HIV-1 phenotype and genetic variation in vertical transmission

Abstract

HIV-1 vertical transmission is increasing worldwide despite that successful strategies are available. Reduction of vertical transmission has been established by antiretroviral prophylaxis and elective Caesarian section, but this has been limited to the developed countries. In the developing world the lack of resources, the necessity of breastfeeding, social and ethical factors, have all contributed to high transmission rates. Thus, a better understanding of the basic mechanisms of vertical transmission of HIV-1 is important for future preventions in the developing world.

This thesis was based on blood samples from HIV-1 infected children and mothers collected prospectively between March 1987 and September 1994, following a national program for pregnant women living in Sweden. In addition, sequence data from the HIV database in Los Alamos National Laboratory was analyzed to evaluate our results with data representative of globally circulating HIV-1.

HIV-1 isolates were subjected to co-receptor determination in U87 cell lines, expressing CD4 and chemokine receptors CCR5 or CXCR4, and additional receptors used by HIV-1. Coreceptor use was determined by syncythia induction together with p24 ELISA assays or env V3 directed PCR and by infection in GHOST(3) cell lines expressing these chemokine receptors. PCR was also used to confirm virus growth together with p24 ELISA assays. Samples were subjected to limiting dilution nested PCR, also directed towards env V3, to create single molecular clones for sequencing in conjunction with sequencing of the isolates. Phylogenetic analysis was used to obtain the correct genetic clade for each patient. It was also used to map the viral populations in two mother-child pairs and in another four subtype D infected women.

In the first study, we analyzed the co-receptor use of HIV-1 in 24 children over time, to relate clinical parameters with the emergence of X4 virus. In four children, the X4 phenotype emerged after immune deficiency was established, and was hence not the directly contributing cause of AIDS. In the second study, we wanted to examine the phenotypes in mother-child pairs. A statistically significant link was found between the presence of X4 in the mothers and the development of X4 in the children. To evaluate the nature of this link, we studied two mother-child pairs and related the maternal virus from during the pregnancies with the virus in the children. We could in detail describe the transmission of R5 from both mothers, followed by a transmission-independent evolution of X4 from the established R5 populations in the children. We preliminary observed a possible concordance between the V3 loop N-linked glycosylation motif and the R5 phenotype in these mother-child pairs, while a high V3 net charge was in accordance with X4 and dual-tropic viruses.

We expanded our phylogenetic mapping of virus populations with four infected mothers of uninfected children. Here we could show a highly statistically significant association among R5 sequences to the N-linked glycosylation motif in the V3 loop of gp 120. There was also an accumulation of X4 virus within this patient group in the context of subtype D. To examine the fidelity of the association between the V3 loop N-linked glycan and R5 viruses, we analyzed 176 individual sequences of group-0, -N and most group-M clades of HIV-, including recombinants. The glycosylation motif NNT was highly conserved among the M-group clades (99.2%). This strongly supported our conclusion of the importance of the N-linked glycosylation motif of the V3 loop for CCR5 binding, and hence possibly for transmission.

In summary, we have shown the population dynamics of HIV-1 in detail with regard to coreceptor use over time in HIV-1 infected mothers and children, and have identified a possible marker for transmissibility.