Previous Article | Next Article ![]()
Journal of Virology, January 2003, p. 685-695, Vol. 77, No. 1
0022-538X/03/$08.00+0 DOI: 10.1128/JVI.77.1.685-695.2003
Copyright © 2003, American Society for Microbiology. All Rights Reserved.
Laboratory of Molecular Virology and Epidemiology, AIDS Research Center, National Institute of Infectious Diseases, Shinjuku-ku, Tokyo 162-8640, Japan,1 Kunming Institute of Zoology, Chinese Academy of Sciences, Kunming, Yunnan 650223, People's Republic of China2
Received 3 September 2002/ Accepted 26 September 2002
|
|
|---|
|
|
|---|
Three CRFs play a critical role in the HIV-1 epidemic in Asia. CRF01_AE was originally identified in Thailand (13, 15) and is being spread throughout Southeast Asia (22). Two closely related CRFs, CRF07_BC and CRF08_BC, have been identified among injecting drug users (IDUs) in China (16, 19). CRF07_BC (prototype strains, 97CN54 and 97CN001) was distributed among IDUs in Xinjiang Province in northwest China (19), while CRF08_BC (prototype strain, 97CNGX6F) was circulating widely among IDUs in Guangxi Province in southeast China (16). Each CRF appears to be associated with a different overland heroin trafficking route (1): CRF07_BC has spread northwestward to Xinjiang Province, while CRF08_BC has spread eastward to Guangxi Province, from their common origin, presumably in Yunnan, where subtypes B' and C are cocirculating (16, 19). Yunnan Province in southwestern China is thus thought to be an epicenter of the HIV-1 epidemic in China, where the cumulative number of HIV cases would be expected to reach 10 million by 2010 with the current rate of increase (30%) (20).
It has been reported that the HIV-1 epidemic among IDUs in Yunnan was initiated with both HIV-1 subtype B (same lineage of subtype B strains isolated in the United States and Europe) and subtype B' (Thailand variant of subtype B, also referred to as Thai-B) strains (8, 21, 22) in the late 1980s. Subtype B' appeared to replace subtype B of U.S.-European lineage, increasing from 20% of all subtype B in strains 1990 to 90% in 1996 (8, 15, 21, 22). Subsequently, HIV-1 subtype C strains were identified among IDUs in China by the early 1990s (11). A recent study revealed unique geographical differences in distribution of HIV-1 subtypes among IDUs in Yunnan Province (24). HIV-1 subtype B' and various forms of unique HIV-1 intersubtype B'-C recombinants were distributed in western Yunnan (Dehong Prefecture). In contrast, CRF08_BC predominated and CRF07_BC was detected at low prevalence in eastern Yunnan (Wenshan and Honghe Prefectures) near the border with Guangxi Province. Moreover, unique recombinant forms that showed structural similarity to CRF07_BC were detected in eastern Yunnan (24).
We report here the identification and characterization of a new class of HIV-1 recombinants comprised of two previously established CRFs (CRF07_BC and CRF08_BC) among unique recombinant forms detected in eastern Yunnan Province and discuss the spread of these novel recombinants and their biological implications.
As a part of cohort study of HIV-hepatitis C virus coinfection in Yunnan Province (27), 57 consenting HIV-1-positive IDUs were recruited from drug detoxification and rehabilitation centers in Honghe (44 IDUs) and Wenshan (13 IDUs) Prefectures during the period August 2000 to March 2001. The subjects included 43 males and 14 females with an age range of 17 to 43 years (mean, 29.1 years). The mean duration of drug use was 5 years (range, 1 to 11 years). Fourteen additional specimens from IDUs in Dehong Prefecture in west Yunnan Province were analyzed in parallel (24). EDTA-treated blood samples were drawn from HIV-1-positive individuals, and the peripheral blood mononuclear cells (PBMCs) were separated by Ficoll-Hypaque (Pharmacia, Piscataway, N.J.) density gradient centrifugation. For virus isolation, PBMCs from HIV-1-positive individuals were cocultured with phytohemagglutinin (2 µg/ml)-stimulated CD8+-T cell-depleted PBMCs from HIV-negative healthy donors in RPMI 1640 containing 10% fetal calf serum and interleukin-2 (20 U/ml). Virus production was detected by virion-associated reverse transcriptase (RT) assay as described previously (9). Plasma specimens were saved for genotype screening based on the nucleotide sequence determination of virion HIV-1 RNAs (24).
