Abstract

The existence of extrahepatic sites of hepatitis C virus (HCV) replication has been proposed as a mechanism responsible for the poor antiviral immune response found in chronic infection. Dendritic cells (DCs), as unique antigen-presenting cells able to induce a primary immune response, are prime targets of persistent viruses. From 24 blood samples obtained from HCV-seropositive patients, peripheral blood DCs (PBDCs) were purified. HCV genomic sequences were specifically detected by reverse-transcription polymerase chain reaction in 6 of 24 PBDC pellets, and replicative-strand RNA also was found in 3 of 24 cell purifications. Analysis of the HCV quasi-species distribution in the PBDC population of 1 patient showed the presence of a dominant variant different from that found in plasma with respect to the primary amino-acid sequence and physicochemical profile of the hypervariable region 1 of glycoprotein E2. These data strongly suggest that PBDCs constitute a reservoir in which HCV replication takes place during natural infection

Hepatitis C virus (HCV) belongs to the Flaviviridae family and is the main etiologic factor of non-A, non-B hepatitis. Its genome is a positive-sense, single-strand RNA with an average length of 9.4 kb, including 2 untranslated regions at the 5′ and 3′ ends and an open-reading frame encoding a large polyprotein that is processed into structural and nonstructural proteins [1]. This virus is highly variable and exists in infected persons as a quasi species that usually consists of a predominant virus variant and a pool of highly related but genetically distinct variants [2]

The World Health Organization estimates that ∼170 million persons are infected with HCV worldwide. Viral infection is characterized by a high rate (50%–70%) of chronicity that, in 20% of cases, eventually leads to liver cirrhosis, with complications that may result in the development of hepatocellular carcinoma [3]. Although immune correlates associated with spontaneous or therapeutic resolutions are not yet clearly defined, recent data suggest that rapid, strong, and maintained Th1 cellular immune responses are needed to control viral infection [4, 5]. In contrast, patients who remain chronically infected either mount an early and vigorous T cell response that temporarily controls viral replication but eventually wanes or do not develop detectable CD4+ and CD8+ T cell–mediated responses during acute infection. This feature strongly suggests that some defects in the priming of the antiviral immune response occur very early during acute infection [4]

Although HCV preferentially infects hepatocytes, evidence of extrahepatic replication sites has accumulated during the past decade. This feature may very likely be related to reinfection of liver grafts after transplantation and to the various autoimmune manifestations frequently associated with HCV infection. This also has been proposed as a potential explanation for HCV persistence in vivo. The infection of cell populations such as dendritic cells (DCs), which play a key role in the induction of both innate and adaptive immune responses, could induce functional defects that may result in an inefficient antiviral immune response [6]. We have shown elsewhere that DCs derived in vitro from monocytes from chronic HCV carriers can be a reservoir for HCV [7]. Furthermore, these cells display a deficient allostimulatory function [7–9 ], which may explain the poor antiviral adaptive immune response. Recent data support the concept that the antigen-specific immune response is largely determined by the innate immunity induced during the initial phase of an infection and involves the secretion of cytokines such as type I interferon (IFN) [10]

In the peripheral blood, 2 major subpopulations of DCs have been described. The myeloid DCs (MDCs) are most likely involved in the capture and presentation of parenterally transmitted pathogens to specific T cells. The second subpopulation, the so-called plasmacytoid DCs (PDCs), was shown to release large amounts of IFN-α/β in response to enveloped viruses and to represent a major link between the innate and adaptive immune response [11, 12]. Infection of these circulating DCs in the setting of HCV infection could have major consequences for disease outcome. In the present study, we sought evidence of HCV replication within the peripheral blood DCs (PBDCs) of patients chronically infected with HCV

Subjects, Materials, and Methods

SubjectsWhole blood (400–500 mL) was collected on anticoagulant from 19 patients with iron overload undergoing therapeutic phlebotomy (table 1). Whole blood from seronegative healthy volunteers (National Blood Center, Lyon, France) served as negative controls in our experiments

