CROI 2017 Abstract e-Book

Abstract eBook

Poster and Themed Discussion Abstracts

Methods: Forty-four hunters infected with a gorilla SFV were studied. Serial dilutions of their plasma were incubated with SFV strains and residual viral infectivity was quantified with an indicator cell line expressing the beta-galactosidase gene under the control of the viral LTR. Genomic DNA was extracted from the buffy coat (BC). Env fragments were amplified from BC-DNA using specific primers by nested PCR. PCR products were directly sequenced by MWG Operon (Ebersberg, Germany). Results: Neutralizing activity of plasmas was tested against two zoonotic primary gorilla SFV strains, representative of two env genotypes circulating in Central Africa. Neutralizing activity was detected in > 90% of SFV-infected donors, and titers ranged from 1:22 to 1:14724. Two-thirds of donors recognized only the isolate from the same genotype as their own virus. One third of donors recognized the two genotypes. For half of them, env sequences from the two genotypes were amplified, demonstrating the co- infection by at least two different SFV strains. Most donors infected with a gorilla SFV neutralized chimpanzee SFV, showing the conservation of epitopes targeted by neutralizing antibodies. Patients with broader responses had higher blood viral DNA levels. Conclusion: Gorilla SFVs transmitted to humans led to the generation of high titers of neutralizing antibodies. The high titers of neutralizing antibodies in some of them suggest that active SFV replication may occur in humans with persistent zoonotic SFV infections. This study revealed that a significant proportion of our study population was co-infected by at least two SFV strains. This result is relevant for the emergence of retroviruses in the human population. 191 ESTIMATING THE HERITABILITY OF AN HIV-1 TRAIT IN A CLINICAL SETTING Nadine Bachmann 1 , Teja Turk 1 , Alex Marzel 2 , Claus Kadelka 2 , Thomas Klimkait 3 , Vincent Aubert 4 , Jürg Böni 2 , Huldrych F. Günthard 2 , Roger Kouyos 2 1 Univ Hosp Zurich, Zurich, Switzerland, 2 Univ of Zurich, Zurich, Switzerland, 3 Dept of Biomed, Basel, Switzerland, 4 Univ Hosp Lausanne, Lausanne, Switzerland Background: Parent-offspring(PO) regression is a central tool to determine the heritability of phenotypic traits; i.e., the extent to which those traits are controlled by genetic factors. Its applicability to viral traits is unclear because the directionality of viral transmission is typically unknown and viral phylogenies are sparsely sampled. We hypothesized that PO regression is robust to these potential problems and can provide reliable estimates for the impact of viral genetics on set-point virus load. Furthermore, we test the hypothesis that heritability is not due to clustering of host-demographic factors on the viral phylogeny. Methods: We assessed the applicability of PO regression in a realistic setting using Ornstein-Uhlenbeck simulated data on phylogenies built from 11,442 Swiss HIV Cohort Study(SHCS) partial pol sequences. The SHCS is one of the most densely sampled populations available. As approximation for parent and offspring we used transmission pairs of the phylogeny and randomly identified donor(parent) and recipient(offspring). Set-point virus load(SPVL) data was available for 3,293 patients. Results: We found that misidentification of donor and recipient plays a minor role when measuring heritability and also showed that imperfect sampling does not influence the heritability estimated by PO regression. Our results also provided evidence that heritability is a trait of a specific population and that pairs represent non-random samples from the entire population, implying that heritability measured on pairs differs from that measured on the entire population. A mixed-effect model approach yielded the same results as PO regression but could be extended to clusters of size > 2 and allowed for the correction of confounding effects. Finally, we applied both methods to SPVL data from the SHCS and obtained heritability estimates ranging from 8% to 47% (Figure) that did not change substantially after adjusting for host-demographic factors in the mixed-effect model (±2%). Conclusion: Our study underlines the utility of PO regression and mixed-effect models as flexible and robust tools to estimate the contribution of viral genetics to viral traits under real-life settings. Estimated heritability depends on transmission pair selection criteria - despite lower statistical power, conservative criteria should be preferred, as true heritability is reached at zero genetic distance. We find a strong effect of viral genetics on SPVL and that this effect is not confounded by host demographics.

Poster and Themed Discussion Abstracts

192 NGS ANALYSIS OF HIV-1 GROUP O RT RESIDUE 181C PREVALENCE AND EVOLUTION OVER TIME

Marie Leoz 1 , Myriam Vezain 2 , Elodie Alessandri 1 , Sophie Coutant 2 , Guillemette Unal 3 , Isabelle Tournier 2 , Francois Simon 4 , Jean-Christophe Plantier 1 1 Rouen Univ Hosp, Rouen, France, 2 Normandie Univ, Rouen, France, 3 COREVIH Haute-Normandie, Rouen, France, 4 Saint Louis Hosp, Paris, France Background: HIV-1 group O viruses are endemic in Cameroon and found sporadically in other countries. They are subdivided into subgroups HIV-1/O-H and HIV-1/O-T. Their genetic divergence from HIV-1/M causes polymorphisms on residues associated to HIV-1/M antiretroviral resistance. In all the HIV-1, SIVcpz and SIVgor radiation, a 181Y residue is found in the Reverse Transcriptase (RT), but a Non-Nucleotide RT Inhibitors (NNRTI) resistance mutation Y181C can be naturally present in HIV-1/O, and associated to the recently emerged HIV-1/O-H subgroup. The presence of minor variants at this position, and its evolution over time was investigated. Methods: We used Next Generation Sequencing to study the polymorphism on residue 181 and associated signature residue from the NNRTI-binding pocket in 75 NNRTI-naïve HIV-1/O patients. Evolution of residue 181 over time – with or without NNRTI – was investigated in 8 patients, as well as that of associated residues. Results: Residue 181C was found in 40/75 NNRTI naive patients. Its association with HIV-1/O-H was confirmed (p<0.001). A 181C/Y mixture was found in 7 unlinked individuals. Residues at signature positions were diverse in 181Y viruses but a 28K-103K-142I-174D-178L pattern was highly conserved in 181C viruses. Evolution of residue 181 was similar to what observed in HIV-1/M for one 181Y HIV-1/O-T virus: NNRTI-associated selection of 181C, and reversion after NNRTI interruption. Three HIV-1/O-H viruses selected 181C due to NNRTIs, but conserved it several years after NNRTI interruption. Four HIV-1/O-H viruses evolved without NNRTIs (181C/Y => 181C, n=2, 181C => 181C/Y, n=2). Residues at signature positions did not evolve concomitantly when the 28K-103K-142I-174D-178L pattern was present in a first place, suggesting that this pattern was compatible with both 181C and 181Y residues. Conclusion: Mutation 181C presence and evolution in HIV-1/O is linked to the virus genetic background. It is associated to the emergent H subgroup where it can be naturally present, or conserved after NNRTI selection, probably due to a favourable pattern on associated signature residues.

75

CROI 2017