Establishment of persistent hepatitis C virus infection and ... - CiteSeerX

Journal of General Virology (1997), 78, 2467–2476. Printed in Great Britain ...........................................................................................................................................................................................................................................................................................

Establishment of persistent hepatitis C virus infection and replication in vitro Stefanie Seipp,1 Hubert M. Mueller,2 Eberhard Pfaff,2 Wolfgang Stremmel,1 Lorenz Theilmann1 and Tobias Goeser1 1 2

Department of Internal Medicine, University of Heidelberg, Bergheimerstr. 58, D-69115 Heidelberg, Germany Federal Research Center for Virus Diseases of Animals, Paul-Ehrlich-Str. 28, D-72076 Tu$ bingen, Germany

Hepatitis C virus (HCV) is a major cause of chronic viral hepatitis. Development of anti-viral strategies has been hampered by the lack of efficient cell systems to propagate HCV in vitro. To establish a long-term culture system, we tested human hepatoma (HuH7, HepG2) and porcine non-hepatoma (PK15, STE) cell lines, as well as several culture and infection conditions. As a marker for virus replication, minus-strand HCV RNA in infected cells was detected by an enhanced detection system using nested RT–PCR followed by hybridization analysis. Short-term efficiency of HCV infection (10 days) was slightly increased by addition of polyethylene glycol (PEG) and/or dimethyl sulfoxide (DMSO) to culture media during inoculation of HuH7, PK15 and STE cells, but no augmentation in long-term culture was achieved, suggesting enhanced attachment of HCV to cells rather than more efficient infection. A

Introduction The biological characteristics of hepatitis C virus (HCV), the major causative agent of non-A, non-B hepatitis (Kuo et al., 1989), remain obscure. Although there has been good progress with molecular analysis (Wang et al., 1994 ; Bartenschlager et al., 1995), and diagnostic tools have been well established (Lunel et al., 1995 ; Roth et al., 1996), the mechanisms leading to persistent infection are still unclear. The liver is the main target for replication of HCV in vivo (Nouri-Aria et al., 1993 ; Gastaldi et al., 1995), which occurs via minus-strand RNA as replicative intermediate (Houghton et al., 1991). In addition, the presence of both HCV RNA molecules Author for correspondence : Stefanie Seipp. Fax 49 6221 564922. e-mail StefaniejSeipp!krzmail.krz.uniheidelberg.de

0001-4757 # 1997 SGM

stabilizing effect on HCV propagation was observed for 50 days in a serum-free medium with stimulation of the low-density lipoprotein (LDL) receptor expression by lovastatin. Using partially serum-free culture conditions, long-term persistence of HCV in cells and release of virions into supernatant was achieved for up to 130 days. Infectivity of released virions in supernatants after long-term culturing (day 30–80) was shown by successful infection of fresh cells. In conclusion, supplementation with PEG, DMSO and lovastatin during inoculation did not enhance virus replication substantially, but continued stimulation of LDL-receptor expression resulted in infections which persisted for over 4 months. These data support the hypothesis of an LDL-receptor mediated uptake of HCV into cells in vitro.

in peripheral blood mononuclear cells (PBMC) of patients infected with HCV can be demonstrated (Zignego et al., 1992 ; Mueller et al., 1993 ; Mihm et al., 1996), suggesting that replication occurs in these cells also. The lack of a reliable cell culture system allowing persistent in vitro virus propagation is still hampering screening of antiviral strategies. Some cell lines, particularly of lymphoid origin, are susceptible to HCV infection and permissive for HCV RNA replication (Bertolini et al., 1993 ; Iacovacci et al., 1993 ; Lanford et al., 1994 ; Nissen et al., 1994 ; Shimizu & Yoshikura, 1994 ; Cribier et al., 1995 ; Kato et al., 1995 ; Mizutani et al., 1996 ; Nakajima et al., 1996 ; Shimizu et al., 1996). Although virus production has been achieved by long-term culture of primary hepatocytes of infected patients (Ito et al., 1996), efforts to propagate the virus by infection of adherent cells such as hepatoma cell lines have so far been unsatisfactory. Based on previous results of HCV in vitro infections we systematically

CEGH

S. Seipp and others

tested several approaches to establish reproducible in vitro infection systems for HCV. HepG2 and HuH7 human hepatoma cell lines were chosen because of the hepatic tropism of HCV in vivo. PK15 cells (porcine kidney cell line) were used based on data indicating the putative susceptibility of this cell line for HCV in vitro infection (M. Beach, personal communication). Since our preliminary results had revealed STE cells (porcine testis cell line) to be non-permissive for HCV infection, this cell line was chosen as a negative control. Successful infection and propagation of the virus was assessed by detection of HCV plus- and minus-strand RNA using nested RT–PCR with specific primers. To improve in vitro infections, the effects of polyethylene glycol (PEG) and dimethyl sulfoxide (DMSO) were investigated, as these substances have been shown to increase the efficiency of in vitro infection with other viruses such as hepatitis B virus (Gripon et al., 1988, 1993), Sendai virus (Hoekstra et al., 1989), herpes simplex virus types 1 and 2 (Subramanian et al., 1995) and mouse hepatitis virus (Kooi et al., 1991). Since HCV was reported to be associated with very low density lipoprotein (VLDL) fractions (Thomssen et al., 1992), cell culture methods stimulating expression of LDL receptors (Havekes et al., 1983 ; Sviridov et al., 1990 ; Kostner, 1993) were examined as well. These various conditions were compared to the effect of a highly supplemented culture medium that was reported to allow establishment of a long-term, persistently HCV transfected HuH7 cell line (Yoo et al., 1995).