Table 1

Clinical characteristics of 19 patients with hepatitis C virus (HCV) infections with iron overload who were undergoing therapeutic phlebotomy

Table 1

Clinical characteristics of 19 patients with hepatitis C virus (HCV) infections with iron overload who were undergoing therapeutic phlebotomy

Purification of PBDCsPeripheral blood mononuclear cells (PBMCs) were isolated by density gradient separation with Ficoll-Paque (Pharmacia). PBDCs were purified from PBMCs by means of the blood DC isolation kit, as recommended by Miltenyi Biotec. In brief, immunomagnetic depletion of T cells, monocytes/macrophages, and NK cells was followed by the isolation of PBDCs from the remaining B cells with use of direct CD4-coupled microbeads. An abnormally low proportion of CD11b+CD11c+ MDCs within the purified PBDC fraction (20% MDCs vs. 80% PDCs) was observed, which was most likely due to the anti-CD11b antibody present in the depletion cocktail. Such an antibody appears to deplete a great proportion of the MDCs that express CD11b at their cell surface. The purity and phenotype of the purified PBDCs were assessed by double-staining with fluorescein isothiocyanate–conjugated mouse monoclonal antibodies (MAbs) against HLA-DR and CD11c and phycoerythrin-conjugated mouse MAbs against CD3, CD20, CD14, CD56, and CD123. A total of 5000 events was registered with use of a Galaxy Flow cytometer (Dako)

Extraction of RNAViral RNA was extracted from 140 μL of plasma or cell pellets, containing 105 to 1.4×106 cells, according to procedures recommended for the QIAmp viral RNA and RNeasy kits, respectively (Qiagen). The RNA pellet was resuspended in 30 μL of RNase-free water

Reverse-transcription polymerase chain reaction (RT-PCR) amplificationFor the detection of HCV positive-strand RNA and β-actin mRNA, cDNA synthesis and PCR amplification were done with the Onestep RT-PCR kit (Qiagen) and the NC4/NC3 and actin 5/actin 3–specific primers described elsewhere [13]. The enzymes contained in this kit, either the Omniscript and Sensiscript for retrotranscription or the HotStarTaq DNA polymerase for PCR, allow highly specific and sensitive detection of RNA. The volume of RNA used for RT-PCR is 2 μL for β-actin detection and varies for the detection of HCV genomic sequences, depending on the different compartments tested: 10 μL of RNA was used for plasma and DCs. The number of PBMCs and contaminating cells used for RT-PCR was adjusted to the number of PBDCs (referred to as “100% equivalent DCs”)

Conditions for RT were as follows: prehybridization at 70°C for 10 min; RT at 50°C for 30 min; denaturation of the reverse transcriptases; and activation of the HotStarTaq DNA polymerase at 95°C for 15 min. Conditions for PCR within the 5′ noncoding region (5′ NCR) and β-actin were as follows: initial PCR cycle at 95°C for 5 min; 40 cycles of 94°C for 50 s, 64°C (5′ NCR) or 63°C (β-actin) for 50 s, and 72°C for 1 min 40 s; and a final extension at 72°C for 10 min. Conditions for PCR within the hypervariable region 1 from E2 (HVR1) were as follows: 95°C for 5 min; 40 cycles of 95°C for 1 min, 60°C for 1 min, and 72°C for 1 min; and a final extension at 72°C for 10 min. Detection of HCV negative-strand RNA was done, as described elsewhere [14], on 10 μL of RNA from the different cellular compartments tested and with use of retrotranscriptase/DNA polymerase thermus thermophilus to avoid false-positive amplifications

A second round of PCR amplification (nested PCR) was done on a 1:10 dilution of the first PCR product with use of inner primers and the High Fidelity Taq DNA polymerase (Invitrogen). Conditions for the nested PCR amplification were described elsewhere [7]. Typically, 10 μL of PCR product was analyzed on agarose gels. Bands were visualized by ethidium bromide staining. The sensitivity of the 1-step RT-PCR technique was evaluated as described elsewhere [15]. This technique allows the detection of 104 copies of HCV genomic RNA by first-round PCR and 103 copies by nested PCR

