Potential conflict of interest: Nothing to report.
This article first published online 16 December 2011. It has since been updated. The abbreviations in Table 1 have been corrected.
Several enveloped viruses including human immunodeficiency virus type 1 (HIV-1), cytomegalovirus (CMV), herpes simplex virus 1 (HSV-1), Ebola virus, vaccinia virus, and influenza virus have been found to incorporate host regulators of complement activation (RCA) into their viral envelopes and, as a result, escape antibody-dependent complement-mediated lysis (ADCML). Hepatitis C virus (HCV) is an enveloped virus of the family Flaviviridae and incorporates more than 10 host lipoproteins. Patients chronically infected with HCV develop high-titer and crossreactive neutralizing antibodies (nAbs), yet fail to clear the virus, raising the possibility that HCV may also use the similar strategy of RCA incorporation to escape ADCML. The current study was therefore undertaken to determine whether HCV virions incorporate biologically functional CD59, a key member of RCA. Our experiments provided several lines of evidence demonstrating that CD59 was associated with the external membrane of HCV particles derived from either Huh7.5.1 cells or plasma samples from HCV-infected patients. First, HCV particles were captured by CD59-specific Abs. Second, CD59 was detected in purified HCV particles by immunoblot analysis and in the cell-free supernatant from HCV-infected Huh7.5.1 cells, but not from uninfected or adenovirus serotype 5 (Ad5) (a nonenveloped cytolytic virus)-infected Huh7.5.1 cells by enzyme-linked immunosorbent assay. Last, abrogation of CD59 function with its blockers increased the sensitivity of HCV virions to ADCML, resulting in a significant reduction of HCV infectivity. Additionally, direct addition of CD59 blockers into plasma samples from HCV-infected patients increased autologous virolysis. Conclusion: Our study, for the first time, demonstrates that CD59 is incorporated into both cell line-derived and plasma primary HCV virions at levels that protect against ADCML. This is also the first report to show that direct addition of RCA blockers into plasma from HCV-infected patients renders endogenous plasma virions sensitive to ADCML. (HEPATOLOGY 2012)
The complement system plays a central role in both innate and adaptive immune defenses against infectious pathogens. This system can be activated by three distinct pathways known as the classical, alternative, and mannose-biding lectin pathways.1 Activation of the complement system is tightly regulated by regulators of complement activation (RCA), which restrict host self-complement activation thereby preventing self-injury.1 Several enveloped viruses including human immunodeficiency virus type 1 (HIV-1), cytomegalovirus (CMV), herpes simplex virus 1 (HSV-1), Ebola virus, vaccinia virus, and influenza virus have been shown to escape antibody-dependent complement-mediated lysis (ADCML) by incorporating and hijacking host RCA proteins into the viral envelopes (Env).2-6 The presence of RCA proteins including CD46, CD55, and CD59 on the external surface of the viral Env provides resistance to ADCML of virions. These findings provide a possible molecular explanation for why certain human pathogenic viruses are not lysed by ADCML, in spite of potent complement activation and high levels of viral-specific antibodies (Abs) present in the circulation of infected persons. For example, patients infected with HIV-1 are well known to mount a vigorous and sustained Ab response at all stages of viral infection,7 yet fail to control virus proliferation or protect themselves from developing AIDS.7 Recent studies have suggested that HIV-1 resistance to ADCML is dependent on RCA molecules, particularly CD55 and CD59, two glycosylphosphatidylinositol-anchored proteins (GPI-APs).2, 5, 6 These molecules either interfere with the complement activation sequence or inhibit the activation of the terminal complement components, thus halting the formation of the membrane attack complex (MAC) and preventing ADCML. Studies in vitro have shown that HIV-1 virions incorporate CD55 and CD59 on the viral surface at levels that protect against ADCML triggered by anti-HIV-1 Env Abs or serum/plasma from HIV-1-infected patients.2, 5, 6 Similarly, these two proteins also provide resistance for HIV-1-infected cells from ADCML by sera from HIV-1-infected patients.8 Blocking CD55 or CD59 function with specific Abs or removing these molecules by treatment of phosphatidylinositol-specific phospholipase C (PI-PLC), a protein enzyme used for release of GPI-APs from cell membrane, renders virions and infected cells sensitive to complement attack.2, 5, 6, 8