A total of 25 HIV-1 strains were isolated from 57 HIV-1-positive specimens from eastern Yunnan, and their genotypes were screened on the basis of the nucleotide sequences of the gag-RT regions (24), where CRF07_BC and CRF08_BC, which are circulating widely among IDUs in China, display distinct profiles of recombination between subtypes B' and C (16, 19, 24). The phylogenetic analyses revealed that the majority of HIV-1 strains circulating in eastern Yunnan province belonged to CRF08_BC (21 of 25; 84%) and that CRF07_BC was found only at a low prevalence (2 of 25; 8%). However, the remaining two isolates (00CN-HH069 and 00CN-HH086) were classified as outliers placed outside the clusters of CRF07_BC and CRF08_BC (24) (data not shown). They were isolated from 24- and 29-year-old asymptomatic male IDUs in Honghe Prefecture in eastern Yunnan Province who started their injecting drug use relatively recently (from October 1998 and July 1999, respectively). The specimens for the present study were collected in August 2000. The interperson distance of 21 CRF08_BC strains identified in eastern Yunnan was remarkably low: 1.4% ± 0.4% (mean ± standard deviation) (range, 0.4 to 2.4%) for the 1.4-kb pol region and 1.5% ± 0.6% (range, 0.7 to 2.4%) for the 2.6-kb gag-RT region (n = 21). According to a study based on specimens collected from IDUs in neighboring Guangxi Province in July 1996 and July 1997, CRF08_BC prevailing in this region was highly homogeneous, with an interpatient diversity of 0.4% (for all structural genes) and 0.9% (for regulatory genes), compared with the reference strains of subtype C from Botswana and subtype A from Uganda, Somalia, and Tanzania (7.6 and 8.1%, respectively) (16). These results suggest that a common ancestor of CRF08_BC has been introduced into IDU populations in eastern Yunnan and Guangxi Provinces very recently (16). It is plausible that this ancestor might have been arisen in Yunnan, since subtypes B' and C circulated among IDUs in Yunnan in the early 1990s (11, 22, 23), while subtype C was first introduced into IDUs in Guangxi around 1996 and 1997 (3, 25, 26), as suggested by Piyasirisilp et al. (16).
To elucidate the complete genome structures of the two Honghe isolates with outlier genotypes (00CN-HH069 and 00CN-HH086), we cloned and determined the nucleotide sequences of their near-full-length HIV-1 proviral genomes amplified by PCR as described previously (10, 18). Briefly, DNAs were extracted from CD8-depleted phytohemagglutinin-stimulated PBMCs infected with the HIV-1 strains of interest by use of a blood and cell culture DNA midi kit (Qiagen, Hilden, Germany). The near-full-length HIV-1 genomes were amplified by using the Expand long-template PCR system (Boehringer GmbH, Manheim, Germany) with the primer set of pbs-496A (sense, 5'-AGTGGCGCCCGAACAGG-3' [the NarI site is underlined]) (7) and U5-497B (antisense, 5'-GGTCTGAGGGATCTCTAGTTACCAG-3'). Purified PCR fragments were ligated with a pBR322-based vector carrying an Xcm I site for TA cloning (12) (for 00CN-HH069) or with the pCR TOPO XL vector (Invitrogen, Carlsbad, Calif.) (for 00CN-HH086). Positive clones with 9-kb near-full-length inserts were selected, and the nucleotide sequences of near-full-length HIV-1 genomes were determined on both strands by the direct sequencing method with fluorescent dye terminators in an automated ABI PRISM310 DNA sequencer (Applied Biosystems, Inc., Foster City, Calif.), using the primer-walking approach. The resulting near-full-length DNA clones were designated 00CN-HH069.1 and 00CN-HH086.1. Both clones had intact open reading frames for all nine HIV-1 genes. Phylogenetic tree analysis based on the near-full-length nucleotide sequences revealed that they were placed between the monophyletic clusters of CRF07_BC and CRF08_BC, suggesting their close relationship to these two CRFs (Fig. 1).