Quasi-species distributionNested PCR products obtained from plasma and PBDCs of patient 19 were purified from a low–melting-point agarose gel by use of the QIAquick gel extraction kit (Qiagen) and were subcloned in pCR-4 TOPO, according to the manufacturer’s recommendations (TOPO TA Cloning Kit for sequencing; Invitrogen). For each sample, 30 independent colonies were selected. After purification of cDNA (QIAgen Plasmid Mini Kit), the corresponding sequences were determined (Genome Express)

Sequence analysisNucleotide and amino acid multiple sequence alignments were determined with CLUSTALW version 1.6 software [16] with use of the Network Protein Sequence Analysis Web site facilities (http://npsa-pbil.ibcp.fr) [17]. Phylogenetic analyses were done with PHYLIP version 3.5c (J. Felsenstein, 1995) with use of Pasteur Institute Web site facilities (http://bioweb.pasteur.fr/seqanal/phylogeny/phylip-fr)

Results

HCV genomic sequences are detected in PBDCs from HCV-infected patientsFor 19 patients with iron overload, we had the unique opportunity to obtain large blood samples, which allowed purification of the PBDC population. We searched for the presence of HCV genomic and antigenomic RNA in purified PBDCs

Purification of PBDCs from HCV-infected patients PBDCs were successfully isolated from PBMCs, with a yield ranging from 105 to 1.4×106 PBDCs/500 mL of blood. The purity of the PBDCs ranged from 87% to 98% (mean, 94%), contaminating cells being B cells (1%–8%) and a mixture of T cells, NK cells, and monocytes/macrophages (1%–8%; data not shown). In our experimental conditions, the proportions of MDC and PDC subsets in the purified PBDCs were 20% and 80%, respectively (data not shown)

Detection of HCV genomic sequences in PBDCsWe searched for the presence of HCV genomic sequences (positive-strand RNA) in purified PBDCs but also in plasma, PBMCs, B cells, and the fraction containing T lymphocytes, NK cells, and monocytes/macrophages. In these experiments, the number of cells used for the RT-PCR for the different cell populations was adjusted, for each sample, to the number of cells in the PBDC pellet (referred to as “100% equivalent DCs”). Because the maximum proportion of cell contamination observed in the purified PBDC pellets was 8% (see above), to determine whether the signals observed in the PBDC pellets were specifically due to the PBDCs or could be attributed to the contaminating cell populations, we systematically performed PCR amplifications with use of a number of contaminating cells equal to 8% of the respective PBDC fractions tested (referred to as 8% equivalent DCs). Data are summarized in table 2. From these data, different detection patterns could be identified

Table 2

Summary of reverse-transcription polymerase chain reaction (RT-PCR) analysis of hepatitis C virus (HCV) genomic sequences in different peripheral blood cell populations

Table 2

Summary of reverse-transcription polymerase chain reaction (RT-PCR) analysis of hepatitis C virus (HCV) genomic sequences in different peripheral blood cell populations

For patients who were long-term responders at the time of blood sampling (samples 13.3, 13.4, 17, and 18), no viral sequences were detected either in the plasma or in the other cellular compartments, including PBDCs (table 2), as expected

For 17 of 24 samples (samples 1–4, 6–12, 14–16, and 19), genomic sequences were detected in the PBDCs but also in all cellular compartments tested, including plasma (table 2; 100% equivalent DCs). For 3 of 17 samples (samples 1, 3.2, and 19), we could demonstrate that viral sequences retrieved from the PBDC pellets were effectively contained in this subpopulation, because no signal could be detected in the 8% of contaminating cells (see table 2; 8% equivalent DCs). In the remaining 14 cases, it was not possible to prove the origin of the viral sequences found in the PBDC fraction, because contaminating cells also contained HCV genomic sequences (see table 2; 8% equivalent DCs). Finally, for the remaining 3 of 24 samples (samples 5, 13.1, and 13.2), a signal corresponding to the genomic RNA was specifically observed only in the PBDC fraction (table 2; 100% equivalent DCs) and in no other compartments. Altogether, these data show that HCV genomic sequences could be unambiguously detected in the circulating DCs purified from 6 of 24 samples from HCV-infected persons (table 2)