HCV is an enveloped positive-sense single-stranded RNA virus of the family Flaviviridae. Similar to HIV-1 infection, Abs against a wide variety of both HCV structural and nonstructural (NS) protein epitopes become detectable within weeks postinfection.9, 10 Patients who become chronically infected develop high-titer and crossreactive nAbs,9 yet fail to clear the virus. It is unknown whether HCV virions also incorporate RCA to protect against ADCML. HCV particles have been found to contain many host lipoproteins11, 12 which play important roles in the HCV life cycle at various stages.13 In fact, the cellular lipid droplet has been shown to be an important organelle for HCV production, as inhibition of lipid biosynthesis efficiently blocks HCV replication.14 HCV particles bind to specific cell-surface proteins (CD81, SR-B1, low density lipoprotein receptor, glycosaminoglycans, CD209, and CD209 ligand) to trigger virus internalization by clathrin-dependent endocytosis.15 HCV translation is then initiated to produce a single polyprotein that is cleaved by cellular and viral proteases into 10 small proteins including three structural proteins (Core, E1, and E2), and seven NS proteins (P7, NS2, NS3, NS4A, NS4B, NS5A, and NS5B).16, 17 Viral RNA replication takes place in a membrane-associated replication complex (the membranous web) that consists of viral NS proteins, replicating RNA, and cellular membranes.17 The newly synthesized viral genomes bud into the endoplasmic reticulum (ER) lumen to form viral particles. Thus, the ER plays a critical role in formation of HCV envelope and viral maturation, although the molecular mechanism of this is not fully understood. The ER also plays a central part in the synthesis and modification of GPI-APs such as CD59 in all eukaryotic cells.18 Therefore, it is possible for HCV to encounter and obtain CD59 and other GPI-APs in the ER and other cellular organelles.
The current study was undertaken to determine whether HCV virions incorporate CD59 into the viral Env at biologically functional levels. The results indicate that CD59 is incorporated into both cell line-derived and plasma primary HCV virions at levels that protect against ADCML. This is the first demonstration that HCV virions contain and hijack RCA proteins to escape the humoral immune response.
Ab, antibody; Ad5, adenovirus serotype 5; ADCML, antibody-dependent complement-mediated lysis; CMV, cytomegalovirus; ELISA, enzyme-linked immunosorbent assay; Env, envelope; ER, endoplasmic reticulum; FACS, fluorescence-activated cell sorting; FBS, fetal bovine serum; FFU, focus-forming unit; GFP, green fluorescence protein; GPI-AP, glycosylphosphatidylinositol-anchored protein; GVB, gelatin veronal buffer; HCV, hepatitis C virus; HIS, heat-inactivated human serum; HIV-1, human immunodeficiency virus type 1; HSV-1, herpes simplex virus 1; IFN-α, interferon-alpha; mAbs, monoclonal antibodies; MAC, membrane attack complex; MOI, multiplicity of infection; nAbs, neutralizing antibodies; NHS, normal human serum; NS, nonstructural protein; pAbs, polyclonal antibodies; PHHs, primary human hepatocytes; PI-PLC, phosphatidylinositol-specific phospholipase C; qPCR, real-time quantitative RT-PCR; RCA, regulators of complement activation.
Materials and Methods
Huh7.5.1 cells, a human hepatocyte cell line that supports replication of HCV in vitro, were a gift from Apath (Brooklyn, NY). Huh7.5.1 cells were grown in RPMI 1640 medium (Invitrogen, Carlsbad, CA) supplemented with 10% fetal bovine serum (FBS), L-glutamine, penicillin, and streptomycin. Cryopreserved primary human hepatocytes (PHHs) from six individual donors were purchased from ZenBio (Research Triangle Park, NC) and used in the current study.
