Aging augments the impact of influenza respiratory tract infection on mobility impairments, muscle-localized inflammation, and muscle atrophy

Abstract

Although the influenza virus only infects the respiratory system, myalgias are commonly experienced during infection. In addition to a greater risk of hospitalization and death, older adults are more likely to develop disability following influenza infection; however, this relationship is understudied. We hypothesized that upon challenge with influenza, aging would be associated with functional impairments, as well as upregulation of skeletal muscle inflammatory and atrophy genes. Infected young and aged mice demonstrated decreased mobility and altered gait kinetics. These declines were more prominent in hind limbs and in aged mice. Skeletal muscle expression of genes involved in inflammation, as well as muscle atrophy and proteolysis, increased during influenza infection with an elevated and prolonged peak in aged mice. Infection also decreased expression of positive regulators of muscle mass and myogenesis components to a greater degree in aged mice. Gene expression correlated to influenza-induced body mass loss, although evidence did not support direct muscle infection. Overall, influenza leads to mobility impairments with induction of inflammatory and muscle degradation genes and downregulation of positive regulators of muscle. These effects are augmented and prolonged with aging, providing a molecular link between influenza infection, decreased resilience and increased risk of disability in the elderly.

Keywords: aging; disability; influenza; muscle atrophy; resilience.

Figure 1. Prolonged weight loss and elevated lung viral titers in aged mice during influenza infection

Young and aged C57BL/6 mice were intranasally infected with 500 EID₅₀ of PR8 influenza. (A) Weight loss was monitored throughout the infection and percent weight loss was calculated from day 0 prior to infection. Significant weight loss (compared to day 0) was observed day 6 through 15 (not indicated in figure) and differences between young and aged mice were observed at time points indicated (* = p<0.05). Data shown as mean ± SEM and analyzed via two-way ANOVA with Bonferroni post hoc corrections. (B) On day 0, 3, 7, 11, and 15 whole lung tissue was harvested and RNA was isolated. Total influenza PA copy number was determined via RT-qPCR. Significant viral burden was observed following flu infection (compared to day 0, p<0.05, indicated by brackets above data) and differences between young and aged mice at time points indicated (* = p<0.05). Data analyzed via two-way ANOVA with Bonferroni post hoc corrections. Data shown from one independent experiment with individual samples as dots, mean and SEM indicated by line and error bars, respectively.

Figure 2. Influenza infection induced functional decrements in voluntary locomotor activity and gait kinematics that is more pronounced in the hind limbs and in the aged mice

Young and aged C57BL/6 mice were intranasally infected with 500 EID₅₀ of PR8 influenza. On days 0, 3, 7, 11, 15, and 20 mice were tested for functional performance. (A) Spontaneous voluntary activity was assessed via the open field test on a photobeam activity system. Beam breaks were recorded as mice traveled at 16”x16” open field and locomotor activity was assessed as beam breaks per minute. Gait parameters were assessed utilizing DigiGait, a ventral plane videography treadmill system. Postural gait parameters (Stance width of the fore (B) and hind (E) limbs and midline distance of the fore (C) and hind (F) limbs) were altered during flu infection with more prominent differences in the hind limbs of aged mice. Kinematic gait parameters were also altered with flu infection. Gait symmetry of the fore/hind limbs (D) was increased. Stride length variability of the fore limb (G) did not change, however the aged mice had increased stride length variability in the hind limbs later in the infection (J). Maximal rate of change of paw area contact during the breaking phase (Max dA/dt) and propulsion phase (Min dA/dt) is altered in the fore (H and I, respectively) and hind limbs (K and L, respectively) with more dramatic results in the hind limbs. All data analyzed via two-way ANOVA with Bonferroni post hoc corrections with effect of flu infection over time (compared to day 0, p<0.05) indicated by brackets above data and differences between young and aged mice (p<0.05) at time points indicated by asterisk.

Figure 3. Influenza infection induced muscle-localized inflammatory gene expression in the gastrocnemius that is prolonged and elevated in aged mice

Young and aged C57BL/6 mice were intranasally infected with 500 EID₅₀ of PR8 influenza. At day 0, 3, 7, 11, and 15, mice fasted for 4-6 hours prior to sacrifice and gastrocnemius muscle was harvested and RNA was isolated. Gene expression was analyzed via RT-qPCR and normalized to reference genes and expression of young mice at day 0 to indicate fold changes. Influenza induced increased expression of IL6 (A) and IL6RA (B). Increased TNF (C) and CXCL10 (D) expression was observed in the aged mice. All data was log-transformed and analyzed via two-way ANOVA with Bonferroni post hoc corrections with effect of flu infection over time (compared to day 0, p<0.05) indicated by brackets above data and differences between young and aged mice (p<0.05) at time points indicated by asterisk.

