Genetic variation continues to occur in well-controlled HIV-1-infected patients

Background: HIV-1 viral persistence in light of long-term HAART is a major hurdle to a cure.Genomic editing techniques, such as the CRISPR/Cas9 system, hold the promise to permanently excise integrated virus from a host cell.Targets are defined by a 20-25 nucleotide guide RNA (gRNA) molecule complementary to the desired genomic region.However, due to the rapid mutation rate intrinsic to HIV replication, the virus in patients exists as a collection of distinct genomic variants, termed quasispecies.Presented here is a methodology for designing gRNA sequences which will cleave a patient's HIV quasispecies.Methods: PBMC genomic DNA was isolated from patients in the Drexel Medicine CNS AIDS Research and Eradication Study (CARES) Cohort and the long terminal repeat (LTR) of the HIV-1 quasispecies was sampled using Illumina Next Generation Sequencing (NGS).gRNAs were designed using a known positional binding profile to optimally target the quasispecies present in the sample.Using the entire collection of sequences from the CARES cohort, a package of three gRNAs (CARES-3) were designed.Results: The CARES-3 package alone could cleave 70% of the observed quasispecies in the cohort which accounts for >90% of the observable integrated virus of from any patient examined to date.Supplementing this with a package of personalized gRNAs of less than three additional sequences allowed us to target all patients to below our level of detection.Examining longitudinal samples collected over a five year period suggests that the personalized gRNAs will lose their effectiveness if treatment is delayed more than 6 months.Conclusions: Utilizing NGS sampling techniques and computational gRNA design, a package of less than six gRNAs can be developed which will cleave all observable quasispecies within the PBMC compartment.This work presents a step towards understanding the complex task of using excision therapy to target HIV-1 quasispecies in the infected human population.

Abstract licence: CC BY-NC-ND

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.