Previous Article | Next Article ![]()
Journal of Virology, January 2008, p. 569-574, Vol. 82, No. 1
0022-538X/08/$08.00+0 doi:10.1128/JVI.01443-07
Copyright © 2008, American Society for Microbiology. All Rights Reserved.

Valerie Castet,1,
Lydiane Pichard-Garcia,1
Czeslaw Wychowski,2
Eliane Meurs,3
Jean-Marc Pascussi,1
Camille Sureau,4
Jean-Michel Fabre,1,5
Antonio SaCunha,6
Dominique Larrey,1,7
Jean Dubuisson,2
Joliette Coste,1,8
Jane McKeating,9
Patrick Maurel,1* and
Chantal Fournier-Wirth1,8
Inserm, U632, 34293 Montpellier, France; Univ. Montpellier 1, EA3768, 34000 Montpellier, France,1 CNRS, UPR2511, Unité Hépatite C, Institut de Biologie de Lille & Institut Pasteur de Lille, 59021 Lille, France,2 Institut Pasteur, Unité Hepacivirus, 75724 Paris, France,3 Institut National de Transfusion Sanguine, 75139 Paris, France,4 Hôpital Saint Eloi, Service de Chirurgie Digestive II, 34095 Montpellier, France,5 Hôpital Haut Lévèque, Service de Chirurgie Digestive, 33075 Pessac, France,6 Hôpital Saint Eloi, Service d'Hépatogastroentérologie, 34095 Montpellier, France,7 Etablissement Français du Sang, 34094 Montpellier, France,8 Division of Immunology and Infection, Institute of Biomedical Research, University of Birmingham, Birmingham B15 2TT, United Kingdom9
Received 3 July 2007/ Accepted 4 October 2007
|
|
|---|
|
|
|---|
The current study was undertaken to evaluate the role of CD81 in HCVser infection of highly differentiated normal primary human hepatocytes. Hepatocyte isolation, culture conditions, infection with serum from patients chronically infected with HCV, and treatments were performed as described elsewhere (9, 15, 26, 30). These cultures retain liver phenotypic markers (3, 8, 11, 13, 16, 28, 31) and are sensitive to HCV or hepatitis delta virus (HDV) infection (1, 9, 15, 26, 39). Intracellular accumulation of replicative and genomic HCV RNA strands, assessed by rTth-reverse transcription (RT)-PCR and quantitative RT-PCR, respectively, were used as experimental end points (9, 15, 20, 21, 26). Statistical analysis of data was performed using the nonparametric signed-rank Wilcoxon test (34). We verified that hepatocytes cultured under our standard conditions express both CD81 and CD9, a closely related tetraspanin (not shown). On average, after HCVser infection of CD81+/CD9+ hepatocytes at a genome equivalent per cell (Geq/cell) ratio of 0.025, the number of HCV genome copies per hepatocyte is 0.18 ± 0.10 (n = 13; range, 0.04 to 0.37).
Anti-CD81 MAbs JS81 and JS64 inhibited HCVser (genotypes 1 to 3) infection of hepatocytes in a concentration-dependent manner (Fig. 1A to C). On average, MAb JS81 produced 92 ± 9% inhibition. Anti-CD9 MAb M-L13 or the mouse immunoglobulin G1 isotype control produced no inhibition. When cells were treated with MAb JS81 at the time of inoculation or shortly after (30 min), the accumulation of both negative and positive HCV RNA strands was clearly reduced (>90% on the positive strand) (Fig. 1D and E). However, as the interval between HCVser inoculation and anti-CD81 treatment increased, the inhibitory effect gradually decreased. When antibodies were added 8 h after inoculation, the inhibition was markedly reduced. Neither anti-CD81 nor anti-CD9 MAbs prevented HDV infection of hepatocytes (not shown). The inhibitory effect of anti-CD81 MAbs therefore interferes with an early step of HCVser cell entry and is specific to HCV infection.