![]() View larger version (35K): [in a new window] |
FIG. 1. Neighbor-joining tree analysis based on near-full-length nucleotide sequences of 00CN-HH069 and 00CN-HH086 with HIV-1 group M (subtypes A to D, F to H, J, and K) and CRF reference sequences (http://hiv-web.lanl.gov/content/hiv-db/SUBTYPE_REF/Table1.html). SIVCPZGAB was used as an outgroup but is not shown for simplicity. Bootstrap values (>80) are shown at the corresponding nodes. Subtypes and CRF clades are shown outside the tree. The clusters of CRF07_BC and CRF08_BC are circled. Boxed are 00CN-HH069 (HH069) and 00CN-HH086 (HH086).
|
![]() View larger version (43K): [in a new window] |
FIG. 2. Bootscanning analyses of near-full-length HIV-1 nucleotide sequences of 00CN-HH069 and 00CN-HH086. Two sets of bootscanning plots were performed. (A to D) Bootscanning plots (17), depicting the relationship of CRF07_BC (97CN001) (A), CRF08_BC (97CNGX6F) (B), 00CN-HH069 (C), and 00CN-HH086 (D) to the representative strains of HIV-1 subtypes (A, B', C, and D) and CRF01_AE. (G and H) The second set of bootscanning plots, delineating the relationship of 00CN-HH069 (G) and 00CN-HH086 (H) to the reference strains for CRF07_BC (97CN001) and CRF08_BC (97CNGX6F) with the respective representatives of HIV-1 subtypes A and D and CRF01_AE. (E and F) Control plots for the reference strains of CRF07_BC (97CN001) and CRF08_BC (97CNGX6F), respectively. The bootstrap values are plotted for a window of 500 bp moving in increments of 50 bp along the alignment. The reference strains used are shown at the bottom.
|
To confirm their recombinant structures, exploratory tree analyses were performed along the HIV-1 genomes of these two strains. Based on the data on the putative recombination breakpoints estimated by the profiles of bootscanning plots (Fig. 2G and H), the HIV-1 genomes of these two strains were divided into eight segments (I through VIII in Fig. 3). Each segment was then analyzed by constructing a neighbor-joining tree to confirm the genotype (Fig. 3 and Table 1). The stability of the nodes was assessed by bootstrap analysis (5) with 100 replications, using maximum parsimony (6). As shown in Fig. 3 and Table 1, each query strain showed congruence to either the CRF07_BC or CRF08_BC cluster, with high bootstrap confidence (93 to 100%) in most of the areas (segments I, II, III, V, VI, and VIII). These two strains formed a significant cluster with CRF08_BC, with a modest level of bootstrap confidence (66 to 67%) in segments IV, while those segments were very short (159 bp for 00CN-HH069.1 and 100 bp for 00CN-HH086.1) (Fig. 3 and Table 1). Since the middle parts of the env regions were approximately equidistant phylogenetically to those of CRF07_BC and CRF08_BC (segments VII in Fig. 3), the bootstrap values were lower than those in the other areas (Fig. 2G and H). However, even in these areas, 00CN-HH069.1 and 00CN-HH086.1 were more closely related phylogentically to CRF07_BC and CRF08_BC (with bootstrap confidence of 97 to 100%) than the other subtype C reference strains (segments VII in Fig. 3).
![]() View larger version (64K): [in a new window] |
FIG. 3. Exploratory tree analysis. (A) 00CN-HH069; (B) 00CN-HH086. Using bootscanning and distance scanning techniques (Fig. 2G and H), the plausible recombination breakpoints were identified and HIV-1 genomes were then divided into eight segments (I through VIII). Each segment was then used in separate phylogenetic analyses based on the neighbor-joining method to confirm the subtype (or CRF) origin of the segment. The stability of the nodes was assessed by using maximum parsimony (6) with a bootstrap value of 100 replications (5). SIVCPZGAB was used as an outgroup but is not shown for simplicity. Nucleotide positions are numbered by the HXB2 numbering engine (http://hiv-web.lanl.gov/content/hiv-db/NUM-HXB2/HXB2.MAIN.html). The analysis starts from gag open reading frame. The bootstrap values with which the cluster containing each query strain is supported are marked with ovals at the corresponding nodes. , 00CN-HH069; , 00CN-HH086. The genotype assignments in the separate phylogenetic analyses are indicated within the box at the right of the respective tree. 07, CRF07_BC; 08, CRF08_BC; 07/08, equidistant to both CRF07_BC and CRF08_BC.
