Background
HCV is associated with a number of diseases, including hepatocellular carcinoma, B-cell lymphomas, and neuropathy. There is an emerging list of diseases that may have some association with this virus. Approximately 8% of HCV-infected individuals in the United States are infected with genotype 3 [
1]. The chances of liver damage due to HCV infections may not vary by genotype in untreated individuals [
2,
3], and infections with HCV-3 are more likely to respond earlier to ribavirin/α-interferon combination therapy than HCV-1 [
3‐
5]. There is evidence that individuals infected with HCV-3 are likely to progress rapidly to liver steatosis [
6], and fibrosis [
7] compared to infection with HCV-1. Individuals infected with HCV are also frequently infected with other viruses. Hematopoietic cells e.g., HCV infected T-cells, are capable of being co-infected with HIV-1 and HHV-6 [
8]. All of the co-infecting viruses continue to replicate in these cells.
Although synthetic constructs are commonly used for HCV related studies, we have a system for studying the natural virus isolated from infected patients. Reports using constructs viz., Replicon, pseudo-particles etc. may have produced interesting data, but these results lack meaning in the area of human diseases and public health. Meaningful data must come from viruses isolated from patients since no Replicon disease yet exists.
The 5'UTR of HCV controls replication through cap-independent translation [
9‐
11]. For this report, HCV was isolated from the blood of patients infected with HCV-3 and transmitted into macrophages, B-cells, and T-cells. The 5'UTR of the progeny viruses was analyzed and compared to the sequences of the HCV RNA found in these patients' sera. In our previous reports, the 5'UTR of HCV-1 isolates from patients was analyzed and compared to HCV cultured
in vitro and minor differences were found between the HCV in the isolates and patients' sera [
12,
13]. This suggested that the virus in culture was similar to that found in patients' blood.
Discussion
There are numerous reports about differences between different strains or types of HCV. We are reporting the isolation and replication of HCV from patients infected by type 3 strains of HCV. These new isolates can be cultured in both B and T cells. By contrast to type 1 strains of HCV, sequence comparisons of the 5'UTR of HCV found in patients' sera and their corresponding in vitro isolates suggests significant changes in the sequences of type 3 strains. The replication of other HCV genotypes such as 2, 4, 5, and 6 in our system is awaiting studies.
As we have published before, macrophages are important in selecting HCV-1 for replication [
12,
16]. Since these cells are focus of our studies, we would like to name them as cells that are functionally highly phagocytic and cytochemically stain intensely for non-specific esterase. This would include both fixed and free cells such as histiocytes in connective tissue, Kupffer cells of liver, microglial cells of neuronal tissue, dendritic cells of skin, and alveolar macrophages to name a few. The presence of HCV in monocytes or macrophages has been shown in HCV-infected individuals [
17‐
21]. In addition, like HIV, infected macrophages may act as a reservoir of biologically active, infectious HCV
in vivo. Although, we have had some success in isolating HCV-3, the system is not optimal. Other types of macrophages, e.g. Kupffer cells, are probably better for replication of genotype 3. Results presented here show that we were able to isolate and culture HCV from patients infected with HCV-3 to a limited extent. However, the HCV-3 produced by macrophages, B-cells, and T-cells were significantly different from HCV-3 in the patients' sera (Figures
3 and
4).
We were unable to find one set of PCR primers and conditions that worked for all of our samples. For genotype 1, we routinely use the same set of conditions for the analysis of the 5'UTR. For genotype 3, we found that some samples would not work for any particular set of PCR conditions. This is presumably due to a high degree of variability of HCV-3.
Our studies indicate that the macrophages preferentially select HCV-1, making them the dominant virus type (Figure
4), and HCV-3 may poorly infect macrophages from cord blood. The reduced sequence complexity of HCV-3 cultured in B-cells and T-cells suggests that macrophages are selecting against this genotype. Sera from patients 314 and 384 had HCV-3 sequences, while the HCV in macrophages and other cell types was only HCV-1.
The 388 T1 sample (B-cells) had 8 C's starting at position 120, compared to 7 for the 388 serum and 388 primary (macrophage) samples. Although we only sequenced four clones for this sample, every one of them had an extra C. We have previously observed an additional C in several HCV-1 samples [
13]. In addition, we have found a large deletion in this area for one sample [
16]. This region is located between stem-loops II and III, thus apparently allowing greater variability. HCV needs to be infectious, and the level of replication of these infectious agents will depend upon a number of factors, most importantly the target cells.