![]() View larger version (44K): [in a new window] |
FIG. 1. Effect of anti-CD81 MAbs and of siRNA-mediated silencing of CD81 on HCVser infection of primary human hepatocytes (PHH). Three days postplating, hepatocytes FTBP (Biopredic) from a 52-year-old male with metastasis of a colic tumor were exposed or not (UT) to increasing concentrations (0.5 to 6 µg/ml) of anti-CD81 MAb JS81 or JS64, anti-CD9 MAb M-L13, or mouse immunoglobulin G1 isotype MOPC-21 for 30 min before inoculation with HCV-positive serum S298 (genotype 3a; 1.5 x 106 IU/ml). Following overnight exposure, cells were washed three times and the culture was continued. Five days postinoculation, cells were washed three times and total cellular RNA was extracted. (A) One µg of total RNA was analyzed by rTth-RT-PCR for the replicative RNA strand. (B) One µg of the same RNA sample was analyzed by real-time PCR for the genomic strand. RNA quality control was evaluated by performing a quantification of GAPDH mRNA in each sample. Amounts of HCV RNA were normalized to GAPDH mRNA. Results are expressed as the percent intracellular viral RNA relative to the amount obtained in in vitro-infected hepatocytes in the absence of treatment (0). Data are means from three independent assays. These results are representative of observations made with all other hepatocyte cultures from different donors used in this work. (C) Three days postplating, hepatocytes FT168 (75-year-old female with metastasis of a colic tumor) were exposed to HCV-positive serum S42 (genotype 1b, 6 x 106 IU/ml), S31 (genotype 2; 13 x 106 IU/ml), or S34 (genotype 1b; 60 x 106 IU/ml); hepatocytes FT196 (35-year-old female with metastasis of a colic tumor) were exposed to serum S155 (genotype 1a; 1.1 x 106 IU/ml); hepatocytes FT 212 (52-year-old male with metastasis of a colic tumor) were exposed to serum S183 (genotype 3a, 2.6 x 106 IU/ml); and hepatocytes FT249 (52-year-old male with metastasis of a colic tumor) were exposed to serum S192 (genotype 3a; 2.5 x 106 IU/ml) in the absence (UT) or presence of anti-CD81 MAbs JS81 at 4 µg/ml (mAb); further analysis was carried out as indicated above. Amounts of HCV RNA were normalized to GAPDH mRNA. Results are expressed as the percent intracellular viral RNA relative to the amount obtained in in vitro-infected hepatocytes in the absence of treatment (UT). Data are means from three independent assays. These results are representative of observations made with other hepatocyte cultures from different donors, FT161 (35-year-old male organ donor), 167 (57-year-old male with metastasis of a colic tumor), 176 (69-year-old female with metastasis of a colic tumor), 195 (17-year-old male organ donor), and FTBP. (D and E) Three days postplating, hepatocytes (FT196) were exposed to HCV-positive human serum S155. Cells were left untreated (UT) or treated with anti-CD81 MAbs JS81 (4 µg/ml) added either at the time of inoculation with HCV-positive serum (0), or 30 min, 1 h, 2 h, 4 h, or 8 h later, and further analysis was carried out as indicated above. (D) One µg of total RNA was analyzed by rTth-RT-PCR for the replicative RNA strand. (E) One µg of the same RNA sample was analyzed by real-time PCR for the genomic strand. Amounts of HCV RNA were normalized to GAPDH mRNA. Results are expressed as percent intracellular viral RNA relative to the amount obtained in in vitro infected hepatocytes in the absence of treatment (UT). Data are means from three independent assays. These results are representative of observations made with another hepatocyte culture from a different donor (FT195). (F) Hepatocytes FT259 (76-year-old male with hepatocellular carcinoma) were transduced overnight with suspensions of lentivirus expressing hCD81 siRNA 412, 619, or 1262, or expressing human CAR siRNA 382, at a multiplicity of infection (MOI) of 0.5 or 1. Nontransduced cells were used as the control (NT). Seven days later, GFP-positive hepatocytes were stained with anti-CD81 MAb JS-81 conjugated with Zenon and analyzed by flow cytometry. The results are expressed as the percentage of GFP-positive cells expressing (+) or not expressing (–) CD81 protein. Hepatocytes were then infected, on day 8, with HCV-positive serum sample S268 (genotype 1a; 4.9 x 106 IU/ml), and at day 3 postinoculation, the intracellular HCV RNA was analyzed by quantitative real-time RT-PCR as indicated above. All experiments were carried out in duplicate. These results are representative of those for various cell cultures and infectious serum samples S297 (genotype 3a; 1.5 x 106 IU/ml) for culture FT256 (43-year-old female with adenoma) and S268 and S301 (genotype 3a; 3.3 x 106 IU/ml) for culture FT259. Statistical analysis and P values were obtained using the Wilcoxon test.