Methods + Cell culture. The following cell lines were used : PK15, porcine kidney cell line (1) ; STE, swine testis cell line (2) ; HepG2 and HuH7, human hepatoma cell lines (1). Cells were grown in six-well plates with DMEM–10 % FCS (1) or Liebowitz–EMEM medium supplemented with 10 % FCS, tryptose phosphate and essential amino acids (2). The cells were subcultured every 3 days using 1 % EDTA–trypsin. For long-term approaches, cells of all lines were cultured in FCS-free supplemented Ham’s F12 medium as described by Yoo et al. (1995). Following subculture, cells were grown to semi-confluence in standard medium with

10 % FCS ; the medium was then changed to Ham’s F12 and cells were grown further for 2–6 days (one generation equals 6–10 days). + Sera. Sera were taken from HCV-positive patients and immediately stored at ®20 °C. Serum 1 : female patient aged 69, HCV genotype 1b, virus titre 2¬10&}ml, chronic active hepatitis with cirrhosis. Serum 2 : female patient aged 25, HCV genotype 1a, virus titre 3¬10&}ml, chronic persistent hepatitis. Serum 3 : male patient aged 53, HCV genotype 1b, virus titre 8¬10& per ml, chronic persistent hepatitis. HCV genotypes were determined by the Innolipa system (Innogenetics). Virus titres in infectious sera were determined by the Amplicore Monitor test system (Hoffmann-La Roche). + In vitro infections. Cells were plated in six-well plates and grown to semi-confluence with 2 ml standard medium plus 10 % FCS. Preincubations were carried out by adding the following supplements to fresh medium (standard media or FCS-free Ham’s F12) : PEG to a final concentration of 4 % ; DMSO to a final concentration of 1±5 or 2 % when combined with PEG ; and lovastatin to a final concentration of 10 µM (Table 1). Cells were further incubated in 2 ml supplemented medium for 48 h at 37 °C. Cell layers were washed twice with FCS-free medium (standard or Ham’s F12) and infections were performed with 250 µl HCVpositive serum and 250 µl FCS-free supplemented medium per well for 90 min (Table 1, step 1). Supplemented medium and FCS were then added to 2 ml and 10 %, respectively, and cells were further incubated overnight at 37 °C (Table 1, step 2). At day 1, cells were washed three times with culture medium and incubation was continued in standard or FCS-free Ham’s F12 medium with daily medium changes. All inocula were retested for HCV RNA and found positive. Cell lines infected in standard media were subcultured on days 2, 5, 8, 11, 14, 18, 23, 26, 30, 33 and 37 (1 : 3) and harvested daily (days 1–10) or at each generation. Cell lines infected in Ham’s F12 medium were subcultured on days 2, 8, 16, 24, 31, 41, 51, 61, 71, 80, 89, 100, 108, 115 and 122 (1 : 3) and harvested at each cell generation. To test for infectivity of released virions, 1 ml culture medium was sonicated for 20 s and then clarified by centrifugation (5000 r.p.m. for 5 min). Fresh cells of the same cell line were plated in six-well dishes, grown to semi-confluence and inoculated overnight with the clarified infectious culture medium. On day 1, cells were washed three times and incubation was continued for 7 days, as described above. These cells were subcultured at day 3 (1 : 3), and medium and cells were harvested at days 3 and 7. For RNA extraction, culture media were centrifuged for 20 s and supernatants stored at ®70 °C. Cells were washed twice in cold PBS, lysed in 500 µl lysis buffer for 20 min (Nucleon total-RNA kit, Scotlab) and stored at ®70 °C.

Table 1. Parameters for preincubation and infection Supplement (48 h)

Preincubation (90 min)

Infection step 1 (12 h)

Infection step 2 (overnight)

PEG* DMSO* PEG}DMSO*

– 1±5 % DMSO 2 % DMSO

FCS-free Lovastatin

FCS-free standard medium FCS-free standard medium 10 µM lovastatin

FCS-free, 4 % PEG FCS-free, 1±5 % DMSO FCS-free, 4 % PEG 2 % DMSO FCS-free –

4 % PEG 1±5 % DMSO 4 % PEG 2 % DMSO – –

* Used with standard media or FCS-free Ham’s F12 medium.

CEGI

HCV in vitro infection

Fig. 1. Quantification of HCV plus-strand RNA. Results of nested RT–PCR and hybridization signals of in vitro transcribed HCV plus-strand RNA. Standardized amounts of in vitro transcribed RNA were used in RT–PCR reactions. Nested RT–PCR : staining with ethidium bromide in a 2 % agarose gel. Southern blot : hybridization of 10 µl product of RT–PCR (PCR 1) with a fluorescently labelled GBV-C probe. Dot blot : hybridization of 1 µl product of nested RT–PCR (PCR 2). Non-radioactive hybridization was carried out as described. Sensitivity of nested RT–PCR detected in agarose gels was 2¬103 molecules template RNA. Maximum sensitivity of RT–PCR plus Southern blot hybridization was 2¬102 molecules template RNA/RT, equivalent to 103 HCV g.e./106 cells, or 5¬103 g.e. RNA/ml culture medium. Maximum sensitivity of dot blotted nested RT–PCR was 2¬101 molecules template RNA/RT (102 g.e./106 cells or 5¬102 g.e./ml culture medium).

Fig. 2. Short-term HCV infection of HuH7, HepG2, PK15 and STE cell lines in FCS-containing standard medium. Results of nested RT–PCR. Hepatoma cell lines were infected with serum 2, non-hepatoma cell lines with serum 1. M, medium ; C, cells. Supplementations (see Table 1) : –, no supplementation ; PD, PEGDMSO ; P, PEG ; D, DMSO ; lov, lovastatin. Results : i, inoculummedium ; shaded area, nested RT–PCR positive for plus-strand RNA ; –, nested RT–PCR positive for minus-strand RNA ; X, not detected. Days (d) in Arabic, cell generations (gen) in Roman numerals.

