Age and Sex of Mice Markedly Affect Survival Times Associated with Hyperoxic Acute Lung Injury

Abstract

Mortality associated with acute lung injury (ALI) remains substantial, with recent estimates of 35-45% similar to those obtained decades ago. Although evidence for sex-related differences in ALI mortality remains equivocal, death rates differ markedly for age, with more than 3-fold increased mortality in older versus younger patients. Strains of mice also show large differences in ALI mortality. To tease out genetic factors affecting mortality, we established a mouse model of differential hyperoxic ALI (HALI) survival. Separate genetic analyses of backcross and F2 populations generated from sensitive C57BL/6J (B) and resistant 129X1/SvJ (X1) progenitor strains identified two quantitative trait loci (QTLs; Shali1 and Shali2) with strong, equal but opposite, within-strain effects on survival. Congenic lines confirmed these opposing QTL effects, but also retained the low penetrance seen in the 6-12 week X1 control strain. Sorting mice into distinct age groups revealed that 'age at exposure' inversely correlated with survival time and explained reduced penetrance of the resistance trait. While B mice were already sensitive by 6 weeks old, X1 mice maintained significant resistance up to 3-4 weeks longer. Reanalysis of F2 data gave analogous age-related findings, and also supported sex-specific linkage for Shali1 and Shali2. Importantly, we have demonstrated in congenic mice that these age effects on survival correspond with B alleles for Shali1 (6-week old mice more sensitive) and Shali2 (10-week old mice more resistant) placed on the X1 background. Further studies revealed significant sex-specific survival differences in subcongenics for both QTLs. Accounting for age and sex markedly improved penetrance of both QTLs, thereby reducing trait variability, refining Shali1 to <8.5Mb, and supporting several sub-QTLs within the Shali2 interval. Together, these congenics will allow age- and sex-specific studies to interrogate myriad subphenotypes affected during ALI development and progression and identify intermediary injury biomarkers that can predict outcome.

Fig 1. Survival time comparisons of X1 and B controls stratified by age and sex.

Survival times for all exposed mice (6–12 weeks old) of each progenitor strain were combined and sorted by sex into three distinct age groups: ≤ 6 weeks old, 7–9 weeks old, and ≥ 10 weeks old at the initiation of exposure. Resistant X1 control mice (solid bars) are displayed at the left half of the figure, with sensitive B control mice (striped bars) shown at the right half. The number of mice tested in each group is given at the base of each bar plot. For direct within-strain sex comparisons, females (pink bars) and males (blue bars) are positioned side-by-side. Comparisons were also made between age groups (females or males) and between strains for the same age group, with Bonferroni corrections for multiple testing. *, p < 0.05; **, p < 0.001.

Fig 2. Survival time comparisons of F₂ mice stratified by age and sex.

Survival times for all exposed F₂ mice (n = 840) generated from B and X1 progenitors, were sorted by sex into the three age groups: ≤ 6 weeks old, 7–9 weeks old, and ≥10 weeks old at the initiation of exposure. The number of F₂ mice represented in each group is given at the base of each bar plot. Comparisons were made between age groups for females or males and between sexes for the same age group, and Bonferroni corrected for multiple testing. *, p < 0.05; **, p < 0.001.

Fig 3. QTL analysis of age and sex cohorts of the total (X1xB) F₂ dataset.

The total F₂ data set was sorted using a 56 d.o. age threshold for younger versus older cohorts and QTL analysis performed for total mice (black lines), and separately for males (blue lines) and females (red lines) in each age group. A. Younger mice (≤ 56 d.o.; n = 546) identified Shali1 and Shali2, with significant differences between males (blue arrow) and females (red arrows) for each QTL. B. Older mice (> 56 d.o.; n = 294) did not identify Shali1 or Shali2, but instead identified the male-specific Shali5 on distal Chr1 (blue arrow).

Fig 4. Age and sex-specific survival differences of F₂ mice carrying reciprocal alleles for Shali1 and Shali2.

The (X1xB) F₂ dataset (n = 840) was sorted based on resistance (X1-B) or sensitivity (B-X1) alleles (i.e., genotype effect) at both of the peak markers representing Shali1 (D1Mit303) and Shali2 (D4Mit308). Mice with the most resistant (X1-B; n = 50) and most sensitive (B-X1; n = 58) allelic combinations at these two loci were then segregated by age (using a 56 d.o. threshold for younger vs. older mice) and sex, and compared at each stage. Numbers near the top of each bar represent mean survival times (hours) and SDs rounded to the nearest hour, and those at the bottom of bars are numbers of mice with the corresponding variables (genotype/age/sex). Significant differences are as indicated, using the t-test for group comparisons and Bonferroni corrections for multiple testing.

Fig 5. Sex-specific survival time comparisons of ≤ 56 d.o. Shali1 congenic lines.

Mice aged 56 d.o. and younger carrying different intervals of Shali1 (boxed region on chromosome at the left) were exposed to >95% O₂ and survival times determined and compared with age- and sex-matched X1 controls. A significant increase in sensitivity was determined in both males and females for four of six interval-specific congenics (bold red font), which narrowed the Shali1 interval to ~8.5 Mb (as depicted by the dashed line and green box at 60.05–68.55 Mb). In all cases, females were also significantly more sensitive than males.

Fig 6. Sex-specific survival time comparisons of > 56 d.o. Shali2 congenic lines.

Mice > 56 d.o. carrying different subintervals of Shali2 (boxed region on chromosome at the left) were exposed to > 95% O₂ and survival times determined and compared with age- and sex-matched X1 controls. A significant increase in resistance was found in males only for four congenics and for both males and females for three congenics (bold blue font). One congenic (4BB.ab) was significantly more sensitive than controls for males (bold red font), with a trend for increased sensitivity in females. Congenics support multiple sub-QTLs in the Shali2 interval.