|
Recombinant CD81-LEL glutathione S-transferase fusion proteins from humans, African green monkeys, and mice were synthesized as described previously (17). We first verified that hCD81-LEL binds glycoprotein E2 and that LEL and anti-CD81 MAbs JS81 and JS64 markedly inhibit (>90%) JFH1/HCVcc infection of Huh-7.5 cells (not shown), as also observed by others (23, 43). Interestingly, hCD81-LEL did not inhibit HCVser infection of hepatocytes (Fig. 2A and B). As this result was rather unexpected, we compared HCVser and JFH1/HCVcc infections of human hepatocytes. We first established that under our standard conditions of infection (9, 15, 26), JFH1/HCVcc infects hepatocytes at Geq/cell ratios of 0.03 and 0.1, as assessed by both genomic and replicative HCVcc RNA strand accumulation in cells and production of infectious virions in the cell supernatant (not shown). The JFH1/HCVcc genomic strand level was 3.3 ± 1 (n = 3; range, 2.5 to 4.1) copies per hepatocyte after infection at a Geq/cell ratio of 0.03 (and 9.7 at a Geq/cell ratio of 0.1). This is an order of magnitude greater than that observed (0.18 ± 0.10) after HCVser infection at the same Geq/cell ratio. Unexpectedly, however, JFH1/HCVcc infection of human hepatocytes was inhibited significantly but only partly (40 to 60%) by both hCD81-LEL (Fig. 2C and D) and anti-CD81 MAbs (compare Fig. 3A and B and Fig. 1B), in sharp contrast with the strong inhibition (>90%) observed with JFH1/HCVcc infection of Huh-7.5 cells (not shown) and with HCVser infection of hepatocytes (Fig. 1) (92%).
![]() View larger version (34K): [in a new window] |
FIG. 2. Effect of recombinant mouse, green monkey, and human CD81-LEL on HCVser and JFH1/HCVcc infections of primary human hepatocytes (PHH). (A and B) HCV-positive serum S297 or S298 was preincubated for 30 min at 4°C with 10, 20, or 40 µg/ml green monkey (G), mouse (M), or human (H) CD81-LEL fusion proteins and, 3 days postplating, hepatocytes FT257 (30-year-old female with hydatid cyst) or FT259 were inoculated with these mixtures under our standard conditions. After exposure, cells were washed three times to remove excess inoculum, and the culture was continued for 3 days. Total cellular RNA was extracted, and 1 µg was analyzed by rTth-RT-PCR for the replicative RNA strand (FT257/S297) or by real-time PCR for the genomic strand. Amounts of HCV RNA were normalized to GAPDH mRNA. All experiments were carried out in duplicate. These results represent average data obtained from these four series of experiments (FT257/S297 or FT257/S298; FT259/S297 or FT259/S298). These experiments were repeated with two different lots of CD81-LEL fusion proteins with same results. (C and D) JFH1/HCVcc was preincubated or not (UT) with 10, 20, or 40 µg/ml of green monkey (G), human (H), or mouse (M; only 40 µg/ml) CD81-LEL for 30 min and used to inoculate hepatocytes FT273 (64-year-old male with metastasis of a colic tumor), FT276 (62-year-old female with metastasis of a colic tumor), and FT281 (56-year-old male with metastasis of a colic tumor) at different Geq/cell ratios (0.03 and 0.1) 3 days postplating. Three days later, cells were washed three times, and total cellular RNA was extracted. One µg of the same RNA sample was analyzed by real-time PCR for the genomic strand. RNA quality control was evaluated by performing a quantification of GAPDH mRNA in each sample. Amounts of HCV RNA were normalized to GAPDH mRNA. Results are expressed as Geq copies per cell. Data are means from three independent assays. Statistical analysis and P values were obtained using the Wilcoxon signed-rank test.
