CKR-L3, a deletion version CCR6-isoform shows coreceptor-activity for limited human and simian immunodeficiency viruses

Cells

The human glioma cell line NP-2 expresses neither CD4 nor coreceptors, thus serving as an ideal host for the study of coreceptor activity [16]. Eagle’s minimum essential medium (EMEM) (NISSUI Co., Inc. Tokyo, Japan) supplemented with 10% fetal bovine serum (FBS) was used to maintain NP-2 cells and its derivatives. A panel of established human cell lines was examined for the expression of mRNA of CKR-L3- and CCR6-ORFs. T-cell lines, ATL-1 K [17], C8166, Jurkat [18] and a B-cell line, Daudi [19] were cultured in RPMI 1640 medium (NISSUI Co., Inc.) containing 10% FBS. Hepatoma cell lines, huH1 [20] and HepG2 [21] were cultured in Dulbecco’s modified EMEM (DMEM) (NISSUI Co., Inc.) containing 10% FBS. Human peripheral blood mononuclear cells (PBMC) were isolated from the blood of a healthy donor using Ficoll-Paque density gradient centrifugation (Pharmacia, Uppsala, Sweden), then stimulated with phytohemagglutinin (PHA) and cultured in RPMI 1640 medium supplemented with 20% FBS and 100 IU/ml of recombinant interleukin-2 (IL-2). A human T-cell line, C8166 was transduced with CCR5 to prepare C8166/CCR5 cells, which were used for the preparation of viral stocks of HIV and SIV strains [16]. The amphotropic packaging cell line Phoenix-Ampho [22] was maintained in DMEM containing 10% FBS.

Viruses

A panel of laboratory-adapted HIV-1 strains, GUN-1WT, GUN-1V, GUN-4WT, GUN-4V, IIIB, BaL; and HIV-2 strains, CBL23, ROD, SBL6669, and SIV strains, mac251 and mndGB-1, were used to examine coreceptor activities [13]. Primary isolates of HIV and SIV were obtained from the National Institute for Biological Standard and Control (NIBSC, London, UK). The characteristics of these isolates with NIBSC-reference codes are as follows: HIV-1 strains, MVP-5180 (Cameroon, subtype O, EVA167), 93BR020 (Brazil, subtype F, ARP179.25), 92US723 (USA, subtype B, ARP1039.3), HAN2 (Germany, subtype B, EVA158). An HIV-2 strain, MIR (Guinia Bissau, EVA171), and an SIV primary isolate, smE660 (ARP1040), originated from a sooty mangabey, were also included. A primary isolate GUN11 was isolated from Gunma University Hospital and used in this study.

Amplification of CKR-L3 open reading frame

The coding region DNA sequence of CKR-L3 was obtained from the GenBank database (Accession number, Z79784.1). Oligonucleotide primers were designed and synthesized (Proligo K.K., Tokyo, Japan) covering its ORF and to detect reverse-transcribed mRNA by PCR. Primers sequences, orientations and positions in the respective ORF were as follows: CKR-L3-F, 5′ ATGAATTTCAGCGATGTTTTCGACTCCAG 3′(sense: from the 1^st to the 29^th) and CKR-L3-R, 5′ TCACATAGTGAAGGACGACGCATTGTCG 3′(antisense: 1083^rd-1110^th). The native CCR6-ORF (GenBank accession number, U60000.1) was amplified accordingly. As controls, the mRNA expression of CD4 (M12807.1), CCR5 (U54994.1), CXCR4 (AY242129.1) and gleceraldehyde-3-phospahte dehydrogenase (GAPDH, M17851.1) were also examined. Candidate ORFs amplified by RT-PCR were detected by electrophoresis through 1% (w/v) agarose gel.

Cloning of CKR-L3 as coreceptor candidates

Amplified CKR-L3-ORF was cloned into a TA-cloning plasmid, pGEM-T Easy (Promega, Madison, WI), and the derivative plasmid was designated as pGEM-T Easy/CKR-L3. Sequencing of cloned DNA fragment was performed by a 5500-sequencer (Hitachi, Tokyo, Japan) using fluorescent primers labeled with Texas Red. Restriction-digested CKR-L3 ORF was re-cloned into an expression plasmid, pMX-puro [23] and transfected into NP-2/CD4 cells to produce and establish NP-2/CD4/CKR-L3 indicator cell line as described earlier [16]. Similarly, NP-2/CD4/CCR6 cell line was used as running control, NP-2/CD4/CCR5 and NP-2/CD4/CXCR4 as positive controls and NP-2/CD4 was used as a negative-control.