CEGJ

S. Seipp and others

Fig. 3. Long-term in vitro infection of cell lines using FCS-free Ham’s F12 medium and serum 3. Results of nested RT–PCR. M, medium ; C, cells. Supplementations (see Table 1) : –, no supplementation ; PD, PEGDMSO ; P, PEG ; D, DMSO ; lov, lovastatin ; FCS, FCS-free standard medium. Results : i, inoculummedium ; shaded area, nested RT–PCR positive for plus-strand RNA ; –, nested RT–PCR positive for minus-strand RNA ; D, supernatants used for infections of fresh cells ; X, not detected, sec., secondarily infected cells. Days (d) in Arabic, cell generations (gen) in Roman numerals.

+ HCV RNA extraction. RNA from one well of a six-well plate (approx. 10' cells) infected in standard media was extracted by the guanidinium isothiocyanate–phenol–chloroform method described by Chomczynski & Sacchi (1987) and dissolved in 40 µl diethyl pyrocarbonate (DEPC)-treated water. RNA from 10' cells in vitro infected in Ham’s F12 medium was extracted by the nucleon total-RNA kit (Scotlab) and dissolved in 50 µl DEPC-treated water to obtain equal amounts of RNA per 10 µl aliquot. RNA from 140 µl culture medium and serum, respectively, was isolated by the QIAamp HCV–RNA extraction kit following the manufacturer’s instructions (QIAgen) and dissolved in 50 µl DEPC-treated water. + Detection of plus- and minus-strand RNA by nested RT–PCR. Strand-specific RNA was detected by nested RT–PCR resulting in an amplified fragment of about 300 bp. Primers used for amplification were specifically designed to identify the 5’UTR}core region, as reported previously by our group (Mueller et al., 1993). For highly specific detection of minus-strand RNA, a tagged primer (tag28, nt 14–33, 5« TAGCGGATGGCGAATAAGCTTTGGGGGCGACACTCCACCAT) and the tag 5« TAGCGGATGGCGAATAAGCTT were used in the RT reaction and PCR 1, respectively (Lanford et al., 1994). RT was performed in a volume of 20 µl at 42 °C for 1 h using 10 µl heat-denatured RNA, 25 pmol primer (29® or tag28) and 3 units AMV reverse transcriptase (Stratagene). Both amplification reactions were performed in a volume of 50 µl using the following protocol : PCR 1 comprised 1 µl of first-strand DNA, 12±5 pmol of each primer (plus strand, 2829 ; minus strand, tag29), 1 unit Taq-Polymerase (Stratagene), plus 1 unit Taq-Extender (Stratagene) for detection of minus-strand RNA (cycles : 1 min 93 °C, 1 min 45 °C, 2 min 72 °C, 10 min last-step delay, 40 cycles). PCR 2 comprised 2±5 µl PCR 1, 12±5 pmol of each primer (primers 23 and 24 for detection of both RNA molecules), and 1 unit Taq-Polymerase (Stratagene) (cycles : 1 min 93 °C, 1 min 60 °C, 2 min 72 °C, 10 min laststep delay, 30 cycles). Amplified products of PCR 2 were visualized by ethidium bromide staining in a 2 % agarose gel.

CEHA

+ Enhanced detection of HCV RNA by additional nonradioactive hybridization. For enhanced detection of minus-strand RNA, 40 µl of nested PCR products were precipitated and Southern blotted on Hybond N membranes (USB-Amersham) following a standard protocol : 10¬SSC (saline sodium citrate), overnight. Plusstrand RNA was detected by dot blot hybridization of 1 µl nested PCR product on a Hybond N membrane. The following controls were included to ensure specificity : nested RT–PCR of RNA extracted from (a) positive serum, (b) negative serum, (c) uninfected cells, (d) FCS and (e) culture medium ; nested RT–PCR of positive RNA without primers in RT and}or PCR reaction ; and nested RT–PCR and PCR amplifications without template. Procedures recommended for avoiding contamination were strictly followed. Hybridizations were carried out using the nonradioactive Gene Images kit (USB-Amersham). DNA amplified by nested RT–PCR using RNA from the sera used for in vitro infections was gelpurified and 60 ng eluted DNA was fluorescein-labelled at 37 °C for 12 h according to the manufacturers’ instructions. Five ng labelled DNA per ml hybridization buffer (5¬SSC) was used for hybridization at 60 °C for 16 h. Filters were washed twice at 60 °C in 1¬SSC–0±1 % SDS and 0±5¬SSC–0±1 % SDS, and incubated in 10 % blocking reagent for 1 h. Chemoluminescence detection was performed using a polyclonal alkaline phosphatase–anti-fluorescein antibody (USB-Amersham, RPN 3311) in a concentration of 1 : 50 000 for 1 h. Membranes were washed three times in diluent buffer (100 mM Tris–HCl, pH 7±5, 300 mM NaCl) containing 0±3 % Tween 20 and incubated in detection reagent for 4 min. Exposure lasted for 20 min to 6 h on day 0 or day 1. The sensitivity of the system was determined as described below. + RNA quantification of in vitro infections. The amount of plusstrand RNA from cells and supernatants was measured semi-quantitatively. Plus-strand RNA was transcribed in vitro from a cloned fragment of the HCV genome, encompassing the 5«UTR, core region and the 5« end of E1 (unpublished data) with nucleotides 155–270 deleted. RNA was quantified by ethidium bromide staining and standardized

HCV in vitro infection (a)

(b)