Peripheral blood was collected in BD Vacutainer tubes (containing 143 USP units of sodium heparin per 10-mL tube, BD, Franklin Lakes, NJ) from patients chronically infected with HCV or from healthy donors. Plasma was separated from blood cells by centrifugation and then stored at −80°C until use. Plasma HCV viral loads were measured using the COBAS TaqMan HCV Test (Roche, Pleasanton, CA) in a certified clinical laboratory. Anti-HCV E2 Abs in these plasma samples were titrated by an enzyme-linked immunosorbent assay (ELISA) assay modified from a previous report19 (and Supporting Material). Informed consent was obtained from each participant and all investigational protocols were approved by Institutional Review Boards for Human Research at the Indiana University School of Medicine (Indianapolis, IN).
Preparation of Cell Line-Derived and Primary HCV Virions.
JFH-1, a unique HCV genotype 2a replicon derived from a viral isolate of a patient with fulminant HCV, was a gift from Apath (Brooklyn, NY). JFH-1 was inoculated into Huh7.5.1 cells at a multiplicity of infection (MOI) of 1. On day 5 of incubation cell-free supernatants were collected by a centrifugation at 15,000 rpm at 4°C for 10 minutes and then filtered through a 0.2-μm-pore-size filter, followed by virus purification using 20% versus 60% two-layer sucrose gradient ultracentrifugation as described.12 Four fractions collected by aspiration from the top were subjected to: (1) ELISA for measuring HCV core concentration, (2) real-time quantitative reverse-transcription polymerase chain reaction (qPCR) for measuring viral RNA copies, and (3) western blot for measuring CD59. Fraction 3 was also subjected to HCV capture.
The virus purification protocol described above was also used to purify HCV particles from plasma samples from five HCV-infected individuals (Pt1 to Pt5) (Table 1). The purified virus particles from fraction 3 were then subjected to western blot for CD59 detection.
Table 1. Profiles of HCV-lnfected patients
HCV viral load (copies/ml plasma)
Anti-E2 Ab titer
Note: M, male; F, female; C, Caucasian; B, black; AP, alkaline phosphatase; ALT, amino alanine transferase; AST, aminotransferase; NA, not available; ND, not determined.
Human CD59 Blockers.
Two CD59 blockers, BRIC229 and rILYd4, were used in the current study to abrogate CD59 function. BRIC229, a mouse antihuman CD59 monoclonal Ab (mAb) (IBGRL, Bristol, UK), is a widely used CD59-blocking Ab, whereas rILYd4 is a recombinant form of the fourth domain of intermedilysin (ILY), and has been identified as a high-affinity inhibitor of human CD59.6
CD59 ELISA, Virus Capture, and qPCR.
Microplates were coated with rabbit antihuman CD59 polyclonal Abs (pAbs, Santa Cruz, Santa Cruz, CA) or control IgG at 1 μg/mL in phosphate-buffered saline (PBS) overnight at 4°C. After washing and blocking, plates were used for CD59 measurement or HCV capture. To measure CD59 levels, the wells were incubated with Triton X-100-pretreated supernatant from JFH-1-infected or uninfected Huh7.5.1 cells (100 μL/well) at 37°C for 1 hour. After incubation and washing, the bound CD59 was detected by incubating with BRIC229 or control IgG, followed by incubating with HRP-labeled secondary Ab. Cell-free supernatant samples containing HIV-1 particles from two HIV-1-infected human monocytic cell lines, THP-1 (CD59-positive) and U1 (CD59-negative), were used as positive and negative controls, respectively, as HIV-1 particles derived from CD59-expressing THP-1 cells contain CD59, whereas HIV-1 particles derived from CD59-deficient U1 cells are CD59-negative.6 Cell-free supernatant from Ad5 (adenovirus serotype 5)-infected Huh7.5.1 cells were also included in the ELISA detection to rule out the possibility of cell-derived CD59. Ad5, a nonenveloped (naked) DNA virus, causes cytolytic infection and has no Env for CD59 incorporation. Therefore, any CD59 detected in these supernatant samples would be derived from dead cells and cell debris due to Ad5 cytolytic infection rather than Ad5 virions.