HIV and SIV infection assay

Receptor and coreceptor transduced NP-2/CD4/CKR-L3, NP-2/CD4/CCR6, NP-2/CD4/CCR5, NP-2/CD4/CXCR4 and NP-2/CD4 cells were seeded into 24-well microtiter plates in a density of 50,000 cells/mL. After overnight, medium was drained off with the addition of viral inoculums to the seeded-cells. Following a six hour incubation, the cells were washed three times with EMEM containing 10% FBS to remove free virus, then refilled with 500 μL fresh medium per well and cultured at 37°C in a 5% CO₂ incubator. In 3–5 day intervals, the cell monolayers were disrupted with trypsin-EDTA (0.25% trypsin, 1 mM EDTA), cell viability was checked by 0.4% trypan blue staining solution, counted via hemacytometer and re-seeded equal number of cells to maintain uniform cell growth rate and maintained for up to 8 weeks. The presence of viral antigens in the infected cells was detected by indirect immunofluorescence assay (IFA) [24]. Retroviral infection was further assured by detection of proviral DNA by PCR using the genomic DNA of infected cells as templates [25]. Infection induced cell fusion was observed under light microscopy. Formation of syncytia was detected by Giemsa staining (Muto Pure Chemicals, Tokyo, Japan). Subsequently, virus releases through different coreceptors were quantified with measurement of reverse transcriptase (RT) activities in used culture supernatants [17]. Briefly, culture supernatants of virally infected cells were precipitated with 30% polyethylene glycol and lysed with standard disruption buffer containing 0.026% triton X-100. Lysates were mixed with 50 μL of assay reaction mixture consisting of 40 mM Tris HCl (pH 7.8), 4 mM dithiothreitol, 45 mM KCl, poly(rA), oligo(dT) (Pharmacia Biotech, NJ), 15 μM [³H] (5 Ci/mmol) (Amersham International, UK), and 10 mM MgCl₂. The mixtures were incubated for one hour at 37°C. Incorporation of radioactivity into cold trichloroacetic acid–insoluble fractions was assayed in a liquid scintillation counter (Beckman Coulter LS 6500, CA). Ionization rate in counts per minute (cpm) of culture supernatants was adjusted by subtracting the background cpm. The assay was carried out in duplicates.

Sequencing of viral DNA

The primer pair for proviral DNA of HIV-2MIR was HIV2-F (forward 5′-ATGTGATAAGCACTATTGGGATGATA-3′) and HIV2-R (reverse 5′-CATGCTTGTTTAGGTCTTGTATTGAT-3′), located between positions of 7084-7109-nt and 7433-7458-nt, respectively, covering the C2-V3 env region (nucleotide positions of HIV-2CRIK, DQ307022.1). For SIVsmE660, a proviral DNA fragment of 773 nucleotides spanning the V1-V3 regions of env gene was amplified as described earlier [26]. Upon detection, the DNA bands were cloned into a TA-cloning plasmid, pGEM-T Easy, and inserts were sequenced and blasted in the NCBI nucleotide database (http://blast.ncbi.nlm.nih.gov) to confirm the DNA integrity of the inoculated viruses. Sequence alignments have been done using the BioEdit program (version 7.2) for windows.

Statistical analysis

Mean percentages of the antigen-positive cells in two independent infection assay and RT values in duplicate samples were analyzed by paired t test using IBM SPSS statistics data editor (version 17).

Nucleotide sequence accession numbers

The viral nucleotide sequences used in this study were submitted to GenBank and were assigned accession numbers: JN107567-69 for HIV-2MIR-CCR5, HIV-2MIR-CKR-L3 and HIV-2MIR-CCR6, and the number JN107564-66 for SIVsmE660-CCR5, SIVsmE660-CKR-L3, SIVsmE660-CCR6 clones, respectively.