Fig. 4. Long-term infection of cell lines cultured in supplemented Ham’s F12 medium. (a) Results of dot blot (plus-strand RNA) and Southern blot (minus-strand RNA) hybridization of nested RT–PCR products (PCR results and legend shown in Fig. 2) Results : i, inoculummedium ; shaded area, nested RT–PCR positive for plus-strand RNA ; –, nested RT–PCR positive for minusstrand RNA ; D, supernatants used for infections of fresh cells ; X, not detected, sec., secondarily infected cells. (b) Southern blot hybridization of minus-strand nested RT–PCR with tagged primers (40 µl precipitated amplification product of PCR 2). Hybridization signals of nested RT–PCR of two cell lines (STE and HuH7) and two control reactions are shown. , positive RNA ; ®, negative RNA. HuH7 cells were preincubated with FCS-free medium and lovastatin prior to infection. Only the PCR product derived from HuH7 cell lysates of generation VIII could be stained positive for HCV minus-strand RNA by ethidium bromide after gel electrophoresis. Exposure, 3 h on day 1.

amounts of RNA (‘ standard RNA ’, 2¬10!–2¬10( copies) were reverse transcribed and amplified using the method described above. Amplified products from nested RT–PCR reactions of RNA isolated from infected cells and supernatants (‘ test-RNA ’) were separated on a 1±5 % agarose gel and transferred to a membrane by standard protocols (10 µl, PCR 1) or blotted directly by the dot blot technique (1 µl, PCR 2). Products from PCR of standard RNA (2¬10"–2¬10& copies) were included in every blot performed. DNA was detected by non-radioactive hybridization as described. The amount of test-RNA initially applied to RT reactions was determined by comparison of the signals of Southern blotted products of PCR 1 of test RNA and standard RNA. Hybridization signals of dot blot products from PCR 2 were used for detection of the lowest amount of RNA (2¬10" RNA copies}RT) (Fig. 1). Quantification was done twice and was proven to be reproducible. Minus-strand RNA was not quantified but the amount of RNA was judged according to results derived from nested RT–PCR and}or Southern blot hybridization (nested RT–PCR plus hybridization negative ¯ ‘ no minus-strand RNA ’ ; amplified DNA detectable by hybridization only ¯ ‘ small amount of minus-strand RNA ’ ; amplified DNA detectable in agarose gel or with reproducibly strong hybridization signals ¯ ‘ higher amount of minus-strand RNA ’). 10) in vitro transcribed plus-strand RNA molecules and 10( native (fulllength) plus-strand RNA molecules could be reverse transcribed with primer 28tag, amplified and hybridized without detection of artefactual positive results.

Results Short-term HCV in vitro infections detected by nested RT–PCR

In order to establish an HCV in vitro propagation system, HuH7, HepG2, PK15 and STE cells were infected, initially using standard FCS-containing culture media. To modulate parameters for virus–cell interactions, several supplements were added prior to and during infection, including supplementation with PEG and}or DMSO, and also lovastatin for its effect in increasing LDL-receptor expression. Thus all four cell lines were repeatedly infected in vitro with two HCVpositive sera (1 and 2) using preincubations and supplementations as described in Table 1. After inoculation, the appearance of viral plus- and minus-strand RNA was assayed in cell lysates and culture media daily during the first 10 days. As shown in Fig. 2, detection of both plus- and minus-strand HCV RNAs in HuH7, PK15 and STE cells could be slightly enhanced by supplementation with PEG and}or DMSO. However, detection of plus- and minus-strand RNA remained sporadic in cells and supernatants and could not be reproduced in the same

CEHB

S. Seipp and others (a) 105 g.e./mio

g.e./mio

105 104

103

102

102

105

105

104

104

g.e. /mio

g.e./mio

103

104

103 102

103 102

0

I

II

III

IV V VI Generations

VII VIII

IX

X

0

I

II

III

IV V VI Generations

VII VIII

IX

X

5¬ 104 5¬ 103 5¬ 102

5¬ 104 5¬ 103 5¬ 102

g.e. /ml

5¬ 105

5¬ 105

5¬ 105

5¬ 104 5¬ 103 5¬ 102

5¬ 104 5¬ 103 5¬ 102

g.e./ml

g.e./ml

g.e. /ml

(b) 5¬ 105

0

I

II

III

IV V VI Generations

VII VIII

IX

X

0

I

II

III

IV V VI Generations

VII VIII

IX

Fig. 5. Quantification of plus-strand RNA of long-term in vitro infected cell lines. All cell lines were cultured using FCS-free Ham’s F12 medium (Figs 3–4). (a) Cell lysates. (b) Culture media (minus-strand RNA detected in cell lysates). HuH7 cells were also preincubated in PEG and/or DMSO (P, PD, D), without FCS (FCS-free), and supplemented with lovastatin (lovastatin). Generations I–VIII and I–X represent approximately days 1–60 and 1–80, respectively. Plus-strand RNA lower than

CEHC

X


way, although the enhancing effects were reproducible. In cells supplemented with lovastatin no augmenting effect could be demonstrated. Long-term HCV in vitro infections detected by nested RT–PCR

When culture of these infected cell lines was prolonged up to day 35, neither cells nor medium was positive for viral minus-strand RNA, and only very few cells showed the presence of HCV plus-strand RNA (data not shown). Thus we sought to enhance viral propagation using a specially designed culture medium. Yoo et al. (1995) introduced a highly supplemented Ham’s F12 medium for selection of HuH7 cells which were persistently transfected with HCV RNA and allowed the virus to replicate. Additionally, this medium increases LDL-receptor expression in cultivated cells because it lacks FCS (Wade et al., 1988). We therefore used this medium to enhance the efficiency of our supplementation, mainly for long-term cultivation and to study further the influence of increased numbers of LDL-receptors on HCV infection in vitro. Therefore, infections were repeated using the same four cell lines and culture parameters but the medium was changed to FCS-free Ham’s F12 for preincubations, infections and further culture. HuH7 cells were chosen to study the effect of supplementation with DMSO, PEG and lovastatin, as these cells express LDL-receptor molecules and have been shown to allow HCV replication on a low level. Cells were preincubated, infected and further cultured ; cells and medium were harvested weekly. As shown in Fig. 3, no cell line revealed significant enhancement of plus-strand RNA detection over 130 days. In this approach, STE cells were non-permissive for HCV. As found for cells infected and cultured in standard medium, minus-strand RNA was detectable during the early days of cell culture (days 7, 15 and 23) when HuH7 cells were supplemented with PEG or DMSO. However, mainly in unsupplemented hepatoma cells and in HuH7 cells which had been stimulated to increase LDL-receptor expression, replicative intermediates were sporadically detectable by nested RT–PCR after several weeks in culture (days 30–60), indicating persistent infection with low efficiency of propagation. Release of infectious particles into culture media in long-term cultures