Replicative intermediate is present in PBDCsAlthough our data suggest that HCV can infect circulating DCs, we wondered whether viral replication could take place in these cells. Therefore, we searched for the presence of HCV negative-strand RNA, which is considered to be the replicative intermediate, in PBDC samples in which we were able to detect HCV positive-strand RNA (n=20; samples 1–16 and 19), as well as in the different fractions of contaminating cells. Results are detailed in table 3. Negative-strand RNA was detected in PBDCs from 3 samples (samples 4, 13.2, and 19) and also was detected in the B cell fraction from samples 9, 16, and 19 (table 3)

Table 3

Summary of reverse-transcription polymerase chain reaction (RT-PCR) analysis of hepatitis C virus (HCV) antigenomic sequences in different peripheral blood cell populations

Table 3

Summary of reverse-transcription polymerase chain reaction (RT-PCR) analysis of hepatitis C virus (HCV) antigenomic sequences in different peripheral blood cell populations

The presence of negative-strand RNA in PBDCs of samples 13.2 and 19 confirms the results of the positive-strand detection (table 2). For sample 4, the exclusive detection of negative-strand RNA in the PBDCs provides evidence that the sequences retrieved from the PBDC pellet were not caused by contaminating cells

Specific virus quasi species are retrieved from PBDCs To characterize the nature of the viral genomes detected in the PBDCs, we analyzed the quasi-species distribution of the virions detected in the plasma and the PBDC fraction from patient 19 in 2 regions of the HCV genome: the 5′ NCR, because of its function as an internal ribosome entry site [18], and the HVR1 of the E2 envelope protein, because of its potential role in the interaction between the virus and a cellular receptor. Unfortunately, the lack of available cells for sample 4, as well as the absence of HCV sequences in the plasma of sample 13.2, precluded this analysis for these 2 samples containing replicative HCV sequences in PBDCs

From analysis of the 5′ NCR, we could not find any difference between the quasi species from the PBDCs and that from the plasma. However, from the analysis of this sequence and a sequence of 80 nt from the E1 region, we were able to assert that all sequences from patient 19 originated from a single HCV strain, because >90% identity between all of the sequences could be observed (data not shown). In contrast, a clear-cut dichotomy was revealed both by the phylogenetic (data not shown) and sequence analysis of HVR1 (nt 1491–1572) (figure 1). The 56 sequences encoded 8 different HVR1 amino acid sequences, which separated into 2 closely related groups (figure 1A). The variants of group 1 (S1–S3) mainly fit the consensus sequence and included the 27 clones derived from the PBDCs plus 4 clones derived from plasma. In contrast, sequences from group 2 (S4–S8) were observed in the plasma but never in the PBDCs. Consequently, variants of group 1 seemed to be preferentially associated with the DC type. Although variants of group 2 were quite homogeneous, they exhibited ∼60% amino acid changes, compared with those of group 1 (figure 1A). However, this amino acid variability is fully compatible with the HVR1 hydropathy pattern defined elsewhere [19] (figure 1B) and with the structural “s” and functional “f” positions assumed to be mostly involved in the structural and functional features, respectively, of HVR1