To detect HCV capture, the wells were incubated with untreated cell-free supernatant from JFH-1-infected, uninfected Huh7.5.1 cells (100 μL/well), or purified virus fraction 3 at 37°C for 1 hour. After incubation, free virus was removed by washing with PBS, and the remaining bound HCV was lysed with 200 μL of TRIzol (Invitrogen) for isolation of viral RNA, which was subjected to qPCR for measuring HCV RNA copy numbers as described in our Supporting Material. Cell-free supernatant samples from HIV-1-infected and uninfected THP-1 cells were used as positive and negative controls, respectively, as antihuman CD59 Abs have been reported to efficiently capture intact HIV-1 virions.5
PHHs and Huh7.5.1 cells treated or untreated with PI-PLC were lysed in 1× cell lysis buffer (Cell Signaling Technology, Danvers, MA), and then subjected to protein extraction and western blot as described in our Supporting Material. HCV particles purified from cell-free supernatant of JFH-1-infected Huh7.5.1 cells or from plasma samples of HCV-infected patients (Pt1 to Pt5, Table 1) were lysed with 15 μL of 1% Triton X-100, and then subjected to western blot.
Complement-Mediated Lysis of Huh7.5.1-Derived HCV Virions.
The sensitivity of HCV virions in the cell-free supernatant from JFH-1-infected Huh7.5.1 cells to ADCML in the presence or absence of CD59 blockers (BRIC229 and rILYd4) was assessed using a protocol modified from our previous report.6 Briefly, HCV-containing supernatant (50 μL) were preincubated with (1) BRIC229 (1.25-20 μg/mL), (2) rILYd4 (1.25-20 μg/mL), (3) irrelevant IgG control (1.25-20 μg/mL), (4) PBS, or (5) Triton X-100 at 37°C for 30 minutes. After preincubation, anti-HCV E2 pAbs or irrelevant pAbs (anti-HIV-1 gp120/160 pAbs; Abcam, Cambridge, MA) were added, followed by exposure to either complement-competent human sera or heat-inactivated complement (CompTech, Tyler, TX) diluted in gelatin veronal buffer (GVB) (Sigma-Aldrich, St. Louis, MO). Virolysis of HCV was quantified by measuring HCV core release using the QuickTiter HCV Core ELISA Kit as per the manufacturer's description, except that the lysis buffer included in the ELISA kit was not used. Therefore, only HCV core released from the lysed viral particles by ADCML was quantified, whereas the core in the intact HCV virions was embedded in the outer Env, and thereby was not detected. HCV virions treated with Triton X-100 and PBS were used as 100% and blank of virolysis, respectively. The percentage of virolysis was calculated as follows: (core released by CD59 blocker − core released by PBS) / (core released by Triton X-100 − core released by PBS) × 100%.
Direct Lysis of HCV Virions in Patient Plasma.
Plasma samples from six HCV-infected subjects (Pt28, Pt42, Pt49, Pt84, Pt99, and Pt369) are described in Table 1 and were directly treated with: (1) BRIC229 (20 μg/mL), (2) rILYd4 (20 μg/mL), (3) irrelevant IgG control (20 μg/mL), (4) PBS, or (5) Triton X-100 at 37°C for 1 hour. Plasma samples from another five HCV-infected subjects (Pt1 to Pt5; Table 1) were completely used for virus purification; they were not available to be included in the direct virolysis experiments. Virolysis was quantified and calculated by measuring HCV core protein release as described above. All samples were run in triplicate.
Flow Cytometric Analysis (FACS).
