Forced selection of tRNA(Glu) reveals the importance of two adenosine-rich RNA loops within the U5-PBS for SIV(smmPBj) replication

Abstract

Simian immunodeficiency virus (SIV) and human immunodeficiency virus (HIV-1) preferentially select and use tRNA(Lys,3) as the primer for initiation of reverse transcription. Previous studies have shown that HIV-1 can be forced to use tRNA(Glu) if mutations are made within the primer-binding site (PBS) and a region upstream, A-loop, to be complementary to the 3'-terminal 18 nucleotides and anticodon loop of tRNA(Glu). To examine the primer preference of SIV, mutations were made within the PBS of SIV(smmPBj) to be complementary to tRNA(Glu). Analysis of the production of infectious virus revealed that SIV(smmPBj) with the PBS complementary to tRNA(Glu) retained approximately 80% infectivity of the wild type. However, modification of the U5 of SIV(smmPBj) to alter nucleotides to be complementary to the anticodon of tRNA(Glu), in combination with the PBS complementary to tRNA(Glu), drastically reduced the production of infectious SIV(smmPBj) to less than 1% that of wild type. The replication of SIV(smmPBj) with the PBS complementary to tRNA(Glu) was similar to that of the wild type virus, while the replication of SIV(smmPBj) with PBS and A-loop complementary to tRNA(Glu) was delayed compared to that of wild type virus. Analysis of the PBS regions revealed that the virus with the PBS complementary to tRNA(Glu) reverted quickly, within 4 days, to be complementary to tRNA(Lys,3), while the virus with PBS and A-loop complementary to tRNA(Glu) retained the PBS for a longer time during in vitro culture although following extended replication both the A-loop and PBS of SIV(smmPBj) reverted to be complementary to tRNA(Lys,3). RNA modeling of SIV(smmPBj) U5-PBS by m-fold revealed two potential A-loop regions. Mutations in either A-loop drastically effected replication in human PBMC. Analysis of the A-loops following in vitro replication revealed that both reverted to the wild type sequence. The results of these studies demonstrate that SIV(smmPBj), like HIV-1, preferentially utilizes tRNA(Lys,3) as a primer for reverse transcription for high level replication, but unlike HIV-1 selection may involve the use of two adenosine-rich loops.

Figure 1. HIV-1 and SIV_smmPBj U5 and PBS

Panel A. Nucleotide sequence of HIV-1 and SIV_smmPBj A-loop and PBS. The nucleotides sequence of the wild type SIV (SIV-WT) and HIV-1 (NL4-WT) A-loop and PBS regions are depicted. The PBS for both is complementary to the 3’-terminal 18 nucleotides of tRNA^Lys,3. A second region consisting of GAAAAT is found in both SIV-WT and NL4-WT. Proviral genomes were generated in which the PBS was made complementary to the 3’-terminal nucleotides of tRNA^Glu (SIV-Glu or NL4-Glu). Proviral genomes were constructed in which the A-loop region of SIV-Glu and NL4-Glu was mutated to be complementary to the anticodon region of tRNA^Glu. This resulted in the alteration of the GAAAAT to TAGAGTTGAGT; proviral clones were named SIV-GluAC and NL4-GluAC. Panel B. Infectivity of viruses obtained following transfection into 293T cells. The proviral genomes were transfected into 293T cells and activity of the viruses were determined. Analysis of the p27 (SIV) or p24 (HIV-1) produced from the transfected cultures revealed no differences between the wild type and proviral genomes containing the A-loop or PBS mutations (data not shown). The amount of infectious virus was determined using the JC53-BL assay. The infectivity produced from transfection of wild type (SIV-WT or NL4-WT) was determined and the amounts produced by the PBS and A-loop mutant proviral genomes were compared to this value. The data presented represents infectivity compared to the wild type (SIV-WT or NL4-WT). The data presented is representative of three independent transfections.

Figure 2. Replication of HIV-1 and SIV_smmPBj with U5-PBS complementary to tRNA^Glu

Panel A. Replication of wild type HIV-1 and A-loop-PBS mutants. Infections were initiated in human PBMCs using equal amounts of infectious virus. Supernatants were assayed at the designated intervals for infectious virus using the JC53-BL assay. The identity of the samples are as indicated. Note that for NL4-WT and NL4-Glu a break occurred in the production of infectious virus such that the wild type by Day 14 had peaked at 25,000 rLu/μL whereas NL4-Glu peaked at Day 31 at 9,000 rLu/μL. The amount of infectious virus produced from NL4-GluAC increased throughout the culture, but peaked at approximately 100 rLu/μL. The data presented is representative of three independent experiments. Panel B. Replication of SIV-WT and A-loop-PBS mutants. Infections were initiated in human PBMCs using equal amounts of infectious virus. At three to four day intervals, supernatants were collected from the infected cultures and analyzed for production of infectious virus using the JC53-BL assay. The identity of the samples are as indicated. The data presented is representative of three independent experiments.

Figure 3. Potential A-loop regions of SIV_smmPBj

A mfold of the potential A-loop regions of SIV_smmPBj is presented. Depicted are two potential A-loop regions designated as A-loop 1 and A-loop 2. Note that A-loop 2 was altered in the construction of SIV-GluAC. The first four nucleotides of the PBS complementary to tRNA^Lys,3 are also depicted.

Figure 4. SIV_smmPBj genomes with mutations in the A-loop 1 or A-loop 2

Panel A. The wild type SIV_smmPBj U5-PBS is presented with the two A-loop regions (boxed); the PBS complementary to tRNA^Lys,3 is shaded. The first mutant SIV-CCA1, contains a mutation in A- loop 1 with two adenosines mutated to cytosines. A second proviral genome, SIV-CCA2 contains mutation of two adenosines to cytosines in A-loop 2. mfolds of the RNA from SIV-CCA1 and SIV-CCA2 revealed similar RNA structures as the wild type with the mutated nucleotide substituted in the loop regions (data not shown). Panel B. Effect of the A-loop mutations on virus production. The proviral genomes were transfected in the 293T cells and the amount of virus determined. There were no different effects on the amount of virus produced as determined by the p27 antigen ELISA capture assay. Panel C. Effect of the A-loop mutations on infectivity. The amount of infectious virus produced from transfection was determined by the JC53-BL assay and was compared to that obtained from transfection of SIV-WT (set at 100%). Both A-loop mutations resulted in a reduction in the amount of infectious virus to 3 and 4% that of the wild type virus. Data is representative of three independent transfections. Panel D. Effect of A-loop mutations on the completion of reverse transcription. Equal amounts of the transfection supernatant was used to infect human PBMCs; 24-hours later, PCR was used to determine the amount of viral DNA present. The agarose gel is shown and the lanes are labeled with the amount of high molecular weight DNA used in the PCR reactions (i.e., 2, 4, and 10 ng). Panel E. The amount of product produced from PCR was determined from the band intensity on a 1% agarose gel divided by that obtained from SIV-WT (set at 100%); two nanogram samples are presented in figure. Values presented are for three independent PCR reactions; values varied 2% from each PCR reaction.

Figure 5. Replication of SIV_smmPBj genomes with mutations in A-loop 1 or A-loop 2

Infections were initiated with equal amounts of infectious virus in human PBMCs. The amount of virus produced was monitored by JC53-BL assay. At the end of the experiment, the U5-PBS was amplified from the integrated provirus of infected cells and DNA sequenced to determine the presence of the mutations. The data presented is representative of three independent experiments.