Effects of Voluntary Wheel Running Exercise on Chemotherapy-Impaired Cognitive and Motor Performance in Mice

Abstract

Chemotherapy-induced cognitive impairment (chemobrain) and muscle wasting (cachexia) are persisting side effects which adversely affect the quality of life of cancer survivors. We therefore investigated the efficacy of physical exercise as a non-pharmacological intervention to reverse the adverse effects of chemotherapy. We examined whether physical exercise in terms of voluntary wheel running could prevent chemotherapy-induced cognitive and motor impairments in mice treated with the multi-kinase inhibitor sorafenib. Adult male BALB/c mice were subdivided into runner and non-runner groups and orally administered with sorafenib (60 mg/kg) or vehicle continuously for four weeks. Mice could freely access the running wheel anytime during sorafenib or vehicle treatment. We found that sorafenib treatment reduced body weight gain (% of change, vehicle: 3.28 ± 3.29, sorafenib: -9.24 ± 1.52, p = 0.0004), impaired hippocampal-dependent spatial memory in the Y maze (exploration index, vehicle: 35.57 ± 11.38%, sorafenib: -29.62 ± 7.90%, p < 0.0001), increased anhedonia-like behaviour in the sucrose preference test (sucrose preference, vehicle: 66.57 ± 3.52%, sorafenib: 44.54 ± 4.25%, p = 0.0005) and impaired motor skill acquisition in rotarod test (latency to fall on day 1: 37.87 ± 8.05 and day 2: 37.22 ± 12.26 s, p > 0.05) but did not induce muscle wasting or reduce grip strength. Concomitant voluntary running reduced anhedonia-like behaviour (sucrose preference, sedentary: 44.54 ± 4.25%, runners: 59.33 ± 4.02%, p = 0.0357), restored impairment in motor skill acquisition (latency to fall on day 1: 50.85 ± 15.45 and day 2: 168.50 ± 37.08 s, p = 0.0004), but failed to rescue spatial memory deficit. Immunostaining results revealed that sorafenib treatment did not affect the number of proliferating cells and immature neurons in the hippocampal dentate gyrus (DG), whereas running significantly increased cell proliferation in both vehicle- (total Ki-67⁺ cells, sedentary: 16,687.34 ± 72.63, exercise: 3320.03 ± 182.57, p < 0.0001) and sorafenib-treated mice (Ki-67⁺ cells in the ventral DG, sedentary: 688.82.34 ± 38.16, exercise: 979.53 ± 73.88, p < 0.0400). Our results suggest that spatial memory impairment and anhedonia-like behaviour precede the presence of muscle wasting, and these behavioural deficits are independent of the changes in adult hippocampal neurogenesis. Running effectively prevents body weight loss, improves motor skill acquisition and reduces anhedonia-like behaviour associated with increased proliferating cells and immature neurons in DG. Taken together, they support physical exercise rehabilitation as an effective strategy to prevent chemotherapy side effects in terms of mood dysregulation and motor deficit.

Keywords: adult hippocampal neurogenesis; chemotherapy; depression; learning and memory; muscle wasting; physical exercise.

Figure 1

Voluntary running exercise attenuated sorafenib-induced weight loss. (A) Mice were housed with (runners) or without (non-runners) running wheels and were treated with sorafenib (60 mg/kg, oral gavage) or vehicle for 28 days continuously. They were then assessed by sucrose preference test (SPT), Y-maze test (YMT), open field test (OFT), rotarod test (RR), and forced swim test (FST). Fixed tissues were collected for histological analyses at the experimental endpoint. (B) In the second cohort, mice were submitted to muscle strength tests, including the inverted grip test (IGT) and grip strength test (GST) after the same treatment. Fresh skeletal muscles were collected at the experimental endpoint. (C) Sorafenib administration significantly reduced body weight. Concomitant voluntary exercise promoted body weight gain in vehicle-treated runners and restored body weight loss in sorafenib-treated runners (Tukey’s post-hoc test: ** p < 0.005 vs. Vehicle + Non-runners & Vehicle + Runners; ^## p < 0.005 vs. Vehicle + Non-runners). Results were expressed as mean ± SEM. n = 8–11 animals per group. (D) The average running distance was significantly reduced in sorafenib-treated runners (unpaired t-test: ** p < 0.005). * p < 0.05.