Figure 1

Analysis of hepatitis C virus (HCV) hypervariable region 1 (HVR1) quasi species based on amino acid sequences of 29 clones from plasma and 27 clones from peripheral blood dendritic cells (PBDCs) of patient 19. A Alignment of 8 different sequences (denoted S1–S8) showing amino acid changes. Consensus corresponds to most frequent amino acid sequence derived from alignment of all clones (56 sequences). Amino acids identical to consensus sequence are represented by a dash (−). No. of clones (n) exhibiting particular sequence in plasma and/or in PBDCs is given in the 2 right-hand columns. B HVR1 consensus hydropathy pattern established by Penin et al. [19]. This consensus was derived from analysis of 1382 HVR1 sequences reported in the European Molecular Biology Laboratory (EMBL) database. Hydrophobic (o), neutral (n), hydrophilic (i), and variable (v) positions are presented. G is a fully conserved Gly residue. s/f pattern summarizes the putative main role of each HVR1 position, with “s” and “f” meaning structural and functional, respectively. C Functional amino acid sequence signatures. To highlight main amino acid differences in “f” positions among 8 sequences reported in panel A residues in “s” positions are represented by a dot ( &b.dot;), whereas identical and very similar residues at given “f” position in all sequences are represented by a colon ( : ). Sequences exhibiting the same amino acid signature are grouped. Basic amino acids are underlined. Nucleotide sequences were deposited in EMBL (accession nos.: plasma sequence, AJ422080; DC sequence, AJ422081)

Figure 1

Analysis of hepatitis C virus (HCV) hypervariable region 1 (HVR1) quasi species based on amino acid sequences of 29 clones from plasma and 27 clones from peripheral blood dendritic cells (PBDCs) of patient 19. A Alignment of 8 different sequences (denoted S1–S8) showing amino acid changes. Consensus corresponds to most frequent amino acid sequence derived from alignment of all clones (56 sequences). Amino acids identical to consensus sequence are represented by a dash (−). No. of clones (n) exhibiting particular sequence in plasma and/or in PBDCs is given in the 2 right-hand columns. B HVR1 consensus hydropathy pattern established by Penin et al. [19]. This consensus was derived from analysis of 1382 HVR1 sequences reported in the European Molecular Biology Laboratory (EMBL) database. Hydrophobic (o), neutral (n), hydrophilic (i), and variable (v) positions are presented. G is a fully conserved Gly residue. s/f pattern summarizes the putative main role of each HVR1 position, with “s” and “f” meaning structural and functional, respectively. C Functional amino acid sequence signatures. To highlight main amino acid differences in “f” positions among 8 sequences reported in panel A residues in “s” positions are represented by a dot ( &b.dot;), whereas identical and very similar residues at given “f” position in all sequences are represented by a colon ( : ). Sequences exhibiting the same amino acid signature are grouped. Basic amino acids are underlined. Nucleotide sequences were deposited in EMBL (accession nos.: plasma sequence, AJ422080; DC sequence, AJ422081)

To further characterize the putative functional differences between the 2 groups of HVR1 sequences, only variable residues in “f” positions are reported in figure 1C. The resulting amino acid sequence signature indicates that sequences retrieved from the PBDCs (group 1) exhibit mainly polar and positively charged residues (H, R, S, and Q), whereas hydrophobic amino acids (I, Y, A, and V) predominate in sequences from plasma (group 2)

In conclusion, both the presence of negative-strand RNA and the preferential quasi-species distribution found in PBDCs strongly suggest not only that HCV can replicate in PBDCs but also that the virus could display a preferential tropism for this subpopulation

Discussion

The mechanisms that determine the outcome of HCV infection are not well understood, although it is widely assumed that dysfunctions of the cellular immune response may play an important role. Accumulating reports suggest that these defects could be directly related to the capacity of HCV to replicate in extrahepatic compartments and, more particularly, in cells from the immune system. One major driver of an antiviral immune response is the DC population. In peripheral blood, 2 major subsets have been defined, namely the MDC and PDC subsets. The latter has been recently shown to link innate and adaptive immune responses, essentially through the production of IFN-α [20, 21]. The present study was designed to search for evidence of HCV replication in circulating DCs from chronically infected patients