The results presented here suggest that HCV-3 may need a different cell type for its primary replication
in vitro. Our previous publications document the selection of HCV-1 in macrophages or similar cells viz., neuronal precursors. Individuals infected with genotype 3 may have small amounts of other genotypes circulating in their blood. It is possible that these other genotypes may also prefer to infect specific cell types for replication
in vitro. Others have shown that different tissues in one particular individual may harbor different genotypes of HCV [
22‐
24], suggesting that cell tropism may establish the tissue specificity of HCV in infected individuals. Variability in diseases of HCV-infected individuals, such as neuropathy and lymphoma, may either be due to variations in the virus or to increased susceptibility of infected cell types, or the presence of other viral agents in circulation. This phenomenon is under further investigation.
Methods
Patient samples
All patient samples were given a code at the source, and a sequential number in our laboratory to preserve their anonymity. Patients 314, 384, and 388 were all AIDS patients doubly infected with HIV and HCV. The HCV was genotyped as type 3 using an INNO-LiPA assay by a clinical testing laboratory (Quest Diagnostics).
In vitro culture system
Our culture system, described earlier, takes advantage of the infectious particles present in the peripheral circulation [
12]. Briefly, the isolation of HCV was done in two stages: (A) HCV derived from patients' blood was used to infect human macrophages; (B) HCV obtained from the macrophages was then used to infect freshly transformed B-cells or T-cells obtained from human fetal cord blood and cultured in the presence of 100 units/ml of IL-2 (Collaborative Biomedical Products, catalog number 40121). The types of samples that were analyzed included: (i) HCV found in serum or plasma of patients; (ii) HCV produced by macrophages (primary isolate); and (iii) HCV produced by B-cells or T-cells (secondary isolates). Each sample was given a unique number that indicated the patient and a suffix designating replication into various cell types. Transfers into fresh uninfected B-cells were given a suffix of T1, transfers into cell lines, such as P3HR1 and CEM, were given a suffix of the cell line name, and secondary transfers into macrophages produced from human cord blood (CB). All isolates were produced in the laboratories at CIMM, and are therefore called CIMM-HCV. The data in this paper is based on these isolates.
CEM.NCI and CEM.SS cells were obtained through the NIH AIDS Research and Reference Reagent Program, Division of AIDS, NIAID, NIH. P3HR1 cells were obtained from ATCC. The cell types were cultured using the methods recommended by ATCC. Single cell clones were established from these cell lines for the sake of uniformity of data and the removal of adventitious material such as mycoplasma.
RT-PCR
RNA was purified using TriReagent as previously described [
12], and a nested RT-PCR was performed. For the patient 314 samples, the procedure was as previously described using Fidelitaq (US Biochemicals) [
12]. For the other patient samples, RT was performed using N
12 random primers for 12 12-minute cycles at 48°C using cyclic RT (Bioneer). The PCR was performed using Bioneer high fidelity TLA PCR premixes. The primers used for the experiments are listed in Table
1.
For each TLA PCR reaction, samples were denatured at 94°C for 3 minutes, and then 30 cycles of amplification were performed with the following temperature profiles: 94°C for 30 seconds, 50 or 55°C for 30 seconds, and 72°C for 1 min for the outer primer set and 94°C for 30 seconds, 55 or 60°C for 30 seconds, and 72°C for 1 min for the inner primer set (Table
2). Normally, 2 ml of RNA was used for the RT, 2 ml of the cDNA used for the first PCR, and 2 ml of the first PCR product for the second PCR. Volumes were adjusted as needed.
Sequencing
Fragments comprising approximately 263 or 276 bp of the 5'UTR generated by nested PCR were cloned using Invitrogen's ZeroBlunt cloning kit. Plasmid DNA from a minimum of 4 clones of each sample were amplified by Templiphi (Amersham) and then sequenced using a Beckman CEQ8000 Genetic Analysis system. In order to ensure high quality analyses, only clones that had identical sequences for both strands were analyzed. All methods followed the manufacturers' protocols.
Analysis of the sequences was performed as described previously [
13]. The numbers for the base positions that are reported here are the bases compared to the positions of the full length genome of HCV H77 [
25,
26].
Complexity of the variation was calculated as Shannon entropy and Pn complexity as described previously [
13]. Only samples with at least 17 sequences were analyzed for variation.
Accession numbers of HCV sequences used for genotyping
The 5' UTR sequences reported in the paper have the GenBank accession numbers HM641722 to HM641732.
Competing interests
All intellectual rights are reserved by the California Institute of Molecular Medicine (CIMM). There are no competing interests between California Lutheran University or any other body and CIMM.
Authors' contributions
SZS performed the biological work. JGP and ASK performed the clinical work, recruitment of patients, and procurement of specimens. MOA designed experiments and performed molecular work. SZS and DR designed and conducted experiments, analyzed the data, and wrote the manuscript. All of the authors have read and approved the final manuscript.