Figure 2

Sorafenib administration impaired spatial memory and induced anhedonia-like behavior. (A) Sorafenib administration impaired spatial working memory in non-runners (Tukey’s post hoc: ^## p < 0.005 vs. Vehicle + Non-runners, * p < 0.05). Running failed to improve cognitive impairment in sorafenib-treated mice since they did not show any preferences for novel over familiar arms. Voluntary running could not restore sorafenib-induced cognitive impairment (Tukey’s post hoc: ^# p < 0.05 vs. Vehicle + Non-runners). (B) Sorafenib administration induced anhedonia-like behaviour as showed with a significant reduction in sucrose preference (Tukey’s post hoc: ^## p < 0.005 vs. Vehicle + Non-runners), whereas voluntary running showed the opposite effects (Tukey’s post hoc: * p < 0.05 vs. Sorafenib + Non-runners). (C) Sorafenib administration did not affect the immobility time spent in FST, while voluntary running reduced immobility time (Tukey’s post hoc: * p < 0.05 vs. Non-runners). Results were expressed as mean ± SEM. n = 7–12 animals per group.

Figure 3

Exercise ameliorated sorafenib-impaired motor learning. (A) Vehicle-treated and non-runners showed improved motor learning on the second day on the constant-speed rotarod (Tukey’s post-hoc test: ** p < 0.005 vs. Day 1: Sorafenib + Runners & Non-runners) ^## p < 0.005. Sorafenib administration impaired rotarod learning (Tukey’s post-hoc test: ^§§ p < 0.005 vs. Day 2: Vehicle + Non-runners), which could be restored by concomitant voluntary wheel running exercise (Tukey’s post-hoc test: * p < 0.05 vs. Day 1: Sorafenib + Runners). (B) Voluntary wheel running exercise improved motor coordination of vehicle-treated runners on the accelerated rotarod with increased latency to fall (Tukey’s post-hoc test: * p < 0.05 vs. Vehicle + Non-runners). Exercise-associated motor coordination improvements were hindered by sorafenib administration in sorafenib-treated runners (Tukey’s post-hoc test: ^# p < 0.05 vs. Sorafenib + Runners). (C,D) There was a significant reduction in distance travelled in (C) sorafenib-treated runners during the first 5 min (Tukey’s post-hoc test: # p < 0.05 vs. Vehicle-treated + Non-runners) and (D) in both sorafenib-treated runners and non-runners during the total 15 min in the open field (Tukey’s post-hoc test: # p < 0.05 vs. Vehicle-treated + Runners & Vehicle-treated + Non-runners). (E) Neither voluntary wheel running exercise nor sorafenib administration affected mean speed. Results were expressed as mean ± SEM. n = 7–12 per group.

Figure 4

Voluntary running failed to increase cell proliferation in DG of sorafenib-treated mice. (A) Voluntary running exercise increased proliferating progenitors in the ventral DG of vehicle-treated mice (Tukey’s post hoc: ** p < 0.005 vs. Vehicle + Non-runners). Sorafenib did not affect the proliferating progenitor number. Although voluntary exercise increased the proliferating cell counts in sorafenib-treated mice (Tukey’s post hoc: * p < 0.05 vs. Sorafenib + Non-runners), the exercise-induced proliferation was significantly lower (Tukey’s post hoc: ^## p < 0.005 vs. Sorafenib + Non-runners) in the ventral DG. (B,C) Similarly, concomitant voluntary exercise promoted cell proliferation in the (B) dorsal and (C) whole DG in vehicle-treated runners (Tukey’s post hoc: ** p < 0.005 vs. Vehicle + Non-runners), but not in sorafenib treated mice (Tukey’s post hoc: ^## p < 0.005 vs. Vehicle + Runners). (D,E) Representative images of Ki-67 positive cells in the (D) ventral and (E) dorsal DG (Scale bars, 100 μm in 200×). Results were expressed as mean ± SEM. n = 6 per group.

Figure 5

Voluntary running promoted immature neuron numbers in DG, but sorafenib showed no effect. (A–C) Sorafenib administration did not affect immature neuron number in DG, while runners increased immature neurons in the (A) ventral, (B) dorsal, and (C) whole DG of sorafenib-treated and vehicle-treated mice (Tukey’s post hoc: ** p < 0.005 vs. Non-runners). (D,E) Representative images of DCX positive cells in the (D) ventral and (E) dorsal DG (Scale bars, 100 μm in 200×). Results were expressed as mean ± SEM. n = 6 per group.