To address this issue, we searched for the presence of HCV genomic and antigenomic RNA in PBDCs purified from peripheral blood from HCV-seropositive patients. In addition, we tried to define the nature of such genomes through the analysis of their quasi-species distribution. We report here that HCV is present and replicates in PBDCs

The conclusion that HCV replicates in PBDCs is supported by 2 sets of observations. First, HCV replicative-strand RNA was detected in PBDCs from 3 of 24 samples. To date, the only virus shown to replicate in isolated PBDCs ex vivo is human immunodeficiency virus type 1 (HIV-1) [22]. Patterson et al. [22] reported that both lymphotropic and macrophage-tropic strains can replicate in both myeloid and plasmacytoid subsets of DCs but do so much more efficiently in the latter. The second piece of evidence is the presence of a specific dominant HCV variant in PBDCs, which was identified by the analysis of the quasi-species distribution of HCV sequences found in PBDCs, compared with that found in the plasma of patient 19. Sixty percent of the amino acids of the HVR1, located within the glycoprotein E2, differ between the quasi species found in the PBDCs and those present in the plasma of this patient. Viral envelopes are a major determinant of virus tropism, because they usually mediate the interaction with a receptor on the host cell and eventually virus entry [23]. In the HIV-1 model, Patterson et al. [24] analyzed virus variants found in circulating DCs and compared them with those infecting CD3+ T cells. Although an overall pattern across the entire group of patients could not be defined, for 1 patient, 2 amino acid positions within the V3 loop of the envelope clearly differed between the 2 groups of sequences, indicating the presence of unique virus variants in the DC population [24]. In our study, the sequences retrieved from circulating DCs, compared with those obtained from plasma, displayed amino acid variations that were essentially targeted to the variable positions of the HVR1 (figure 1B). These positions are most likely involved in molecular interactions with other viral and/or cellular partners and the antigenicity of E2 [19]. Recent data suggest that HVR1 may play an important role in viral attachment and thus possibly in cell tropism [19, 25]. Although our data are still preliminary, because they were obtained from a single patient, they show a dichotomy of HVR1 quasi species between plasma and PBDCs, which suggests that the 2 groups of HCV variants most likely interact with distinct cellular types. This hypothesis is in accordance with the fact that all HCV circulating sequences most likely originate from viruses replicating within the liver [26]. Therefore, it should not be surprising that HCV uses distinct ligands when infecting either DCs or hepatocytes. HVR1 is intrinsically a basic region [19], and our observations indicate that, at the putative functional amino acid positions (figure 1C), the major quasi species from the PBDCs contains a higher number of positively charged amino acids than that from the plasma. Changes in HVR1 physicochemical properties might affect the binding of HCV to its cellular receptor(s). Thus far, 3 putative receptors, namely glycosaminoglycans, CD81, and the low-density lipoprotein receptor, have been shown to bind HCV particles or envelope proteins [27–29 ]. Overall, our results provide the first evidence of HCV replication in PBDCs. Moreover, a longitudinal study performed on patient 13 suggests that PBDCs could be one of the last reservoirs of the virus in peripheral blood during response to therapy (data not shown)

Experiments are currently ongoing to investigate whether HCV can replicate in MDCs and/or PDCs and the functional consequences of such an infection. We and others have previously shown that monocyte-derived DCs from chronically HCV-infected patients display an impaired allostimulatory function [7–9 ]. Although the circulating MDC population may not arise from a monocyte precursor in vivo, infection of MDCs by HCV could have similar consequences for the antigen-presentation function of these cells, which will most likely further affect the antiviral immune response. Finally, the infection of circulating PDCs could have detrimental effects on their capacity to release IFN-α and to polarize the antiviral immune response

Acknowledgments

We would like to acknowledge the clinicians from the Hôtel Dieu in Lyon, for providing blood samples, and Claude Vieux and Benoît Ligeoix, for collecting clinical information

