Evidence of long-lived founder virus in mother-to-child HIV transmission

Abstract

Exposure of the infant's gut to cell-associated and cell-free HIV-1 trafficking in breast milk (BM) remains a primary cause of mother-to-child transmission (MTCT). The mammary gland represents a unique environment for HIV-1 replication and host-virus interplay. We aimed to explore the origin of the virus transmitted during breastfeeding, and the link with quasi-species found in acellular and cellular fractions of breast-milk (BM) and in maternal plasma. The C2-V5 region of the env gene was amplified, cloned and sequenced from the RNA and DNA of BM, the RNA from the mother's plasma (PLA) and the DNA from infant's dried blood spot (DBS) in 11 post-natal mother-infant pairs. Sequences were assembled in Geneious, aligned in ClustalX, manually edited in SeAL and phylogenetic reconstruction was undertaken in PhyML and MrBayes. We estimated the timing of transmission (ETT) and reconstructed the time for the most recent common ancestor (TMRCA) of the infant in BEAST. Transmission of single quasi-species was observed in 9 of 11 cases. Phylogenetic analysis illustrated a BM transmission event by cell-free virus in 4 cases, and by cell-associated virus in 2 cases but could not be identified in the remaining 5 cases. Molecular clock estimates, of the infant ETT and TMRCA, corresponded well with the timing of transmission estimated by sequential infant DNA PCR in 10 of 11 children. The TMRCA of BM variants were estimated to emerge during gestation in 8 cases. We hypothesize that in the remaining cases, the breast was seeded with a long-lived lineage latently infecting resting T-cells. Our analysis illustrated the role of DNA and RNA virus in MTCT. We postulate that DNA archived viruses stem from latently infected quiescent T-cells within breast tissue and MTCT can be expected to continue, albeit at low levels, should interventions not effectively target these cells.

Fig 1. Sample size of VTS cohort.

An illustration of the number of patients in the post-natal vertical transmission study and the subsequent number of cases, comprising mother-infant pairs, selected for this analysis.

Fig 2. Phylogenetic reconstruction of the most recent common ancestor (TMRCA).

This image illustrates the Bayesian maximum clade credibility (mcc) trees inferred in BEAST. Terminal branch colors represent sequences derived from the infant DBS DNA (green), maternal plasma (red), left breast milk DNA (purple), left breast milk RNA (orange), right breast milk DNA (blue) and right breast milk RNA (yellow). Unambiguous internal branches are shaded according to the compartment of origin or shaded black if the origin was ambiguous. Timelines indicate the number of years in the past from the date of sample collection of the infant DBS, indicated as Time 0.

Fig 3. Comparison of TMRCA estimates.

Graphical representations of the upper and lower limit (95% CI) of the most recent common ancestor (TMRCA) estimations of each compartment of the datasets as calculated in BEAST for the infant DBS DNA (green), the mother’s plasma RNA (gray), the left breast milk DNA (purple), left breast milk RNA (yellow), right breast milk DNA (red) and right breast milk RNA (blue).