As negative results on nested RT–PCR of in vitro infections might reflect insufficient sensitivity of our detection system, we used selected supernatants from generations IV–X of longterm studies to infect fresh cells of the same cell line. Surprisingly, we found HCV RNA sporadically by nested

RT–PCR in these secondarily infected cells (Fig. 3, ‘ sec. cells ’ days 3 and 7) indicating the presence of infectious particles in culture media undetected by nested RT–PCR. Therefore we aimed to increase the sensitivity of the detection system. Long-term in vitro infections : RNA detection by an enhanced detection system

To enhance the sensitivity of the assay, hybridization analysis was done after the nested RT–PCR in an attempt to detect plus- and minus-strand RNA of infections in Ham’s F12 medium. Products of nested RT–PCR were dot blotted (plusstrand RNA) or precipitated and Southern blotted (minusstrand RNA) and hybridized with non-radioactively labelled HCV probes, as described. It had previously been determined that any remaining primer DNA was not detected by the probe used. Using this approach, apparently HCV-negative cells and supernatants were found to be positive for both plus- and minus-strand RNA up to generation XVI, indicating persistent low level virus production (Fig. 4 a). In addition, cells and supernatants of cells infected by using culture media (secondary infections) were positive for HCV RNA. Detection of minusstrand RNA in secondarily infected cells on day 7 (generation II) showed that replication had already taken place. Typical results of minus-strand RNA detection are shown in Fig. 4 (b), revealing low amounts of minus-strand RNA in both cell lines presented. Interestingly, the apparently nonpermissive cell line STE contained replicative intermediates in every cell generation and high amounts of minus-strand RNA in supernatants up to generation 6 (day 40). These data correspond to the morphological appearance of infected STE cells, which showed a high rate of cell death and might have released a number of RNA molecules into the culture medium through cell lysis. Long-term in vitro infections : quantification of HCV RNA in infected cells

Cell lines cultured in Ham’s F12 appeared to be persistently infected with HCV, although they showed no increase in the amount of virus RNA detectable by nested RT–PCR. In order to evaluate more subtle differences between cells cultured with various supplements, we aimed to investigate the course of the in vitro infections by quantifying the HCV plus-strand RNA. Therefore, we established a semi-quantitative detection system based on a single-round PCR followed by Southern blot analysis using defined concentrations of HCV RNA as standards (see also Fig. 1). As shown in Fig. 5 (a), most cell lines revealed rapidly varying amounts of plus-strand RNA in the

102 RNA copies/106 cells and 5¬102 copies/ml medium could not be detected. g.e./mio cells, RNA genome equivalents/106 cells ; g.e./ml, RNA genome equivalents/ml culture medium ; , PCR negative for minus-strand RNA ; E, amplification products derived from minus-strand RNA, detectable by hybridization only ; E, amplification products derived from minusstrand RNA, detectable in agarose gel or with reproducible strong hybridization signals.

CEHD

S. Seipp and others

cells. However, in HuH7 cells infected under LDL-receptorstimulating conditions (FCS-free medium and lovastatin), high numbers of HCV RNA molecules were detected constantly up to generation VII (day 50). Assays for plus-strand RNA routinely revealed about 3¬10% genome equivalents (g.e.)}10' cells, although cells were subcultured 1 : 4 six times, indicating active virus replication. HCV plus-strand RNA in supernatants was quantified to estimate virus release from cells. Over time, the amount of HCV RNA in supernatants varied substantially (Fig. 5 b). Again, LDL-receptor-stimulating agents prolonged virus release to about day 40 (six generations). In cells infected with infectious culture media, the amount of plus-strand RNA reached a maximum of 10& g.e.}10' cells and 5¬10& g.e.}ml supernatant (data not shown). These data again indicate proliferation of HCV in secondarily infected cells, which had been infected with less than 10& g.e. and which had been subcultured 1 : 4 after generation I (day 3).

Discussion In this study, we systematically analysed culture conditions for in vitro HCV infection and succeeded in perpetuating HCV infection in vitro for 130 days. Ongoing replication after prolonged culturing was demonstrated as infectious HCV particles capable of infecting fresh cells were released into supernatant. There is much concern about the sensitivity and specificity of HCV RNA detection as the only marker for virus propagation, but the concentration of viral antigens in cells infected in vitro is too low to ensure sufficient sensitivity. Quantification of plus-strand RNA was performed by comparison of two-tube RT–PCR reactions rather than using a onetube competitive RT–PCR (Clementi et al., 1994 ; Ruster et al., 1995), because co-amplification of two cDNAs might alter the yield of both products (Braga & Gendler, 1994). For detection of minus-strand RNA, tagged primers (Lanford et al., 1994) or rTth-polymerase (Lanford et al., 1995) were suggested to avoid false positives by self priming events. To ensure high specificity we used small amounts of tagged primers and higher temperature during reverse transcription, although this has been shown to reduce sensitivity (V. Lohmann, personal communication). To overcome the loss of sensitivity, we decided to add Southern blot analysis of precipitated products of nested PCR in order to detect a small amount of HCV RNA. Repeating the detection tests and the use of highly stringent controls ensured specific reactions. Thus, even our observation of minus-strand RNA in some supernatants (e.g. in STE cells) does not necessarily argue for a reduced specificity ; an association of replicative intermediates with HCV envelope proteins in human sera has been described (Shindo et al., 1994). In addition, the infectivity of supernatants from cells after several passages further confirmed the presence of HCV in supernatant.

