High-resolution functional mapping of the venezuelan equine encephalitis virus genome by insertional mutagenesis and massively parallel sequencing

Abstract

We have developed a high-resolution genomic mapping technique that combines transposon-mediated insertional mutagenesis with either capillary electrophoresis or massively parallel sequencing to identify functionally important regions of the Venezuelan equine encephalitis virus (VEEV) genome. We initially used a capillary electrophoresis method to gain insight into the role of the VEEV nonstructural protein 3 (nsP3) in viral replication. We identified several regions in nsP3 that are intolerant to small (15 bp) insertions, and thus are presumably functionally important. We also identified nine separate regions in nsP3 that will tolerate small insertions at low temperatures (30°C), but not at higher temperatures (37°C, and 40°C). Because we found this method to be extremely effective at identifying temperature sensitive (ts) mutations, but limited by capillary electrophoresis capacity, we replaced the capillary electrophoresis with massively parallel sequencing and used the improved method to generate a functional map of the entire VEEV genome. We identified several hundred potential ts mutations throughout the genome and we validated several of the mutations in nsP2, nsP3, E3, E2, E1 and capsid using single-cycle growth curve experiments with virus generated through reverse genetics. We further demonstrated that two of the nsP3 ts mutants were attenuated for virulence in mice but could elicit protective immunity against challenge with wild-type VEEV. The recombinant ts mutants will be valuable tools for further studies of VEEV replication and virulence. Moreover, the method that we developed is applicable for generating such tools for any virus with a robust reverse genetics system.

Figure 1. Generation of VEEV nsP3 insertion library.

Entranceposon M1 – Kan^R was transposed into pBB300 and then processed to generate a library of clones that contained 15 base-pair inserts in nsP3 (and short flanking regions in nsP2 and nsP4). (A) Diagram of plasmids pBB300 and pBB305 and Entranceposon M1 – Kan^R. T7 - T7 promoter; nsP1-nsP4 – VEEV non-structural proteins; SG – VEEV subgenomic promoter; C, E3, E2, 6K, E1 – VEEV structural proteins. (B) Procedure for making full-length VEEV library with short insertions in nsP3.

Figure 2. Functional mapping by capillary electrophoresis.

Electropherogram data from five different amplicons spanning nsP3 are shown in panels A-E and summarized in panel F. The X axis is the DNA fragment size, and the Y-axis is relative fluorescence (proportional to the number of integrations at that site). From top to bottom in each panel, the electropherograms represent virus grown at 30°C, 37°C, and 40°C, and the unselected control (lib). Regions of interest are shaded in grey. The approximate beginning and end of nsP3 are indicated by dotted lines in panels (A) and (E), respectively. The end of the conserved region, and the beginning of the hypervariable region are indicated in panel (C). (A) Amplicon BBU02+BBU017, nts 3932–4684; (B) Amplicon BBU04+BBU018, nts 4281–5059; (C) Amplicon BBU06+BBU019, nts 4613–5350; (D) Amplicon BBU08+BBU020, nts 4961–5701; (E) Amplicon BBU010+BBU021, nts 5236–5803; (F) Compilation of nsP3 genetic mapping data. Regions intolerant to 15 bp insertions at all temperatures are shown in blue. Regions tolerant to insertions at 30°C, but intolerant at 37°C or 40°C are indicated in green. Regions tolerant to insertions at all temperatures are indicated in yellow. The locations of the macro domain, conserved region, and hypervariable region are shown.

Figure 3. Frequency of insertion sites found in unselected RNA versus vRNA from virus produced at 30°C.

The frequency of transposon insertions at each nucleotide position in the VEEV genome was calculated from the GS-FLX sequencing data and normalized to account for differences in the total number of sequencing reads obtained from each sample. For this histogram, the VEEV genome was divided into bins of 50 nucleotides from 5′ to 3′, and the total number of insertions in each bin was calculated. Insertion frequencies in unselected RNA are shown on top in blue, and vRNA isolated from virus produced at 30°C is shown in red. The approximate location in the genome is indicated between the two datasets. Gray bars indicate some of the regions intolerant to insertions at 30°C.

Figure 4. Frequency of insertion sites found in vRNAs of virus propagated at 30°C or 40°C.

Transposon insertion frequencies were calculated for vRNAs isolated from 30°C and 40°C. Insertion frequencies for 50 nt bins are shown. 30°C frequencies are shown on top in blue, and 40°C frequencies are shown in red. Arrowheads indicate some of the regions in which more insertions were detected at 30°C than at 40°C.

Figure 5. Location of nsP3 temperature sensitive mutants.

The frequency of insertions at nt positions 4000–4250 at 30°C and 40°C is shown. The location of the ts mutants that were generated for the mouse challenge study are indicated with arrows. Mutants used in the study had either a single insertion at one of these locations (ts3-1 and ts3-3) or an insertion at both of these locations (double ts).

Figure 6. Virus replication at 30°C and 40°C.

Single-cycle growth curves of viruses predicted to be ts based on functional mapping. Vero cells were infected in duplicate at an MOI of 1, and incubated at either 30°C (A) or 40°C (B). Aliquots were removed at 0, 22 and 46 hr after infection and virus was measured by plaque assay on Vero cells. Samples in which virus was not detected are indicated with an asterisk (*).

Figure 7. Kaplan-Meier analysis of mouse survival after challenge.

Groups of 10 mice were innoculated with the VEEV ts strains and doses indicated along the right hand side of the graph, or with PBS. 28 days post innoculation, mice were challenged with 10⁴ pfu of wild-type VEEV strain Trinidad donkey and monitored for an additional 28 days. Mice were euthanized when moribund. All mice surviving to day 14 survived to day 28, when the study was terminated.