References

1
Houghton
M
Fields
BN
Knipe
DM
Howley
PM
, et al.  . 
Hepatitis C viruses
Fields virology
 , 
1996
3rd ed
Philadelphia
Lippincott-Raven
(pg. 
1035
-
58
)
2
Simmonds
P
Viral heterogeneity of the hepatitis C virus
J Hepatol
 , 
1999
, vol. 
31
 (pg. 
54
-
60
)
3
Seeff
LB
Natural history of hepatitis C
Am J Med
 , 
1999
, vol. 
107
 
Suppl 6B
(pg. 
S10
-
5
)
4
Gerlach
JT
Diepolder
HM
Jung
MC
, et al.  . 
Recurrence of hepatitis C virus after loss of virus-specific CD4+ T-cell response in acute hepatitis C
Gastroenterology
 , 
1999
, vol. 
117
 (pg. 
933
-
41
)
5
Lechner
F
Gruener
NH
Urbani
S
, et al.  . 
CD8+ T lymphocyte responses are induced during acute hepatitis C virus infection but are not sustained
Eur J Immunol
 , 
2000
, vol. 
30
 (pg. 
2479
-
87
)
6
Reis e Sousa
C
Dendritic cells as sensors of infection
Immunity
 , 
2001
, vol. 
14
 (pg. 
495
-
8
)
7
Bain
C
Fatmi
A
Zoulim
F
Zarski
JP
Trépo
C
Impaired allostimulatory function of dendritic cells in chronic hepatitis C infection
Gastroenterology
 , 
2001
, vol. 
120
 (pg. 
512
-
24
)
8
Auffermann-Gretzinger
S
Keeffe
EB
Levy
S
Impaired dendritic cell maturation in patients with chronic, but not resolved, hepatitis C virus infection
Blood
 , 
2001
, vol. 
97
 (pg. 
3171
-
6
)
9
Kanto
T
Hayashi
N
Takehara
T
, et al.  . 
Impaired allostimulatory capacity of peripheral blood dendritic cells recovered from hepatitis C virus–infected individuals
J Immunol
 , 
1999
, vol. 
162
 (pg. 
5584
-
91
)
10
Guidotti
LG
Rochford
R
Chung
J
Shapiro
M
Purcell
R
Viral clearance without destruction of infected cells during acute HBV infection
Science
 , 
1999
, vol. 
284
 (pg. 
825
-
9
)
11
Cella
M
Jarrossay
D
Facchetti
F
, et al.  . 
Plasmacytoid monocytes migrate to inflamed lymph nodes and produce large amounts of type I interferon
Nat Med
 , 
1999
, vol. 
5
 (pg. 
919
-
23
)
12
Siegal
FP
Kadowaki
N
Schodell
M
, et al.  . 
The nature of the principal type 1 interferon–producing cells in human blood
Science
 , 
1999
, vol. 
284
 (pg. 
1835
-
7
)
13
Lerat
H
Berby
F
Trabaud
MA
, et al.  . 
Specific detection of hepatitis C virus minus strand RNA in hematopoietic cells
J Clin Invest
 , 
1996
, vol. 
97
 (pg. 
845
-
51
)
14
Negro
F
Giostra
E
Krawczynski
K
, et al.  . 
Detection of intrahepatic hepatitis C virus replication by strand-specific semi-quantitative RT-PCR: preliminary application to the liver transplantation model
J Hepatol
 , 
1998
, vol. 
29
 (pg. 
1
-
11
)
15
Besnard
NC
Andre
PM
Automated quantitative determination of hepatitis C virus viremia by reverse transcription–PCR
J Clin Microbiol
 , 
1994
, vol. 
32
 (pg. 
1887
-
93
)
16
Thompson
JD
Higgins
DG
Gibson
TJ
CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice
Nucleic Acids Res
 , 
1994
, vol. 
22
 (pg. 
4673
-
80
)
17
Combet
C
Blanchet
C
Geourjon
C
Deléage
G
NPS@: network protein sequence analysis
Trends Biochem Sci
 , 
2000
, vol. 
25
 (pg. 
147
-
50
)
18
Lemon
SM
Honda
M
Internal ribosome entry sites within the RNA genomes of hepatitis C virus and other flaviviruses