Cells were incubated with BRIC229 or control Ab, followed by FITC-conjugated secondary Ab, and then subjected to FACS using a BD FACSCalibur (BD Biosciences). Data were analyzed using FlowJo software (Tree Star, San Carlos, CA). To measure intracellular level of CD59, PHHs or Huh7.5.1 cells were treated with PI-PLC (Sigma-Aldrich) at 0.5 units/mL at 37°C for 1 hour to remove CD59 from the surface of cells. After washing, cells were permeabilized using a Cytofix/Cytoperm Plus kit (BD PharMingen) according to the manufacturer's instructions. Cells were incubated with BRIC229 or control Ab and followed by FITC-conjugated secondary Ab for FACS as described above.
The paired two-tailed Student's t test was used to compare the means ± standard deviation (SD). Values of P < 0.05 were judged significant.
Human Hepatocytes Express High Levels of Intracellular and Surface CD59.
To determine whether human hepatocytes express CD59, thereby serving as a possible source of CD59 to HCV, we used FACS to analyze surface and intracellular CD59 expression on PHHs and Huh7.5.1 cells. In the absence of any stimulus, both PHHs and Huh7.5.1 cells expressed surface CD59 at levels comparable to that seen on the surface of CD59-expressing THP-1 cells (Fig. 1A). These hepatocytes also expressed a high level of intracellular CD59 that was detected after removal of surface CD59 by PI-PLC treatment (Fig. 1B). Western blot results revealed that a single ≈19 kDa protein band was detected by BRIC229 from PI-PLC-treated and -untreated PHHs or Huh7.5.1 cells, suggesting that cell surface and intracellular CD59 molecules are not significantly changed in these cells (Fig. 1C). Thus, PHHs and Huh7.5.1 cells express substantial levels of surface and intracellular CD59 in the absence of any stimulus, thereby potentially providing a source for HCV to incorporate CD59 in intracellular organelles, plasma membrane, or both.
Huh7.5.1 Cell-Derived HCV Virions Contain CD59.
Next we determined whether HCV virions contained CD59. ELISA results showed that CD59 was not detected in the supernatant from uninfected Huh7.5.1 cells (Fig. 2A), suggesting that, in the naïve condition, CD59 does not appear to have a soluble or secretory form. In contrast, CD59 in the supernatant from HCV-infected Huh7.5.1 cells was easily detected at levels comparable to that seen in the supernatant of HIV-1-infected THP-1 cells (Fig. 2A). CD59 concentration in the supernatant of activated U1 cells was slightly, but not significantly, higher than that in the cell-free supernatant from uninfected Huh7.5.1 or THP-1 cells (Fig. 2A). CD59 was not detected from the supernatant of Ad5-infected Huh7.5.1 cells (Fig. 2A). Ad5 is a nonenveloped cytolytic virus incapable of incorporating cellular proteins onto its surface. Ad5 rapidly and efficiently infected Huh7.5.1 cells as 5.1%, 14.3%, and 34.4% of Huh7.5.1 cells became GFP-positive after overnight infection with 1, 2, 10 MOI of a replication-defective GFP-Ad5, respectively (Fig. S1 and Supporting Material), and 100% of cells became rounded and undetached after 2-3 days of infection (data not shown), indicating that massive cell death occurred. Thus, absence of CD59 in these supernatant samples suggests that, in infected/stimulated conditions, CD59 does not have a soluble or secretory form and dead cells do not release soluble CD59 into the supernatant of cell cultures. Therefore, CD59 detected in the supernatant of HCV-infected cells is most likely derived from HCV virions.
To further assess the presence of CD59 on virus, HCV particles were purified from the supernatant of JFH-1-infected Huh7.5.1 cells using sucrose gradient ultracentrifugation. In agreement with the previous report,12 most of the HCV particles were concentrated in fraction 3, as determined by ELISA of HCV core quantification and by qPCR of HCV RNA copies (Fig. 2B). Fraction 3 corresponded to the 20% to 60% sucrose interphase (Fig. 2B). CD59 was also concentrated in fraction 3 as determined by western blot and its level correlated with HCV core concentration and viral RNA copies (Fig. 2B), indicating that CD59 is related to the HCV particles. All fractions collected from the supernatant of uninfected Huh7.5.1 cells were HCV core and RNA negative and CD59 negative (Fig. 2B).