|
|
View this table: [in a new window] |
TABLE 1. Summary of genotype assignment, based on exploratory free analysis, in different regions of HIV-1 genomes of two putative inter-CRF recombinants (00CN-HH069.1 and 00CN-HH086.1)
|
|
View this table: [in a new window] |
TABLE 2. Informative-site analyses of putative inter-CRF recombinants (HH069 and HH086) comprised of CRF07_BC and CRF08_BCa
|
B sites (NF-
B II and NF-
B III) of these two strains were comprised of two nucleotides (5'-GC-3') that appear to be specific to CRF08_BC strains known to date (four of four). In contrast, three or four spacer nucleotides are commonly observed in subtype C and CRF07_BC strains (mostly GCC or GCT for subtype C strains and GCG for CRF07_BC). (ii) All known CRF07_BC strains (four of four) harbor the unique nucleotide substitution at the 3' end of the proximal NF-
B motif (NF-
B I), being GGGGCGTTCT or GGGGTGTTCT instead of GGGGCGTTCC of the subtype C consensus. In contrast, 00CN-HH069.1 and 00CN-HH086.1 had the proximal NF-
B motif (5'-GGGGCGCCTT-3') specifically found in most subtype C strains and all known CRF08_BC (four of four) strains (Fig. 4B).
![]() View larger version (42K): [in a new window] |
FIG. 4. Phylogenetic relationship and possible sequence signatures of LTRs of 00CN-HH069 and 00CN-HH086. (A) Neighbor-joining tree based on 3'-LTR sequences of 00CN-HH069 and 00CN-HH086 with HIV-1 group M and CRF reference sequences (7) with bootstrapping (100 replications). Bootstrap values (>60) are shown at the corresponding nodes. The bootstrap values with which the cluster containing each query strain is supported are marked in ovals on the corresponding nodes. Subtype and CRF assignments are shown outside the tree. HH069, 00CN-HH-69; HH086, 00CN-HH086. (B) Nucleotide sequence alignment of enhancer-core regions of the 3' LTR in comparison with HXB2 and other subtype C strains. Dashes indicates the sequence identity with HXB2. Periods indicate gaps introduced to improve the alignment. The NF- B-binding motif (consensus sequence, 5'-GGGRNNYYCC-3') is boxed. Lowercase letters in the box with the gray margin are the substitutions that differ from the NF- B consensus. Shaded areas are the putative sequence signatures specific to CRF07_BC (uniform nucleotide substitution from C to T in the proximal NF- B site; G substitution at 3' end of trinucleotide spacer sequence between NF- B II and III) or CRF08_BC (GC dinucleotide spacer sequence between NF- B II and III). ICR, putative second-generation inter-CRF recombinants.
|
![]() View larger version (47K): [in a new window] |
FIG. 5. Schematic representation of subtype structures and deduced profiles of crossovers between CRF07_BC and CRF08_BC in new second-generation inter-CRF recombinants identified in Honghe Prefecture of Yunnan Province. The regions of subtypes B' and C are as indicated at the bottom. The predicted profiles of crossovers between CRF07_BC (07) and CRF08_BC (08) deduced from bootscanning plots (Fig. 2 and 7), exploratory tree analyses (Fig. 3; Table 1), and informative-site analyses (Tables 2 and 3) are superimposed on the subtype structures of plausible inter-CRF recombinants. (A) CRF07_BC (97CNXJ001); (B) CRF08_BC (97CNGX6F); (C) 00CN-HH069.1; (D) 00CN-HH086.1; (E) 00CN-HH003; (F) 00CN-HH004; (G) 00CN-HH029. The dotted lines in the segments of CRF08_BC in the 3' half of the genome, corresponding to env open reading frames, are the areas that are approximately equidistant genetically to both CRF07_BC and CRF08_BC but are distantly related to other subtype C strains. (08) in panel D indicates the area that was detected and supported by both bootscanning plots (Fig. 2) and exploratory tree analysis (Fig. 3) but was not supported by informative-site analysis (Table 2). The subtype and CRF assignments of LTR regions were based on the data presented in Fig. 4. Positions of identical recombination breakpoints between CRF07_BC and CRF08_BC, which were predicted from informative-site analyses (boldface numbers in Tables 2 and 3), are marked with arrowheads with different shadings.