Semin Virology
 , 
1997
, vol. 
8
 (pg. 
274
-
88
)
19
Penin
F
Combet
C
Germanidis
G
Frainais
PO
Deleage
G
Conservation of the conformation and positive charges of hepatitis C virus E2 envelope glycoprotein hypervariable region 1 points to a role in cell attachment
J Virol
 , 
2001
, vol. 
75
 (pg. 
5703
-
10
)
20
Cella
M
Facchetti
F
Lanzavecchia
A
Colonna
M
Plasmacytoid dendritic cells activated by influenza virus and CD40L drive a potent Th1 polarization
Nat Immunol
 , 
2000
, vol. 
1
 (pg. 
305
-
10
)
21
Kadowaki
N
Antonenko
S
Lau
JYN
Liu
YJ
Natural interferon α/β–producing cells link innate and adaptive immunity
J Exp Med
 , 
2000
, vol. 
192
 (pg. 
219
-
25
)
22
Patterson
S
Rae
A
Hockey
N
Gilmour
J
Gotch
F
Plasmacytoid dendritic cells are highly susceptible to human immunodeficiency virus type 1 infection and release infectious virus
J Virol
 , 
2001
, vol. 
75
 (pg. 
6710
-
3
)
23
Schneider-Schaulies
J
Cellular receptors for viruses: links to tropism and pathogenesis
J Gen Virol
 , 
2000
, vol. 
81
 (pg. 
1413
-
29
)
24
Patterson
S
English
NR
Longhurst
H
, et al.  . 
Analysis of human immunodeficiency virus type 1 (HIV-1) variants and levels of infection in dendritic and T cells from symptomatic HIV-1–infected patients
J Gen Virol
 , 
1998
, vol. 
79
 (pg. 
247
-
57
)
25
Hamaia
S
Li
C
Allain
JP
The dynamics of hepatitis C virus binding to platelets and 2 mononuclear cell lines
Blood
 , 
2001
, vol. 
98
 (pg. 
2293
-
300
)
26
Cabot
B
Martell
M
Esteban
JI
, et al.  . 
Longitudinal evaluation of the structure of replicating and circulating hepatitis C virus quasispecies in nonprogressive chronic hepatitis C patients
J Virol
 , 
2001
, vol. 
75
 (pg. 
12005
-
13
)
27
Pileri
P
Uematsu
Y
Campagnoli
S
, et al.  . 
Binding of hepatitis C virus to CD81
Science
 , 
1998
, vol. 
282
 (pg. 
938
-
41
)
28
Garson
JA
Lubach
D
Passas
J
Whitby
K
Grant
PR
Suramin blocks hepatitis C binding to human hepatoma cells in vitro
J Med Virol
 , 
1999
, vol. 
57
 (pg. 
238
-
42
)
29
Meyer
K
Basu
A
Ray
R
Functional features of hepatitis C virus glycoproteins for pseudotype virus entry into mammalian cells
Virology
 , 
2000
, vol. 
276
 (pg. 
214
-
26
)
Presented in part: 8th International Symposium on Hepatitis C Virus and Related Viruses (Molecular Virology and Pathogenesis), Paris, 2–5 September 2001 (abstract O57); 52nd annual meeting of the American Association for the Study of Liver Diseases, Dallas, 9–13 November 2001 (abstract 1244); Congrès du Club Francophone des Cellules Dendritiques, Paris, 10–11 December 2001 (abstract C10); 7th International Symposium on Dendritic Cells, Bamberg, Germany, 19–24 September 2002 (abstract P214)
Informed consent was obtained from all individuals before they were included in the present study
Financial support: Association pour la Recherche sur le Cancer; Agence Nationale pour la Recherche sur le SIDA; Ministère de l’Education Nationale de la Recherche et Technologie Fellowship from the French government (to N.G.)