To further exclude the possibility of host cell protein contamination, a virus capture assay was utilized. In agreement with the previous report,5 HIV-1 particles were captured by anti-human CD59 pAbs, as HIV-1-specific qPCR qualified 167 copies of HIV-1 RNA from an input of 2,000 viral RNA copies in 100 μL of supernatant (8.4% capture rate) (Fig. 2C). Similarly, HCV particles were also captured by the pAbs, although only 26 copies of viral RNA were detected by the qPCR from an input of 2,000 HCV copies in 100 μL of supernatant (1.3% capture rate) (Fig. 2C). HCV capture efficiency was markedly enhanced when the purified viral particles were used, as 215 copies of viral RNA were detected from an input of 2,000 HCV copies of the purified virus fraction 3 resuspended in 100 μL of supernatant from uninfected Huh7.5.1 cells (10.8% capture rate) (Fig. 2D). Thus, anti-human CD59 Abs captured HCV, which directly shows the presence of CD59 on the external membrane of HCV particles.
Primary HCV Virions from the Plasma of HCV-Infected Individuals Contain CD59.
To further investigate whether primary HCV virions also incorporate CD59, we purified HCV particles from the plasma of five HCV-infected individuals by sucrose gradient ultracentrifugation as described above. The purified primary virions were subjected to western blot for measuring CD59. As shown in Fig. 3, CD59 was detected by western blot from virus particles purified from plasma samples of all five HCV-infected patients examined (Pt1 to Pt5; Table 1), but not from any of the three HCV-negative healthy donors (H1 to H3). Importantly, CD59 levels correlated with plasma HCV viral loads (Fig. 3), suggesting that the CD59 signal is derived from HCV particles rather than potential contamination of host proteins coprecipitated from plasma samples.
Abrogation of CD59 Function Enhances ADCML of Huh7.5.1-Derived HCV Virions.
To test whether CD59 incorporation protects HCV against ADCML, we used BRIC229 and rILYd4 to block CD59 and then analyzed HCV lysis in the presence or absence of anti-HCV E2 pAbs with or without competent complement. As shown in Fig. 4A, HCV core was markedly increased in both BRIC229 and rILYd4 treatments in a dose-dependent manner when compared with PBS or IgG control. The increase of HCV core was triggered by ADCML because the effects of BRIC229 and rILYd4 were completely abolished if heat-inactivated complement was used or anti-HCV E2 pAbs were replaced with anti-HIV-1 gp120/160 pAbs (Fig. 4A,B). Notably, moderate levels of HCV core were detected in PBS control groups in the presence (13.6 ± 1.9 ng/mL, n = 3) or absence (12.6 ± 2.6 ng/mL, n = 3) of complement activation when compared with the maximal lysis of Triton X-100 treatments in the presence (36.3 ± 2.9 ng/mL, n = 3) or absence of complement activation (35.7 ± 3.6 ng/mL, n = 3) (Fig. 4A,B), suggesting that a certain amount of broken HCV virions existed in the supernatant of infected cells. Analysis of virolysis showed that both BRIC229 and rILYd4 treatments increased virolysis of HCV virions. Treatment with BRIC229 or rILYd4 at 20 μg/mL increased virolysis from 2.6 ± 1.2% (IgG at 20 μg/mL, n = 3) to 53.7 ± 6.3% (n = 3) and 63.9 ± 9.7% (n = 3), respectively (Fig. 4C). The effects of ADCML by rILYd4 treatment appeared greater than those mediated by BRIC229, although they were not significant (Fig. 4C).