CEHE

PEG and DMSO : short-term cultures

Both PEG and DMSO were reported to enhance mechanisms necessary for membrane (e.g. virus–cell) interactions. PEG is known to be a membrane-fusing agent (Ahkong et al., 1987) and increases the efficiency of infections in a number of virus systems and cell lines (Hoekstra et al., 1989 ; Kooi et al., 1991 ; Subramanian et al., 1995) by fusing viral and cellular membranes and therefore increasing penetration rates. DMSO is used for vesicular fusion reactions (Geron & Meiri, 1985 ; Varga & Martonosi, 1992) and stabilizes cell culture systems (Isom et al., 1985), presumably by initiating formation of gap junctions (T. Yoshizawa, personal communication). Additionally, direct cell–cell contact is supposed to be a major factor for spread of pestiviruses in cell culture (T. Ru$ menapf, personal communication). Thus general mechanisms of virus– cell membrane fusion as well as cell–cell communication during initiation of infection could be the cause of enhanced virus propagation during HCV in vitro infections supplemented with PEG and}or DMSO. However, although a marginal increase in HCV RNA detection was observed within the first 10 days of cell culture if PEG and}or DMSO was added during inoculation, this effect was lost after prolonged cultivation. Therefore, rather than better access of HCV to cells, or increased infectivity, nonspecific attachment of virions to cells without complete infection may be the cause of the higher frequency of viral RNA detection during the early phase of infection. Increase of LDL receptors : long-term cultures

In FCS-free long-term cultures assayed with the enhanced detection system, persistence of HCV RNA could be detected up to day 130. Furthermore, in contrast to studies using PEG and DMSO, the additional stimulation of LDL-receptor expression by lovastatin or using FCS-free medium in HuH7 cells prior to and during infection (Mulder et al., 1991 ; Sviridov et al., 1990) resulted in detection of high amounts of HCV plusstrand RNA for 50 days. Using the same culture conditions without initial stimulation of LDL-receptor expression, this effect was not seen, raising the question whether HCV gains access to cells via the LDL-receptor molecule. There are several other lines of evidence supporting this hypothesis. Firstly, HCV particles in most sera were found to be associated with LDL or VLDL fractions (Miyamoto et al., 1992 ; Thomssen et al., 1993 ; Kanto et al., 1994). Secondly, the LDL-receptor or receptor-related surface molecules are potential entry sites for other virions into cells, such as Rous sarcoma virus (Bates et al., 1993). The apparent inconsistency of our data, showing no HCV minus-strand RNA in lovastatinsupplemented HuH7 cells but nonetheless persistent infection and infectivity of the supernatant, might be explained by the facilitated access of virions to cells without a marked increase in the number of virions present in cells or culture media. If this is correct, continued stimulation of LDL-receptor expression


by culturing infected cells in FCS-free medium would prevent abortion of infection due to a lack of entry sites. In conclusion, our data support the hypothesis of an LDLreceptor-mediated uptake of HCV into cells, while application of non-specific agents such as PEG or DMSO during inoculation did not enhance efficiency of infection in vitro. It remains to be investigated whether constant supplementation during culture of infected cells enhances replication rates. Although long-term cultures have been shown to be persistently infected with HCV for several months, the amount of viral RNA in primarily infected cells remained at low levels of maximally 1 copy}40 cells. However, the HCV RNA load in cells infected with infectious supernatants may be sufficient for further experiments on antiviral strategies. Further experiments are needed to show whether the established HCV infections resemble the in vivo situation or whether adaptation processes have already started in the in vitro cultures. This work was supported in part by grants from the Bundesministerium fu$ r Bildung, Technologie, Forschung und Wissenschaft and the Forschungsschwerpunkt Transplantation Heidelberg. The authors thank Mrs Susanne Selzer for excellent technical assistance and Dr Reinhold Ahl (Federal Research Center for Virus Diseases of Animals, Tu$ bingen, Germany) for kindly providing the STE cell line. We are indebted to Dr J. Kreutzer, Department of Cardiology, University of Heidelberg and Merck, Sharp & Dohme GmbH, Germany for provision of lovastatin.

References Ahkong, Q. F., Desmazes, J. P., Georgescauld, D. & Lucy, J. A. (1987).

P. (1995). Detection by in situ hybridization of hepatitis C virus positive

and negative RNA strands using digoxigenin-labeled cRNA probes in human liver cells. Journal of Hepatology 23, 509–518. Geron, N. & Meiri, H. (1985). The fusogenic substance dimethyl sulfoxide enhances exocytosis in motor nerve endings. Biochimica et Biophysica Acta 819, 258–262. Gripon, P., Diot, C., Theze, N., Fourel, I., Loreal, O., Brechot, C. & Guguen-Guillouzo, C. (1988). Hepatitis B virus infection of adult human

hepatocytes cultured in the presence of dimethyl sulfoxide. Journal of Virology 62, 4136–4143. Gripon, P., Diot, C. & Guguen-Guillouzo, C. (1993). Reproducible high level infection of cultured adult human hepatocytes by hepatitis B virus : Effect of polyethylene glycol on adsorption and penetration. Virology 192, 534–540. Havekes, L., Van Hinsbergh, V., Kempen, H. J. & Emeis, J. (1983). The metabolism in vitro of human low-density lipoprotein by the human hepatoma cell line HepG2. Biochemical Journal 214, 951–958. Hoekstra, D., Klappe, K., Hoff, H. & Nir, S. (1989). Mechanism of fusion of Sendai virus : role of hydrophobic interactions and mobility constraints of viral membrane proteins. Effects of polyethylene glycol. Journal of Biological Chemistry 264, 6786–6792. Houghton, M., Weiner, A., Han, J., Kuo, G. & Choo, Q. L. (1991).

