants of HCV PAMP activity are indicated. The sequence shown is that of the Con1 pU/UC region. The underlined and bolded region indicates a statistically significant correlation with RIG-I activity as shown in Fig. 6. The underlined sections were shown to correlate with the overall RIG-I activation rank as measured by all assays performed; the section in bold correlates with innate immune signaling and innate immune priming in Huh7 cell transfection assays.