To understand the consequence of ADCML, we performed focus-forming experiments to quantitate the number of infectious HCV virions remaining in the ADCML samples (Fig. 4B) (IgG, BRIC229 or rILY4d at 20 μg/mL with complement plus anti-HCV E2 pAbs). Although cells in all conditions were not undetached from wells after 4 days of incubation (Fig. 2S, lower panel), HCV foci were not observed in Huh7.5.1 cells exposed to Triton X-100-treated solution (Fig. 4D; Fig. 2S), indicating that all potentially infective particles were totally lysed. Huh7.5.1 cells exposed to control solutions (PBS or IgG) had greater numbers of HCV foci (Fig. 4D; Fig. 2S), whereas cells exposed to BRIC229- or rILYd4-treated solutions showed lower numbers of HCV foci (Fig. 4D; Fig. 2S), indicating that both BRIC229 and rILYd4 allowed the anti-HCV Abs to regain their ADCML activity, resulting in a reduction of HCV infectivity.
Direct Addition of CD59 Blockers Enhances Lysis of HCV Virions in Patient Plasma.
To test whether abrogation of CD59 function renders plasma primary HCV virions sensitive to complement destruction, we directly added BRIC229 or rILYd4 into patient plasma without adding any artificial buffer and then analyzed HCV virolysis. Six plasma samples (Table 1) from chronically HCV-infected subjects were tested and all showed potent complement activity determined by an Ab-sensitized hemolytic assay (Supporting Material, Fig. 3S), although these plasma samples contained 15 USP units per mL of sodium heparin as an anticoagulant. Summary data of virolysis from all six individuals are illustrated in Fig. 5A. Similar to HCV infection in vitro, moderate levels of HCV core were detected in PBS control groups in all plasma samples tested when compared with those of maximal release of HCV core in Triton X-100 groups, ranging from 5.6 ± 1.1 ng/mL (PBS) versus 8.9 ± 1.8 (Triton X-100) (Pt28) to 69.2 ± 10.6 ng/mL (PBS) versus 163.1 ± 26.1 (Triton X-100) (Pt49). In all samples tested, IgG treatment caused a slight, but not significant, increase of HCV core when compared with PBS (Fig. 5A). In the presence of BRIC229 or rILYd4, all six HCV plasma samples showed increased release of HCV core when compared with those of PBS or IgG treatment, albeit in varied degrees (Fig. 5A). Three samples (Pt49, Pt84, and Pt369) showed a significant increase of HCV core in response to BRIC229 or rILYd4 treatment, whereas the remaining three (Pt28, Pt42, and Pt99) were affected slightly, but not significantly (Fig. 5A). Analysis of virolysis showed that both BRIC229 and rILYd4 treatments significantly increased virolysis of primary HCV virions when compared with IgG treatment at the same concentration (Fig. 5B,C). The effects of ADCML following rILYd4 treatment appeared greater than those mediated by treatment with BRIC229, although they were not significant (Fig. 4C). Taken together, these results indicate that CD59 blockers (BRIC229 and rILYd4) also sensitize plasma primary HCV virions to complement-mediated virolysis, and that CD59 blockers enhance virolysis of HCV virions not only under experimental conditions but also in real clinical environments of blood samples from HCV-infected patients.
This report provides evidence that CD59 is incorporated into HCV virions at levels that protect against ADCML. First, CD59 was detected in the supernatant from HCV-infected, but not from either uninfected or Ad5-infected Huh7.5.1 cells, indicating that the detected CD59 most likely derives from HCV particles rather than dead cells and/or a soluble or secretory form. Second, CD59 was detected in purified HCV particles from cell cultures in vitro and plasma samples from patients chronically infected with HCV, and CD59 level correlated with HCV core concentration and viral RNA copy numbers (Fig. 2B) or plasma HCV viral loads (Fig. 3). Third, anti-human CD59 Abs captured HCV particles from the cell-free supernatant, albeit with less efficiency than that of HIV-1 capture. Possible explanations for this disparity are that (1) HCV simply incorporated less CD59 than HIV-1 due to different cell types used for virus preparations and different mechanisms of virus-cell interaction, and (2) there were more broken HCV particles in the supernatant of HCV-infected Huh7.5.1 cells than that of HIV-1 in the supernatant of infected THP-1 cells, as moderate levels of viral core were detected in the PBS control groups of HCV virolysis, but not in HIV-1 virolysis.5, 6 Broken HCV particles might release CD59-associated proteins, which competed with intact HCV particles to bind to coated anti-CD59 Abs, resulting in less intact HCV particles being captured. Removal of broken HCV particles by sucrose gradient ultracentrifugation significantly enhanced HCV capture efficiency. Our finding of broken viral particles echoes a previous report that HCV virions exist as a highly heterogeneous mixture of closely related viruses (quasispecies) containing a mixture of both infectious and noninfectious particles in ratios ranging from 1:100 to 1:1,000, both in vivo and in cell culture.12 Last, abrogation of CD59 function with its blockers increased the sensitivity of HCV virions from both cell cultures and plasma samples to ADCML, resulting in a significant reduction of HCV infectivity. These results indicate that CD59 is present on the external membrane of HCV particles at levels that protect from ADCML.