Molecular biology of the hepatitis C viruses : implications for diagnosis, development and control of viral disease. Hepatology 14, 381–388. Iacovacci, S., Sargiacomo, M., Parolini, I., Ponzetto, A., Peschle, C. & Carloni, G. (1993). Replication and multiplication of hepatitis C virus

genome in human foetal liver cells. Research in Virology 144, 275–279. Isom, I., Georgoff, I., Salditt-Goergieff, M. & Darnell, J. E., Jr (1985).

Persistence of liver-specific messenger RNA in cultured hepatocytes : different regulatory events for different genes. Journal of Cell Biology 105, 2877–2885. Ito, T., Mukaigawa, J., Hirabayashi, Y., Mitamura, K. & Yasui, K. (1996). Cultivation of hepatitis C virus in primary hepatocyte culture

Movements of fluorescent probes in the mechanisms of cell fusion induced by polyethylene glycol. Journal of Cell Science 88, 389–398.

from patients with chronic hepatitis C results in release of high titer infectious virus. Journal of General Virology 77, 1043–1054.

Bartenschlager, R., Lohmann, V., Wilkinson, T. & Koch, J. O. (1995).

Kanto, T., Hayashi, N., Takehara, T., Hagiwara, H., Mita, E., Naito, M., Kasahara, A., Fusamoto, H. & Kamada, T. (1994). Buoyant density of

Complex formation between the NS3 serine-type proteinase of the hepatitis C virus and NS4A and its importance for polyprotein maturation. Journal of Virology 69, 7519–7528. Bates, P., Young, J. A. T. & Varmus, H. E. (1993). A receptor for subgroup A Rous sarcoma virus is related to the low density lipoprotein receptor. Cell 74, 1043–1051. Bertolini, L., Iacovacci, S., Ponzetto, A., Gorini, G., Battaglia, M. & Carloni, G. (1993). The human bone-marrow-derived B-cell line CE,

susceptible to hepatitis C virus infection. Research in Virology 144, 281–285. Braga, V. M. M. & Gendler, S. J. (1994). Co-amplification of two cDNAs in RT–PCR can alter the yield of both products. BioTechniques 17, 228–229. Chomczynski, P. N. & Sacchi, N. (1987). Single step method of RNA isolation by acid guanidinium thiocyanate–phenol–chloroform extraction. Analytical Biochemistry 162, 156–159.

hepatitis C virus recovered from infected hosts : two different features in sucrose equilibrium density-gradient centrifugation related to degree of liver inflammation. Hepatology 19, 296–302. Kato, N., Nakazawa, T., Mizutani, T. & Shimotohno, K. (1995).

Susceptibility of human T-lymphotropic virus type I infected cell line MT-2 to hepatitis C virus infection. Biochemical and Biophysical Research Communications 206, 863–869. Kooi, C., Cefvin, M. & Anderson, R. (1991). Differentiation of acid-pHdependent and -nondependent entry pathways for mouse hepatitis virus. Virology 180, 108–119. Kostner, G. M. (1993). Interaction of Lp(a) and of Apo(a) with liver cells. Arteriosclerosis and Thrombosis 13, 1101–1109. Kuo, G., Choo, Q. L., Alter, H. J., Gitnick, G. L., Redeker, A. G., Purcell, R. H., Miyamura, T., Dienstag, J. L., Alter, M. J., Stevens, C. E. and others (1989). An assay for circulating antibodies to a major etiologic

Clementi, M., Bagnarelli, P., Manzin, A. & Menzo, S. (1994).

virus of human non-A, non-B hepatitis. Science 244, 362–364.

Competitive polymerase chain reaction and analysis of viral activity at the molecular level. Genetic Analysis Techniques and Applications 11, 1–6. Cribier, B., Schmitt, C., Bingen, A., Kirn, A. & Keller, F. (1995). In vitro infection of peripheral blood mononuclear cells by hepatitis C virus. Journal of General Virology 76, 2485–2491.

Lanford, R. E., Sureau, C., Jacob, J. R., White, R. & Fuerst, T. R. (1994).

Gastaldi, M., Massacrier, A., Planells, R., Robaglia-Schlupp, A., PortalBartolomei, I., Bourliere, M., Quilici, F., Fiteni, J., Mazzella, E. & Cau,

Demonstration of in vitro infection of chimpanzee hepatocytes with hepatitis C virus using strand-specific RT}PCR. Virology 202, 606–614. Lanford, R. E., Chavez, D., Chisari, F. V. & Sureau, C. (1995). Lack of detection of negative-strand hepatitis C virus RNA in peripheral blood mononuclear cells and other extrahepatic tissues by the highly strandspecific rTth reverse transcriptase PCR. Journal of Virology 69, 8079–8083.

CEHF

S. Seipp and others Lunel, F., Mariotti, M., Cresta, P., de la Croix, I., Huraux, J. M. & Lefrere, J. J. (1995). Comparative study of conventional and novel

Shimizu, Y. K., Feinstone, S. M., Kohara, M., Purchell, R. H. & Yoshikura, H. (1996). Hepatitis C virus : detection of intracellular virus

strategies for the detection of hepatitis C virus RNA in serum : amplicor, branched-DNA, NASBA and in-house PCR. Journal of Virology Methods 54, 159–171. Mihm, S., Hartmann H. & Ramadori, G. (1996). A reevaluation of the association of hepatitis C virus replicative intermediates with peripheral blood cells including granulocytes by a tagged reverse transcription} polymerase chain reaction technique. Journal of Hepatology 24, 491–497.

particles by electron microscopy. Hepatology 23, 205–209.

