Genetic Analysis of HIV Type 1 env Gene in Cerebrospinal Fluid and Plasma of Infected Chinese Paid Blood Donors

I n the 1980s and early 1990s, hundreds of illegal commercial plasma collection stations were established in China because of a shortage of blood products. Usually blood sharing the same blood type was mixed together, and refusion of pooled blood cells was common at that time. Such procedures allowed HIV to be transmitted among these plasma donors, which then spread rapidly from paid blood donors (PBDs) to their spouses and other sexual partners. ¹ Henan province has been severely affected by commercial plasma donations, with an infectious rate of 8.9% in 2004. ² Sequence analysis indicated that most of these PBDs in Henna harbored HIV-1 subtype B′. ³

The central nervous system (CNS) is a major target of HIV-1 infection. HIV infection in the CNS can progress to AIDS dementia complex (ADC) in the late stage of infection, characterized by a clinical syndrome of cognitive, motor, and behavioral dysfunction. ⁴ HIV envelope glycoprotein, which consists of the gp120 surface and gp41 transmembrance subunits, is related to neuronal injury and death. ⁵ The genetic evolution of HIV-1 within the brain is distinct from virus present in blood, ⁶ and HIV-1 env gene sequence differences have been identified between brain and blood. ⁷ It was reported that in a group of infected Chinese blood/plasma donors, the C2-V5 sequences of the HIV-1 env gene from AIDS patients showed a higher divergence and lower diversity compared with those from long-term nonprogressor (LTNP) subjects. ⁸ We have previously characterized the C2-V5 sequences in the CNS and blood of infected Chinese patients; all were from Yunnan province and most of them were injecting drug users (IDUs). ⁹ However, the HIV-1 env gene in the CNS of infected Chinese PBDs has not been well characterized. To gain further insight into the HIV-1 env gene in the CNS of infected Chinese patients, the C2-V5 region of the HIV-1 env gene in the CNS and blood of infected Chinese PBDs, all of whom came from Henan province, was sequenced and analyzed in the present study.

Six HIV-infected patients from Henan province were included in the present study. All these patients were former blood donors, and they have been infected for more than 10 years. All subjects had plasma and cerebrospinal fluid (CSF) HIV RNA >2500 copies/ml (Table 1). Blood and CSF samples were collected from subjects between March 2010 and August 2010, with their written consent and approval from the ethical review committee. Viral RNA was purified from blood and CSF using the QIAamp Viral RNA Mini Kit (Qiagen, Duesseldorf, Germany) according to the manufacturer's protocol, and the purified RNA was then subjected to polymerase chain reaction (PCR) amplification using primer pairs of outer forward (ENV-FO: atgggatcaaagcctaaagccatgtgt) and outer reverse (ENV-RO: gcgcccatagtgcttcctgctgctgc); nested PCR with interior forward primer (ENV-FI: ctgttaaatggcagtctagc) and interior reverse primer (ENV-RI: acttctccaattgtccctcat) was used to amplify the C2-V5 regions of the env gene. Multiple HIV-1-negative controls were utilized to detect any possible contamination. PCR products were cloned using the TA cloning system (Takara, Dalian, China). For each patient, approximately six randomly selected clones per sample were sequenced on an ABI 3730 genetic analyzer. Overall, 41 clones from plasma and 37 clones from CSF were sequenced.

OPEN IN VIEWER

Table 1.

Subject Characteristics

Patient number	Gender	Age	Current CD4 (cells/mm3)	Plasma RNA (copies/ml)	Cerebrospinal fluid RNA (copies/ml)
1	Male	37	24	630,000	14,000
2	Male	45	46	3,600,000	80,000
3	Male	37	40	350,000	1,000,000
4	Female	45	195	2,200,000	440,000
5	Male	42	10	120,000	280,000
6	Male	55	10	16,000	2,700

The nucleotide sequences were assembled and error checked (the GenBank accession numbers are JF773483–JF773560). The C2-V5 sequences from each clone were aligned with reference sequences from different HIV-1 subtypes available from the Los Alamos HIV-1 databases and GenBank by the Clustal W multiple sequence alignment program from the Mega 5.0 software package. The sequences exhibited the usual alterations in the C2-V5 region of the env gene, especially in the variable region. Phylogenetic analysis was performed to assign subtype classification. The phylogenetic tree was constructed by the neighbor-joining method and Kimura two-parameter model, and the tree was evaluated by the bootstrap resampling method using 1000 replicates as implemented in Mega 5.0. All sequences clustered closely with subtype B′ (the reference strain is U71182.1) and B (the reference strains are AY945710 and DQ990880) (Fig. 1). None of the sequences displayed any intrapatient contamination, and the sequences from the CNS and blood were intermingled with each other. Of note, sequences from Patient 1 fell on different branches; four clonal sequences of Patient 1 from the CNS clustered closely with the sequences of Patient 5. Previous studies have demonstrated that HIV-1 sequences clustered together by organ in phylogenetic analyses; brain-specific evolution may be due to the host immune response and environmental factors within the CNS that influence selection pressures. ¹⁰

OPEN IN VIEWER

FIG. 1.