HCV exclusively replicates in human hepatocytes and has a high rate of replication with approximately one trillion (1 × 1012) particles produced each day in an infected individual.16, 17 HCV has a limited genome consisting of a single open reading frame with 9,600 nucleotide bases in length.16, 17 With this limited genome, HCV replicates in hepatocytes by relying on cellular systems, thereby co-opting cellular proteins in its life cycle. To date, HCV particles have been found to contain more than 10 host lipoproteins.11, 12 Incorporation of these host proteins into HCV virions may not be random, as other enveloped viruses selectively obtain host proteins. For example, HIV-1 selectively acquires more than 40 host proteins, but excludes some proteins such as CD4, CD45, CXCR4, CCR3, and CCR5, which are all highly expressed on HIV-1-infected cells.20 It is believed that HIV-1 acquires biologically functional RCA proteins during viral budding at the plasma membrane. HCV, however, may acquire CD59 while budding through the membranes of intracellular organelles rather than at the plasma membrane because HCV may exit the cells by way of a secretory pathway.21 FACS and western blot data in this study showed that human hepatocytes expressed high levels of intracellular and surface CD59 without a difference in their molecular weights (Fig. 1), thereby providing a possible source for HCV to encounter and obtain CD59 in intracellular organelles such as the ER. Identifying these interactions is critical for understanding the life cycle of HCV, and may yield novel targets for development of therapeutic strategies.
To date, abrogation of RCA function to regain antivirus Ab activity against enveloped viruses has only been tested in vitro in artificial environments such as GVB systems.2, 5, 6, 8 These systems provide optimized conditions for complement activation in vitro, but have no clinical relevance because they do not adequately replicate in vivo conditions. In this study, CD59 blockers were directly added into patient plasma to abrogate the function of CD59 on the patient's own viral particles, resulting in destruction of primary virions. The enhanced virolysis was likely triggered by ADCML, as all six individuals chronically infected with HCV showed high titers of anti-E2 Abs and potent complement activity. ADCML efficacy, however, significantly varied among these samples, which may be affected by many factors including the nature of the host immune response, HCV virological features, and patient profiles, because they all affect the outcome of HCV infection. For example, HCV from patient Pt42 was one of the most resistant viruses to the ADCML. Accordingly, this patient had the lowest anti-HCV E2 Ab titer among all six patients examined. However, our sample size is too small to analyze the correlation between the Ab titer and virolysis. In addition, anti-HCV E1 Abs in plasma samples from these patients were not titrated. Thus, further investigations with large clinical samples are required to analyze the correlation of anti-HCV E1/E2 Ab titers and virolysis efficacy.
In conclusion, this study is the first report to demonstrate that HCV particles contain CD59 at levels that provide resistance to ADCML. This is also the first report to show that direct addition of RCA blockers into plasma samples from patients chronically infected with HCV render endogenous plasma virions sensitive to complement-mediated destruction. This strategy may be further developed in combination with the current standard of care for treatment of chronic HCV (pegylated IFN-α plus ribavirin) to enhance therapy efficacy.
We thank Apath (Brooklyn, NY) and Dr. Charles M. Rice at Rockefeller University (New York, NY) for JFH-1, pFL-J6/JFH, and Huh7.5.1 cells.