Mizutani, T., Kato, N., Saito, S., Ikeda, M., Sugiyama, K. & Shimotohno, K. (1996). Characterization of hepatitis C virus replication

Defective entry of herpes simplex virus types 1 and 2 into porcine cells and lack of infection in infant pigs indicate species tropism. Journal of General Virology 76, 2375–2379.

in cloned cells obtained from a human T-cell leukemia virus type 1infected cell line, MT2. Journal of Virology 70, 7219–7223. Miyamoto, H., Okamoto, H., Sato, K., Tanaka, T. & Mishiro, S. (1992).

Shindo, M., Di Bisceglie, A. M., Akatsuka, T., Fong, T. L., Scaglione, L., Donets, M., Hoofnagle, J. H. & Feinstone, S. M. (1994). The physical

state of the negative strand of hepatitis C virus RNA in serum of patients with chronic hepatitis C. Proceedings of the National Academy of Sciences, USA 91, 8719–8723. Subramanian, G., LeBlanc, R. A., Wardley, R. C. & Fuller, A. O. (1995).

Sviridov, D. D., Safonova, I. G., Pavlov, M. Y., Kosykh, V. A., Podrez, E. A., Antonov, A. S., Fuki, I. V. & Repin, V. S. (1990). Inhibition of

Extraordinarily low density of hepatitis C virus estimated by sucrose density gradient centrifugation and the polymerase chain reaction. Journal of General Virology 73, 715–718. Mulder, M., de Wit, E. & Havekes, L. M. (1991). The binding of human lipoprotein lipase treated Vldl by the human hepatoma cell line HepG2. Biochimica et Biophysica Acta 1081, 308–314.

cholesterol synthesis by lovastatin tested on six human cell types in vitro. Lipids 25, 177–179.

Mueller, H. M., Pfaff, E., Goeser, T., Kallinowski, B., Solbach, C. & Theilmann, L. (1993). Peripheral blood leukocytes serve as a possible

hepatitis C virus in human sera due to the binding of beta-lipoproteins and immunoglobulins. Medical Microbiology and Immunology 182, 329–334. Varga, S. & Martonosi, A. (1992). Giant sarcoplasmic reticulum vesicles : a study of membrane morphogenesis. Journal of Muscle Research and Cell Motility 13, 497–510. Wade, D. P., Knight, B. L. & Soutar, A. K. (1988). Hormonal regulation of low-density lipoprotein (LDL) receptor activity in human HepG2 cells. Insulin increases LDL receptor activity and diminishes its suppression by exogenous LDL. European Journal of Biochemistry 174, 213–218. Wang, C., Sarnow, P. & Siddiqui, A. (1994). A conserved helical element is essential for internal initiation of translation of hepatitis C virus RNA. Journal of Virology 68, 7301–7307.

extrahepatic site for hepatitis C virus replication. Journal of General Virology 74, 669–676. Nakajima, N., Hijikata, M., Yoshikura, H. & Shimizu, Y. K. (1996).

Characterization of long-term cultures of hepatitis C virus. Journal of Virology 70, 3325–3329. Nissen, E., Hohne, M. & Schreier, E. (1994). In vitro replication of hepatitis C virus in a human lymphoid cell line (H9) [letter]. Journal of Hepatology 20, 437. Nouri-Aria, K. T., Sallie, R., Sangar, D., Alexander, G. J., Smith, H., Byrne, J., Portmann, B., Eddleston, A. L. & Williams, R. (1993).

Detection of genomic and intermediate replicative strands of hepatitis C virus in liver tissue by in situ hybridization. Journal of Clinical Investigation 91, 2226–2234. Roth, W. K., Lee, J. H., Ru$ ster, B. & Zeuzem, S. (1996). Comparison of two quantitative hepatitis C virus reverse transcription PCR assays. Journal of Clinical Microbiology 34, 261–264. Ruster, B., Zeuzem, S. & Roth, W. K. (1995). Quantification of hepatitis C virus RNA by competitive reverse transcription and polymerase chain reaction using a modified hepatitis C virus RNA transcript. Analytical Biochemistry 224, 597–600. Shimizu, Y. K. & Yoshikura, H. (1994). Multicycle infection of hepatitis C virus in cell culture and inhibition by alpha and beta interferons. Journal of Virology 68, 8406–8408.

CEHG

Thomssen, R., Bonk, S., Propfe, C., Heermann, K. H., Kochel, H. G. & Uy, A. (1992). Association of hepatitis C virus in human sera with beta-

lipoprotein. Medical Microbiology and Immunology 181, 293–300. Thomssen, R., Bonk, S. & Thiele, A. (1993). Density heterogeneities of

Yoo, B. J., Selby, M. J., Choe, J., Suh, B. S., Choi, S. H., Joh, J. S., Nuovo, G. J., Lee, H. S., Houghton, M. & Han, J. H. (1995). Transfection

of a differentiated human hepatoma cell line (Huh7) with in vitrotranscribed hepatitis C virus (HCV) RNA and establishment of a longterm culture persistently infected with HCV. Journal of Virology 69, 32–38. Zignego, A. L., Macchia, D., Monti, M., Thiers, V., Mazzetti, M., Foschi, M., Maggi, E., Romagnani, S., Gentilini, P. & Brechot, C. (1992).

Infection of peripheral mononuclear blood cells by hepatitis C virus. Journal of Hepatology 15, 382–386. Received 28 February 1997 ; Accepted 4 June 1997