|
![]() View larger version (28K): [in a new window] |
FIG. 3. Effect of anti-CD81 MAbs on JFH1/HCVcc infection of primary human hepatocytes (PHH). Human hepatocytes (FT273, FT276, and FT281) were inoculated with JFH1/HCVcc at different Geq/cell ratios (0.03 and 0.1) in the absence (UT) or presence of 0.5 to 6 µg/ml of anti-CD81 MAb JS81 or JS64 added for 30 min before inoculation. Control experiments were performed under the same conditions with anti-CD9 MAbs M-L13. Following overnight exposure, cells were washed three times, and the culture was continued. Three days later, cells were washed three times, and total cellular RNA was extracted. One µg of total RNA was analyzed by real-time PCR for the genomic strand. RNA quality control was evaluated by performing a quantification of GAPDH mRNA in each sample. Amounts of HCV RNA were normalized to GAPDH mRNA. Results are expressed as a percentage of the control (UT). Data are means ± standard deviations for three independent assays. Statistical analysis and P values were obtained using the Wilcoxon signed-rank test.
|
In contrast to recombinant CD81-LEL proteins, which are expected to compete with the natural membrane receptor for binding virus particles, antibodies directly target the cellular receptor. Their efficiency is therefore expected to be greater. This is consistent with current results revealing that the efficacy of MAb JS81 is approximately one to two orders of magnitude greater than that of recombinant hCD81-LEL on a molar concentration basis (not shown). Anti-CD81 MAbs inhibited JFH1/HCVcc infection of primary hepatocytes weakly (60%) compared with HCVser infection of primary hepatocytes (92%), and JFH1/HCVcc infection of Huh-7.5 cells (>90%; not shown). It is possible that the higher input of infectious virus in JFH1/HCVcc versus HCVser infection of hepatocytes, the difference in affinity or kinetics of HCV interaction with Huh-7.5 versus hepatocyte membrane receptors, and the difference in membrane protein population in Huh-7.5 cells versus hepatocytes account for these differences in inhibition efficiency. In sum, the data obtained on recombinant CD81-LEL and anti-CD81 antibodies suggest that CD81 has a much greater impact on the entry of JFH1/HCVcc in Huh-7.5 cells than it does on the entry of HCVcc particles in primary hepatocytes. Nevertheless, other data provided here show that CD81 indeed mediates the entry of HCVser in hepatocytes. It is therefore possible that the nature of the cell-virion combination is an important determinant of the virus entry process.
It is interesting that JFH1/HCVcc particles are able to infect primary human hepatocytes as revealed here by the intracellular accumulation of both genomic and replicative strands (not shown). Indeed, although HCVser is infectious for primary hepatocytes, it has to be noted that less than 15% of the tested sera are actually infectious (15). The reason for this observation is unclear but could be related to the presence of anti-HCV antibodies in the sera of chronically infected patients. In addition, the Geq/cell ratio with HCVser can hardly be greater than 0.1 because of the limited amounts of infectious serum. Hence, the possibility to infect primary hepatocytes with unlimited amounts of JFH1/HCVcc at a much greater Geq/cell ratio will strongly facilitate future studies. The rate of genome replication of JFH1/HCVcc in human hepatocytes appears to be one order of magnitude greater than that observed with HCVser, whereas it is one order of magnitude smaller than that observed for Huh-7.5 cells at the same Geq/cell ratio. Thus, although JFH1/HCVcc particles are adapted to Huh-7.5 cells, they can replicate their genome and proliferate in primary human hepatocytes, albeit at a lower rate. This could result from the greater antiviral response of human hepatocytes than Huh-7.5 cells (L. Pichard-Garcia, S. Molina, and P. Maurel, unpublished results). On the other hand, the fact that the rate of genome replication of JFH1/HCVcc in hepatocytes is greater than that of HCVser suggests that either only a small portion of HCVser virions are infectious or antibodies present in the serum strongly affect virion infectivity.
In conclusion, CD81 plays a critical role in an early step of HCVser infection of normal primary human hepatocytes. Comparative studies on HCVser versus JFH1/HCVcc infection of human hepatocytes and on JFH1/HCVcc infection of human hepatocytes versus Huh-7.5 cells suggest that the nature of the cell-virion combination is a determinant factor for virus entry. This would be consistent with the fact that several receptor candidates have been identified for HCV and that their interactions with one another or with other membrane components might modulate the way they associate with the various virion populations.
This work was supported by grant 01031 from the "Agence Nationale de Recherches sur le SIDA et les hépatites virales" (ANRS) and grant 5869 C from the "Association pour la Recherche sur le Cancer" (ARC) to C.F.-W. Both V.C. and S.M. were supported by an ANRS fellowship. J.D. is an international scholar of the Howard Hughes Medical Institute.
Published ahead of print on 17 October 2007. ![]()
These two authors contributed equally to this work. ![]()
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»