Cortical microglia dynamics are conserved during voluntary wheel running

Abstract

We present the first demonstration of chronic in vivo imaging of microglia in mice undergoing voluntary wheel running. We find that healthy mice undergoing voluntary wheel running have similar microglia dynamics, morphologies, and responses to injury when compared to sedentary mice. This suggests that exercise over a period of 1 mo does not grossly alter cortical microglial phenotypes and that exercise may exert its beneficial effects on the brain through other mechanisms. Future work examining how microglia dynamics may be altered during exercise in disease or injury models could provide further insights into the therapeutic benefit of exercise.NEW & NOTEWORTHY We demonstrate the first use of chronic in vivo imaging of microglia over time during physical exercise. We found that microglia movement, morphology, and process motility were remarkably stable during voluntary wheel running (VWR). Additionally, microglia in running mice respond similarly to laser ablation injury compared to sedentary mice. These findings indicate that VWR does not induce changes in microglia dynamics in healthy adults. Exercise may elicit positive effects on the brain through other mechanisms.

Keywords: microglia; two-photon microscopy; voluntary wheel running.

Graphical abstract

Figure 1.

Experimental design and running activity of male and female mice. A: experimental design and timeline of imaging in young adult CX3CR1^GFP/+ mice during 1 mo of voluntary wheel running (VWR) or sedentary (SED) conditions. Created with Biorender.com. B: weight of male and female mice over 1 mo of VWR or SED conditions normalized to the first measurement (baseline imaging session weight for that animal); a.u., arbitrary units. C: change in weight of male and female mice over 1 mo of VWR or SED conditions. VWR males show mild weight loss compared to SED males (P = 0.0413) and VWR males also lost more weight than VWR females (P = 0.0334). D: distance ran (km) by male and female mice in the dark phase over 1 mo of VWR. E: average running activity (km) per day of male and female VWR mice. Only nighttime running was included. F: percentage of day time running in male and female VWR mice. Data are presented as means ± SE with individual animals shown as single data points; n = 4–8 mice per group per sex.

Figure 2.

Microglia soma movement, clustering, and number during voluntary wheel running (VWR). A: representative images of microglia at baseline, week 1, week 2, week 3, and week 4. Scale bar = 100 µm. B: outlines overlayed on microglia somas from week 4 image. C: representative image of binarized microglia somas after thresholding. D–F: number of microglia somas normalized to baseline at week 1, week 2, week 3, and week 4 in sedentary (SED) and VWR mice of both sexes (D) and male (E) and female (F) mice separately. G–I: mean nearest neighbor distances of microglia somas at baseline, week 1, week 2, week 3, and week 4 of SED and VWR mice of both sexes (G) and male (H) and female (I) mice separately. G: when sexes were pooled there was a significant increase in the nearest neighbor distances between baseline imaging and week 1 in the VWR group (*P = 0.0471) that was not apparent in the SED group. J–L: to assay microglial movement, nearest neighbor distances of microglia somas were compared between subsequent imaging sessions in SED and VWR mice of both sexes (J) and male (K) and female (L) mice separately. M: frequency distribution showing the fraction of cells that moved a given distance between baseline and week 4 for male and female SED and VWR mice (bin size 2 µm). Total microglia analyzed: SED female = 1,952; SED male = 3,335; VWR female = 4,066; and VWR male = 3,746. Data are presented as means ± SE with individual animals shown as single data points; n = 4–6 mice per group per sex.

Figure 3.

Microglia soma morphology during voluntary wheel running (VWR). A: representations of soma area as a measure of size. B: representation of soma perimeter used to calculate circularity (4π × area/perimeter²). C: depiction of the major and minor axis of soma used to calculate aspect ratio. D–F: soma size over time in sedentary (SED) and VWR mice of both sexes (D; significant interaction between time and condition; F(4,76) = 2.495, P = 0.0498] and males (E) and females (F). G–I: circularity of somas over time was not different across conditions in mice of both sexes (G) and males (H) and females (I). J–L: the aspect ratio of somas was not significantly different across conditions in mice of both sexes (J) or when males (K) or females (L) were analyzed separately. Total microglia analyzed: SED female = 1,952; SED male = 3,335; VWR female = 4,066; and VWR male = 3,746. Data are presented as means ± SE; n = 4–6 mice per group per sex.