Phylogenetic reconstruction from envelope (C2-V5) clonal sequences in cerebrospinal fluid (CSF) and plasma samples of six HIV-infected Chinese paid blood donors (PBDs). The phylogenetic tree was constructed by the neighbor-joining method from the Kimura two-parameter model and was evaluated by the bootstrap resampling method using 1000 replicates as implemented in Mega 5.0. Reference strains were obtained from GenBank and the Los Alamos HIV database. Sequences from plasma are shown as solid circles and sequences from CSF are shown as open circles. The scale bar represents 1% genetic distance (0.01 substitution per site).

The intraorgan nucleotide diversity of the viral quasispecies was analyzed by the nucleotide distance between each viral quasispecies from the CNS and blood (Fig. 2), The Kimura two-parameter method was calculated for clonal nucleotide sequences using the distance program from the Mega 5.0 software package. The statistical significance of the differences between the CNS and blood was determined using nonparametric statistics, the Mann–Whitney test. We found that the levels of diversity from the CNS were significantly lower than those from blood (p<0.0001). The virus from blood has a broader quasispecies, which may be caused by increased viral replication. However, the levels of diversity from the CNS are lower, likely owing to the constraints on viral replication in the brain. ¹¹ The intraorgan nucleotide divergence was analyzed; however, analysis did not reveal any significant difference between the CSF and plasma samples (data not shown).

OPEN IN VIEWER

FIG. 2.

Diversity analyses of envelope (C2-V5) clonal sequences in cerebrospinal fluid (CSF) and plasma samples of six HIV-infected Chinese paid blood donors (PBDs). The Kimura two-parameter corrected intraorgan diversity was calculated for clonal nucleotide sequences using the distance program from the Mega 5.0 software package. The median (line in the box) and quartile (box) differences between viral quasispecies were plotted as different sample types. Statistical significance of the differences between CSF and plasma was determined using nonparametric statistics, the Mann–Whitney test.

The deduced C2-V5 region of HIV env amino acid sequences was aligned with reference B (GenBank accession number: k03455) by the Clustal W multiple sequence alignment program from Mega 5.0. The envelope region contains genetic sequences that are responsible for cell tropism; the third variable loop (V3) alone appears to be an important determinant for coreceptor usage. The phenotype prediction programs based on V3 sequences have been developed to predict coreceptor usage. We first used the 11/25 charge rule to characterize coreceptor usage in different organs. ¹² In the study, the amino acids are serine (S) at site 11 in all the sequences from CNS, and the amino acids are serine (S) at site 25 in most of the sequences from blood, except for the sequences from blood of Patient 1. As far as the sequences from blood of Patient 1 are concerned, seven clones were sequenced. Two clonal sequences harbor arginine (R), four clonal sequences harbor lysine (K), and one clonal sequence harbors serine (S) at site 25. According to the 11/25 charge rule, all viral quasispecies from the CNS of all patients use CCR5 as coreceptor and all viral quasispecies from blood use CCR5 as coreceptor except for the viral quasispecies of Patient 1. Six HIV strains from blood of Patient 1 whose sequences harbor arginine (R) or lysine (K) at site 25 use CXCR4 as the coreceptor; however, one strain whose sequence harbors serine (S) at site 25 uses CCR5 as the coreceptor. Based on the HIV-1 V3 sequence, coreceptor usage was also analyzed by the Geno2pheno (http://coreceptor.bioinf.mpi-inf.mpg.de/) and PSSM (http://indra.mullins.microbiol.washington.edu/webpssm/) tools online. Consistent with the findings analyzed by the 11/25 charge rule, most of the HIV strains use CCR5 as the coreceptor except for the strains from the blood of Patient 1. On the whole, the genotypic classification based on the 11/25 rule, Geno2pheno, and PSSM is similar; all viral quasispecies from CNS use CCR5 as the coreceptor in the study.

Cytotoxic T lymphocyte (CTL) activities to variable targets may play an important role in viral control, and HLA-A2 and HLA-A11 are two predominant alleles in the Chinese. ¹³ Gag-specific CTL epitopes across the Henan isolate genome have been identified, and CTL responses restricted by HLA-A2 and HLA-A11 have been evaluated among treated PBDs. ¹⁴ HIV strains can escape CTL responses by mutational events, and the sequences in the present study exhibited the usual alterations in the variable region of the env gene. Alexander-Miller et al. identified an HLA-A2-restricted CTL epitope in the V3 region of the env gene, ¹⁵ although when compared with the aligned amino acids in the study, the identified epitope was not conserved (data not shown).

In summary, we characterized the C2-V5 region of the env gene from the CSF and plasma samples, which were obtained from six HIV-infected PBDs. All the sequences from Henan province clustered closely with subtypes B′ and B, and the levels of diversity from the CNS were significantly lower than those from blood (p<0.0001). In addition, all viral quasispecies from the CNS use CCR5 as the coreceptor. These data provide valuable information on the pathogenesis of HIV in the CSF and plasma of infected Chinese PBDs, and our findings could enhance insights into HIV-associated neurological disease.

GenBank Accession Numbers

The GenBank accession numbers of the nucleotide sequences reported in this paper are JF773483–JF773560.