|
![]() View larger version (37K): [in a new window] |
FIG. 6. Neighbor-joining tree analysis based on the nucleotide sequences of 2.6-kb gag-RT regions of HIV-1 samples with outlier genotypes from Yunnan Province. The HIV-1 samples with outlier genotypes (shaded) were placed between the clusters of CRF07_BC and CRF08_BC. 00CN-HH069.1 and 00CN-HH086.1 are boxed. The arrowheads indicate the HIV-1 samples analyzed in the present study: solid arrowheads, HIV-1 samples from Honghe Prefecture; open arrowheads, HIV-1 samples from Wenshan Prefecture.
|
![]() View larger version (61K): [in a new window] |
FIG. 7. Bootscanning analyses based on the nucleotide sequences of 2.6-kb gag-RT regions of HIV-1 samples with outlier genotypes. Analyses were performed as described for Fig. 2. The first sets of the bootscanning plots, depicting the relationship to the representative strains of HIV-1 subtypes (A, B', C, and D) and CRF01_AE, are shown at the left of each panel. The second sets of bootscanning plots, depicting the relationship to the reference strains for CRF07_BC (97CN001) and CRF08_BC (97CNGX6F) with the respective representatives of HIV-1 subtypes A and D and CRF01_AE, are shown at the right of each panel. (A) 97CN001 (CRF07_BC); (B) 97CNGX6F (CRF08_BC); (C) 00CN-HH069.1; (D) 00CN-HH086.1; (E) 00CN-HH003; (F) 00CN-HH004; (G) 00CN-HH029. The bootstrap values are plotted for a window of 200 bp moving in increments of 30 bp along the alignment. Reference strains used for two different sets of bootscanning plots are shown in the boxes at the bottom right: A_92UG037, B'_RL42, C_95IN21068, D_NDK, 01_93TH253, 07_97CN001, and 08_97CNGX6F. Positions of identical recombination breakpoints between CRF07_BC and CRF08_BC, which were predicted from informative-site analyses (boldface numbers in Tables 2 and 3), are marked with vertical arrowheads with different shadings.
|
|
View this table: [in a new window] |
TABLE 3. Summary of informative-site analysis based on the nucleotide sequences of 2.6-kb gag-RT regions of HIV-1 samples with outlier genotypes from eastern Yunnan Provincea
|
The inherent ability of the retroviral replication machinery to switch template leads to the generation of diverse forms of recombinants, and such recombination between heterologous RNA molecules may conceivably generate strains with novel qualities (4). It is believed that nonrecombinant forms of HIV-1 subtype C had been introduced into Yunnan Province by the early the 1990s (11). However, the nonrecombinant form of HIV-1 subtype C was not detected among samples from IDUs in Yunnan in 2000 and 2001, apparently having being replaced by newly emerged recombinants (24). It is thus tempting to speculate that this drastic shift of strains was not caused by chance but may reflect a yet-undefined selective advantage(s) of recombinants over the preexisting parental strains. Recombination events might provide an opportunity to yield more fitness (e.g., more efficient transmission) compared to that of the parental viruses. The regions in southwestern China (24) and northern Myanmar (14) where new recombinants appear to be arising continually might provide unique opportunities to investigate the natural history and biological consequences of HIV-1 recombination. The present study also suggests that the mixing of different lineages of HIV-1 strains in highly exposed populations and in the social networks known in southwestern China could quickly lead to the evolution of new forms of recombinants and even second-generation recombinants between the previously established first-generation CRFs.
The emergence of such new generations of recombinants could further complicate the development of effective vaccines to limit HIV-1 spread. HIV vaccine efficacy, once established, might be quickly challenged by novel recombinant strains acquiring new sets of escape mutations in viral epitopes by shuffling different parts of HIV-1 genomes. The present study may provide insight into the genesis of the HIV-1 epidemic in China and implications for future vaccine strategies.
Nucleotide sequence accession numbers. The near-full-length nucleotide sequences of 00CN-HH069.1 and 00CN-HH086.1 and the nucleotide sequences of the 2.6-kb gag-RT regions of 00CN-HH003, 00CN-HH004, and 00CN-HH029 are available under GenBank accession numbers AP005206, AP005207, AB090997, AB090998, and AB090999, respectively.
This study was supported by a grant-in-aid for AIDS research from the Ministry of Health, Labour and Welfare and the Ministry of Education, Science and Technology of Japan and by the Japanese Foundation for AIDS Prevention (JFAP). R.Y. and S.K. are the recipients of Research Resident Fellowships from JFAP.
|
|
|---|
sray/download.)
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»