Ethics statement

This study was approved by the ethical review committee of the Faculty of Medicine, Gunma University, Japan, and standard written consent was obtained from the enrolled healthy blood donor.

Detection of CKR-L3 and CCR6 expressions

The CKR-L3 ORF is similar to that of CCR6 except for the absence of first five-amino acids in the NTR region. The schematic diagram in Figure 1a indicates the point of deletion at native CCR6 to represent CKR-L3. Expressions of mRNA of both CKR-L3 and CCR6 were detected by RT-PCR in PBMC, T cell lines, ATL1K and C8166 as well as in hepatic cell lines, such as huH1 and HepG2 (Figure 1b). Similar to our findings, previous reports have showed the expression of CKR-L3 in spleen, lymph nodes and PBMC [15], whereas CCR6 was selectively expressed on Th17 cells and regulatory T cells, memory T cells, B cells and dendritic cells [27, 28]. The 15-nucleotide size-difference between the amplified ORFs of CKR-L3 and CCR6 could not be distinguished fairly when they were run through 1.5% agarose gel electrophoresis (Figure 1c). The major coreceptor, CCR5 was found to be expressed on ATL-1 K and PBMC. The expression of CD4 was detected on both the T cell lines and on PBMC.

Establishment of NP-2/CD4/coreceptor cells

Amplified ORFs of CKR-L3, CCR6 and CCR5 were transduced one by one into NP-2/CD4 cells to produce NP-2/CD4/coreceptor cells. The expression of the coreceptor candidates were affirmed by detecting their specific mRNAs in the transfected cell lines by RT-PCR. We ensured the expressions of respective candidates in NP-2/CD4/coreceptor sub-lines by flow cytometry (FCM) following previously described methods [29]. The mother cell, NP-2/CD4 did not express any of the coreceptor candidates.

Coreceptor functions of CKR-L3 and CCR6

The coreceptor activities of CKR-L3 and CCR6 for HIV-2MIR and SIVsmE660 primary isolates were noticed when viral antigen-positive NP-2/CD4/CKR-L3 and NP-2/CD4/CCR6 cells were detected by IFA (Figure 2a). We have described the coreceptor-role of CCR6 in detail previously [13]. The present report highlights the superior coreceptor-role of CKR-L3 relative to its native counterpart, CCR6. HIV-2MIR showed a good pace of replication in NP-2/CD4/CKR-L3 cells and more than 80% cells became antigen positive within 16 days after inoculation (Figure 2b). To reach the same level of infection, CCR6-transfected cells took one week more. In contrast, HIV-2MIR attained the same level of infection in three to four days in CCR5-transfected cells (Figure 2b). SIVsmE660 isolate was also highly infective via the CCR5 coreceptor: more than 80% NP-2/CD4/CCR5 cells were antigen-positive within one week after inoculation. This virus was slow but progressive through the CKR-L3 coreceptor and took six weeks to infect 80% of its cells. In contrast, CCR6 showed much weaker coreceptor activity than CKR-L3 taking 10 days more to produce similar levels of infection (Figure 2c). However, infections via another major coreceptor, CXCR4, is not shown, but described in Table 1. A second independent experiment had been done that revealed similar outcomes.