Figure 4. Influenza infection induced gastrocnemius expression ubiquitin proteasome pathway components that is more dramatic in aged mice

Young and aged C57BL/6 mice were intranasally infected with 500 EID₅₀ of PR8 influenza. At day 0, 3, 7, 11, and 15, mice were fasted for 4-6 hours prior to sacrifice and gastrocnemius muscle was harvested and RNA was isolated. Gene expression was analyzed via RT-qPCR and normalized to reference genes and expression of young mice at day 0 to indicate fold changes. Influenza induced increased skeletal muscle expression of negative muscle regulators (Myostatin (MSTN, C) and Forkhead box protein O1 (FOXO1, D), as well as ubiquitin proteasome components (Atrogin1 (A), MuRF1 (B), Ubiquitin B (UBB, E), and Ubiquitin C (UBC, F). All data was log-transformed and analyzed via two-way ANOVA with Bonferroni post hoc corrections with effect of flu infection over time (compared to day 0, p<0.05) indicated by brackets above data and differences between young and aged mice (p<0.05) at time points indicated by asterisk.

Figure 5. Influenza infection reduced gastrocnemius expression of positive regulators of muscle mass and myogenic regulatory factors to a greater degree in aged mice

Young and aged C57BL/6 mice were intranasally infected with 500 EID₅₀ of PR8 influenza. At day 0, 3, 7, 11, and 15, mice were fasted for 4-6 hours prior to sacrifice and gastrocnemius muscle was harvested and RNA was isolated. Gene expression was analyzed via RT-qPCR and normalized to reference genes and expression of young mice at day 0 to indicate fold changes. Influenza reduced skeletal muscle expression of insulin-like growth factor 1 (IGF1, A), myocyte enhancer binding factor 2C (MEF2C, B), paired box protein 7 (PAX7, C), myogenic differentiation 1 (MYOD1, D), and myogenin (MYOG, E). All data was log-transformed and analyzed via two-way ANOVA with Bonferroni post hoc corrections with effect of flu infection over time (compared to day 0, p<0.05) indicated by brackets above data and differences between young and aged mice (p<0.05) at time points indicated by asterisk.

Figure 6. Influenza-induced weight loss correlated with gastrocnemius gene expression of ubiquitin proteasome pathway components

Percent body mass loss at time of sacrifice and corresponding gastrocnemius gene expression was analyzed via univariate linear regression for all genes that showed significant time effects. Young and aged mice were analyzed separately to determine if relationships vary with age. FOXO1 (A), IL6RA (B), UBB (C), UBC (D), MURF1 (E), ATROGIN1 (F), and IGF1 (G) were significantly correlated with percent body mass in either young or aged, or both (Young (Y) and aged (A) mice regression analysis p and R² values indicated to right of graph, bolded if significant (p<0.05)), while no relationship was seen with percent body mass and expression of IL6, TNF, CXCL10, MEF2C, PAX7, MYOD1, and MYOG (data not shown).

Figure 7. The pathogenesis of influenza-induced myopathies is likely not direct infection of skeletal muscle in vivo as viral copies are not seen in the gastrocnemius muscle

Young and aged C57BL/6 mice were intranasally infected with 500 EID₅₀ of PR8 influenza. (A) On day 0, 3, 7, 11, and 15 whole gastrocnemius was harvested and RNA was isolated. Total number of copies of influenza PA was determined via RT-qPCR with a positive control used (infected mice lung tissue). Uninfected young and aged murine myoblasts were incubated with 0, 10, or 100 EID₅₀ PR8 influenza for one hour and then were cultured in growth media. Myoblast supernatant was analyzed for chemokine/cytokines via multiplex assay. Detectable cytokines were analyzed by 3-way ANOVA (age x infection condition x time point). While a significant age effect was observed (p<0.05), there was no effect of infection and no interaction of infection and time for myoblast secretion of IL6 (C), CXCL10 (D), CXCL1 (E), and CCL2 (F). At 96 hr post infection, total RNA from the myoblast culture was extracted and total number of copies of influenza PA was determined via RT-qPCR. No viral copies were present in in vitro myoblast cultures (B).