CROI 2016 Abstract eBook

Abstract Listing

Poster Abstracts

Methods: Lentiviral vectors were used to transduce Jurkat and primary T cells with chimeric antigen receptors, which act as surrogates for the T cell receptor upon engagement with the corresponding ligand. These receptors include intracellular CD28 and CD3-zeta signaling domains fused to one of several anti-Her2 scFvs displaying different quantifiable affinities for Her2. Flow cytometry was used to isolate transduced cells. Expression of HIV was measured via luciferase or GFP. Integration of HIV was measured via nested Alu RT-PCR. Results: Receptors were successfully transduced into CD4+ T cells. Preliminary data indicates that CAR activation leads to a more than 3 fold increase in HIV expression. Receptor affinity for Her2 does not correlate linearly with the level of HIV expression, indicating that signaling thresholds must be met for robust proviral transcription. Conclusions: Adjusting the intensity of cellular signals at the time of infection allows us to control the robustness of HIV transcription and the course of the viral infection. 194 High-Frequency Illegitimate Strand-Transfers Result in Defective HIV Genomes Xiaojun Li 1 ; Peihu Fan 1 ; Feng Gao 2 1 Jilin Univ, Changchun, China; 2 Human Vaccine Inst, Duke Univ Med Cntr, Durham, NC, USA Background: Two strand-transfers of nascent DNA fragments during reverse transcription are required to convert viral RNA genomes into a double-stranded proviral DNA genome for completion of each genome replication cycle. However, whether strand-transfers occur at illegitimate sites and how this may affect viral genome replication are not well understood. Methods: Both HIV and HAV RNA templates were prepared in an in vitro T7 transcription system. The reverse transcription were carried out with reverse transcriptases from three retroviruses: HIV-1, HIV-2 and murine leukemia virus (MLV). The nascent cDNA fragments were directly cloned without PCR amplification. The sequences from individual reverse transcript clones were compared to the HIV or HAV template sequence to determine if new sequences contained mismatched sequences caused by illegitimate strand-transfers. Results: A total of 1067 nascent reverse transcripts were sequenced. Most of them (72%) matched to the template sequences, although they randomly stopped across the RNA templates (~400bp). However, over one-quarter of reverse transcripts (28%) contained mismatched 3’ sequences due to illegitimate strand-transfers. The majority (81%) of the illegitimate strand-transfers were disassociated from RNA templates and realigned onto opposite complementary DNA strands. Up to three strand-transfers were detected in a single sequences although the majority of them (93%) contained one strand-transfer. Since the majority of illegitimate strand-transfer fragments were generated from templates at two opposite orientations, they resulted in defective viral genomes and could not been detected by previous methods. Further analysis showed that mutations at pause/ disassociation sites in new reverse transcripts resulted significantly higher strand-transfer rates. We also found significantly higher illegitimate strand-transfer rates for HIV-2 RT (38.2%) and MLV RT (44.6%) than for HIV-1 RT (5.1%). Conclusions: Illegitimate strand-transfers frequently occur during reverse transcription and result in defective HIV genomes. These high frequency illegitimate strand-transfers may play an important role in retrovirus genome replication.

Poster Abstracts

195 Gag-Protease-Mediated Replication Capacity Differs According to HIV-1 Subtype MarionW. Kiguoya 1 ; Jaclyn K. Mann 1 ; Saleha Omarjee 1 ; Lyle McKinnon 2 ; Festus M. Kiogora 2 ; Kamini Gounder 1 ; Blake Ball 3 ; Joshua Kimani 2 ;Thumbi P. Ndungu 1 1 Univ of KwaZulu-Natal, Durban, South Africa; 2 Univ of Nairobi, Nairobi, Kenya; 3 Univ of Manitoba, Winnipeg, MB, Canada Background: The HIV-1 epidemic is heterogeneous with various subtypes prevalent.Subtype-specific differences in disease progression rate have been reported; however the mechanisms underlying this heterogeneity are still unknown.We investigated whether there were subtype-specific replication capacity differences in recombinant viruses possessing patient-derived gag-protease genes from the Majengo sex worker cohort based in Nairobi, Kenya. Methods: Antiretroviral therapy-naïve patients were recruited (n=103).Patient-derived gag-protease NL4-3 chimeric viruses were generated and their replication capacities assayed in an HIV-1-inducible green fluorescent protein reporter T cell line.Replication capacities of NL4-3 recombinant viruses bearing subtypes A, C, D and inter-subtype recombinants gag-protease were compared.An exploratory codon-by-codon analysis was performed using Kruskal-Wallis test to identify amino acid residues associated with differences in replication capacity for the subtype A recombinant viruses. Results: There were 57 (56%) subtype A, 16 (15.5%) C, 13 (12.6%) D and 17 (16%) inter-subtype recombinant viruses studied.There were no significant differences in CD4 T cell counts or viral loads according to subtypes (ANOVA; p=0.53 and p=0.91 respectively). There were significant differences in viral replication capacities between the different subtypes (p=0.0001, ANOVA), with subtypes A and C displaying lower replication capacity compared to subtype D (p=0.0006 and p=0.0004 respectively) and inter-subtype recombinants (p=0.0006 and p=0.0006 respectively).We identified six amino acids that were significantly (p<0.05 and q<0.2) associated with reduced replication capacity in subtype A (L75I, I107L, S125S, S126S, N315N and S499S).The polymorphism 107L and consensus amino acid 315N were significantly more frequent in subtype A (Chi square test; p=10 -8 and p=0.01 respectively) and were associated with reduced replication capacity. Conclusions: These data show a hierarchy of Gag-protease driven replicative fitness where subtypes A/C are less fit than D, which is also less fit than inter-subtype recombinants. The data is consistent with reported differences in subtype-specific differences in rate of disease progression.We identified amino acids that may contribute to these differences. Further studies to better understand the mechanisms underlying these differences are warranted.Furthermore, these data may have implications for the uneven spread and expansion of HIV-1 subtypes in the global epidermic.

76

CROI 2016