Productive HIV and SIV infection in susceptible cells in vitro or in vivo is often associated with the formation of syncytia, which are multinucleated giant cells (MGC) [30, 31]. MGC in infected NP-2/CD4/CKR-L3 and NP-2/CD4/CCR6 cells were quantified after fixation and staining with Giemsa. Both HIV-2MIR and SIVsmE660 induced syncytia formation using CKR-L3 and CCR6 as coreceptors (Figure 3a). However, syncytia formation through the CKR-L3 coreceptor occurred more rapidly than with the CCR6 coreceptor. Clusters were variable in sizes and consisting of 10–25 nuclei per syncytium. Infection through CCR5 frequently generated viral-induced MGC, while coreceptor-negative NP-2/CD4 cells did not form any cluster of nuclei (data not shown). Typically when cells become infected by HIV and SIV, viral DNA provirus forms by reverse transcription of genomic RNA into double stranded DNA followed by subsequent integration into host DNA [32]. We further demonstrated infection through the CKR-L3 and CCR6 coreceptors through the detection of proviral DNA in infected cell lines (Figure 3b). Retroviral cytopathic effects (CPE) in cell-line-systems release a large amount of virus in to cell-supernatant. The amount of produced viruses was assessed by RT activities in spent culture media of viral inoculated cells [33]. We observed sharp increases of RT activities associated with CPE and cell ruptures. For HIV-2MIR infection, RT release was measured at different time points with a time dependent gradual increase of antigen-positive cells as detected by IFA (Figure 3c). The RT activities at the final time point of infection cycles were 1.7 × 10⁵, 1.6 × 10⁵ and 1.2 × 10⁵ cpm/mL using CCR5, CKR-L3 and CCR6 coreceptors respectively when 80-90% cells become antigen positive by IFA. All the three cell lines generated four to six fold increases of RT values during the last three days of respective infections (Figure 3c). However, the CKR-L3-transfected cell line took a shorter time to release peak RT values compared to that of CCR6. Contrast to HIV-2MIR, RT activities for SIVsmE660 infections were measured only at the end points of each infection cycles. Culture fluids of infected NP-2/CD4/CCR5 cells on day-seven, NP-2/CD4/CKR-L3 cells on day-42 and NP-2/CD4/CCR6 cells on day-52 were assayed for RT activities when CPE ascended at the level of 80-90% cell infection (Figure 2c). The RT releases using CCR5, CKR-L3 and CCR6 coreceptors for SIVsmE660 were 2.3 × 10⁶, 1.7 × 10⁶ and 1.5 × 10⁶ cpm/mL, respectively. Very negligible RT activities (0.8-1.5 × 10³) were detected when the supernatants of virus-infected NP-2/CD4 cell-cultures were measured.

Sequence analyses of the proviral DNAs

The sequences of HIV-2MIR and SIVsmE660 grown in NP-2/CD4/CCR5, NP-2/CD4/CKR-L3 and NP-2/CD4/CCR6 cells were validated by blasting with NCBI and GenBank nucleotide databases. To analyze the genetic divergence while propagating through different coreceptors, we aligned CCR5-clone to those of CKR-L3- and CCR6-clones. For HIV-2MIR, the amplified C2-V3 region showed good integrity in the majority of the nucleotide sequences among three clones. However, four common nucleotide substitutions were found in the V3 region and three common changes were found in the C2 region of the CKR-L3- and CCR6-clones compared to the CCR5-clone sequence (Additional file 1). All of the V3-substitutions were synonymous, revealing no amino acid change. Unlike the V3 region, two non-synonymous substitutions, namely lysine (K) to arginine (R) and tryptophan (W) to cysteine (C) were observed in the C2 region. Another substitution of lysine to asparagine (N) was found in the CCR6-clone, but not in the CKR-L3-using isolate (Figure 4a).

The env gene for the V1-V3 region of CKR-L3- and CCR6-variants of SIVsmE660 exhibited remarkable nucleotide substitutions relative to that of the parental CCR5-smE660: the levels were 1.4% (11/773) in smE660-CKR-L3 and 1.2% (9/773) in smE660-CCR6 clones. Of these four substitutions were common for both clones (Additional file 2). Some of the nucleotide changes were synonymous. However, non-synonymous substitutions of the CKR-L3-variant were observed in the V1 region, with the change of two threonines (T) into alanine (A) and valine (V) and one alanine (A) into threonine (T). On the other hand, the CCR6-variant demonstrated two changes, arginine (R) to glycine (G) and vice-versa in the V1 region (Figure 4b). In the V2 region, the CKR-L3-clone showed one amino acid change, arginine (R) to lysine (K). The CCR-6 clone remained unaltered. No amino acid substitution was found in the V3 region of either the CKR-L3 or CCR6 clones. Interestingly, the C2 region showed two amino acid changes in the CKR-L3 variant, glycine (G) to glutamic acid (E) and arginine (R) to lysine (K). Likely, the CCR6-variant also showed changes of glycine (G) to glutamic acid (E) and arginine (R) to lysine (K) and in addition one serine (S) to arginine (R) (Figure 4b).