Figure 4.

Assessment of microglia coverage of the same cortical field at different time points during voluntary wheel running. A: representative images of microglia at different imaging time points Scale bar = 25 µm. B–D: percent area covered by microglia in voluntary wheel running (VWR) and sedentary (SED) mice of both sexes (B) and male (C) and female (D) mice. Data are presented as means ± SE; n = 4–7 mice per group per sex.

Figure 5.

Microglia morphology during voluntary wheel running (VWR). A: representative image of microglia. Scale bar = 25 µm B: representative image of binary microglia after thresholding. C: representation of microglia perimeter used to calculate circularity (4π × area/perimeter²). D: depiction of the major and minor axis of microglia used to calculate aspect ratio. E–G: size of microglial cells in sedentary (SED) and VWR mice of both sexes (E) and male (F) and female (G) mice. H–J: circularity of microglia in SED and VWR mice of both sexes (H) and male (I) and female (J) mice. K–M: aspect ratio of microglia in SED and VWR mice of both sexes (K) and male (L) and female (M) mice. L: a significant difference was observed between groups in male mice using a two-way repeated measures ANOVA [F(1,9) = 5.935, P = 0.0376]. However, Bonferroni’s post hoc multiple comparisons test showed no significant differences between groups. Data are presented as means ± SE; n = 4–6 mice per group per sex.

Figure 6.

Microglial complexity during voluntary wheel running (VWR). A: representative image of microglia. Scale bar = 25 µm B: representative image of binary microglia after thresholding. C: representation of Sholl analysis, whereby concentric circles are drawn at equal intervals from the cell body and intersections of processes with these circles are quantified. D–F: Sholl curves of microglia in VWR and SED mice of both sexes (D) and male (E) and female (F) mice. For clarity of presentation only baseline and week 4 curves are shown. G–L: Sholl curves were quantified for area under the Sholl curve (G–I) and maximum intersections (J–L) in VWR and SED mice. A significant difference was observed between groups using a mixed-effects analysis for the maximum number of intersections when sexes were pooled [F(1,19) = 4.847, P = 0.0403 (J)] and in male mice using a two-way repeated measures ANOVA [F(1,9) = 6.445, P = 0.0318 (K)]. However, Bonferroni’s post hoc multiple comparisons test showed no significant differences between groups. Data are presented as means ± SE; n = 4–6 mice per group per sex.

Figure 7.

Microglia process dynamics following 1 mo of voluntary wheel running (VWR). A and B: representative image of microglia at time = 0 min (T0) in magenta (A) and at time = 5 min (T5) in green (B). C: both time points are merged so that white represents pixels that are stable for both time points, while magenta represents retracted processes and green represents extended processes. Scale bar = 25 µm. D and E: microglia motility index for sedentary (SED) and VWR mice of both sexes (D) and when sexes are analyzed separately (E). F: there is no significant correlation between average distance run and microglial motility in male and female VWR mice. G and H: microglia surveillance for SED and VWR mice of both sexes (G) and when sexes are analyzed separately (H). I: there is no significant correlation between average distance run and microglial surveillance in male and female VWR mice. Data are presented as means ± SE; n = 4–7 mice per group per sex.

Figure 8.

Microglial response to laser injury following 1 mo of voluntary wheel running. A: sedentary (SED) and voluntary wheel running (VWR) images at 0 min, 5 min, 30 min, and 55 min following laser ablation. Scale bar = 25 µm. B and C: maximum directional velocity of microglial processes from 5 to 55 minutes in SED and VWR mice of both sexes (B) and male and female mice (C). D: quiver plot of microglial response. Green arrows represent vectors moving toward the ablation core and red arrows represent vectors moving away from the ablation core. Data are presented as means ± SE; n = 4–6 mice per group per sex.

Figure 9.

Open field assessment following 1 mo of voluntary wheel running (VWR). A: latency to leave the central square (seconds) of sedentary (SED) and VWR mice. B: time spent in central square (seconds) of SED and VWR mice. C: number of line crossings of SED and VWR mice. D: urinations and defecations of SED and VWR mice. E–H: latency to leave the central square (E), time spent in central square (F), number of line crossings (G), and urinations and defecations (H) separated by sex. Data are presented as means ± SE; n = 4–8 mice per group per sex.