Experiment
BXD advanced recombinant inbred lines of were used here to identify novel gene variants that impact the severity of several herpes simplex virus type 1 (HSV-1) disease phenotypes including the percent mortality (viral invasion of the central nervous system resulting in death) and herpetic stromal keratitis, in a well characterized mouse model of HSV-1 corneal infection.
The lines of the BXD RI panel result from a cross of C57BL/6J (B6) and DBA/2J (D2) inbred strains of mice that are highly resistant and highly susceptible to HSV infection respectively. Both parental strains have been completely sequenced and the BXD lines have been densely genotyped.
Here BXD lines and the parental strains were infected with HSV-1 strain 17syn+ via the cornea and systematically phenotyped for >30 days post-infection. Infected mice were observed daily, weighed every other day, and tear films were collected on day 4 post infection to evaluate viral titers. Percent mortality was scored from 48 hours through 21 days post infection. Eyes were independently scored for severity of herpetic stromal keratitis. Thirty-two fully inbred BXD strains were phenotyped. The parental strains and most BXD lines were analyzed more than once throughout the multi-year study. In all cases the reproducibility of the disease phenotypes displayed within the individual lines was consistent. In contrast, extensive variability in the phenotypes was displayed between lines.
QTL mapping was performed using the GeneNetwork analysis tools (www.genenetwork.org; NCBI Build 37). A likelihood ratio statistic (LRS) was calculated at each marker. Genome-wide significance was determined by performing 2000 permutations. The LRS is mathematically related to the log of odds (LOD) ratio that is employed in genetic analyses and can be converted to LOD by dividing the LRS score by 4.61.
Weight loss was considered an effective measure of infectious disease severity.
A significant QTL, Hsv1sw1 (herpes simplex virus type 1 susceptibility, weight loss 1), for percent maximum weight loss was detected in both male and female mice on Chromosome 12, peak LRS 22 at ~114.5 Mb. A second QTL, Hsv1sw2 (herpes simplex virus type 1 susceptibility, weight loss 2), was detected on Chromosome 16, peak LRS 21, at ~89.4 Mb in female mice only. B6 alleles influence resistance with an additive effect at both loci. [Figure 2.] The results of pair-wise scans did not suggest any interactions between the Chr 12 and Chr 16 loci. Genes of interest on the distal end of Chr 12 (106-119 Mb) are Igh, Yy1 and Rcor1.
A significant QTL, Hsv1sm1 (herpes simplex virus type 1 susceptibility, mortality 1), associated with percent mortality was detected on the distal portion of Chr 16, LRS=29 at 86.2 Mb. Infected mice of susceptible strains showed signs of central nervous system disease including hunched posture, roughened fur, convulsions and/or ataxia, indicating that death was the result of herpetic encephalitis. Several suggestive QTL were found on Chrs 4, 11, 12 and X. A pair-wise scan revealed possible interaction between the Chr 16 locus and the suggestive locus on Chr X. When controlling for SNP rs4213268 on Chr 16 a significant locus, Hsv1sm2 (herpes simplex virus type 1 susceptibility, mortality 2) was detected on the proximal part of Chr X with a peak LRS of 24. B6 alleles influence resistance additively at both loci. [Fig 3a.3b.]
A significant QTL for severity of herpetic stromal keratitis, Hsks1 (herpetic stroma keratitis severity 1) was localized to the same region of the distal arm of Chr 16 as QTL Hsv1sm1 at 86.45 Mb. The peak LRS for HSK mapped to 86.16 Mb. B6 alleles at this locus influenced resistance. [Fig 4.] Suggestive QTL were also observed on Chrs 3, 8, 11, 12 and 18. Controlling for the QTL on Chr 16 did not reveal any additional significant QTL, nor did pair scans detect any interactions. Grik1 is a top candidate gene for regulation of percent mortality (neuroinvasion) and severity of stromal keratitis.
Percent mortality was significantly correlated to both weight loss and HSK severity, with the former being the stronger association.
Analysis Tools