High-Intensity Interval Training, Caloric Restriction, or Their Combination Have Beneficial Effects on Metabolically Acquired Peripheral Neuropathy

Abstract

Peripheral neuropathy (PN) is a prevalent and debilitating complication of obesity, prediabetes, and type 2 diabetes, which remains poorly understood and lacks disease-modifying therapies. Fortunately, diet and/or exercise have emerged as effective treatment strategies for PN. Here, we examined the impact of caloric restriction (CR) and high-intensity interval training (HIIT) interventions, alone or combined (HIIT-CR), on metabolic and PN outcomes in high-fat diet (HFD) mice. HFD feeding alone resulted in obesity, impaired glucose tolerance, and PN. Peripheral nerves isolated from these mice also developed insulin resistance (IR). CR and HIIT-CR, but not HIIT alone, improved HFD-induced metabolic dysfunction. However, all interventions improved PN to similar extents. When examining the underlying neuroprotective mechanisms in whole nerves, we found that CR and HIIT-CR activate the fuel-sensing enzyme AMPK. We then performed complimentary in vitro work in Schwann cells, the glia of peripheral nerves. Treating primary Schwann cells with the saturated fatty acid palmitate to mimic prediabetic conditions caused IR, which was reversed by the AMPK activator, AICAR. Together, these results enhance our understanding of PN pathogenesis, the differential mechanisms by which diet and exercise may improve PN, and Schwann cell-specific contributions to nerve insulin signaling and PN progression.

Figure 1

Study design and metabolic phenotyping for cohorts 1 and 2. A: For cohort 1, C57BL/6J mice (n = 10 mice/group) were fed an SD or 60% HFD starting at 23 weeks of age. SCN and DRG isolation was performed 4 days after the start of the HFD. B: For cohort 2, C57BL/6J mice (n = 22 mice/group) were fed an SD or 60% HFD starting at 6 weeks until 23 weeks of age. Metabolic and neuropathy phenotyping and SCN and DRG isolation were performed at 23 weeks of age. C: Terminal body weights were measured in SD- or HFD-fed mice at 23 weeks of age. Data are mean ± SD, with each data point representing an individual animal. Two-tailed Student t test, ****P < 0.0001, HFD vs. SD. D: Glucose tolerance tests (GTT) were performed at 23 weeks of age. Data are mean ± SD. Two-way ANOVA test, ****P < 0.0001, HFD vs. SD.

Figure 2

Intraepidermal nerve fiber loss in HFD-fed mice. Representative images of intraepidermal nerve fibers in SD (A) and HFD (B) mice at 23 weeks of age. Individual fibers are indicated with white dots. Scale bar: 100 µm. C: Quantification of IENFD in SD and HFD mice at 23 weeks of age as described in the Research Design and Methods. Data are mean ± SD, with each data point representing an individual animal (n = 6 mice/group). Two-tailed Student t test, **P < 0.01, HFD vs. SD.

Figure 3

IR develops in isolated peripheral nerves after acute and chronic HFD feeding. SCN and DRG were isolated from SD (n = 10 cohort 1, 4 days on diet; n = 22 cohort 2, 17 weeks on diet) and HFD (n = 10 cohort 1, 4 days on diet; n = 22 cohort 2, 17 weeks on diet) mice. The SCN (A and C) and DRG (B and D) tissues harvested from one side of the mice were treated with 100 nmol/L insulin (+), whereas tissues from the other side of the same animal were treated with an equivalent volume of media (−, control). Representative Western blots of SCN (A and C) and DRG (B and D) from SD and HFD mice. Densitometry analysis of Akt phosphorylation (pAkt) at serine 473 (pS473) and ERK1/2 phosphorylation (pERK1/2) at threonine 202 (pT202) and tyrosine 204 (pY204) in SCN (A and C) and DRG (B and D) of SD and HFD mice. Protein expression normalized to tubulin. Data are mean ± SD with each data point representing one SCN or DRG from an individual animal. Two-tailed Student t test, *P < 0.05, **P < 0.01, ***P < 0.001, HFD vs. SD; ns, not significant.

Figure 4

Study design and HIIT treadmill exercise paradigm for cohort 2. A: C57BL/6J mice were fed an SD (n = 8) or 60% HFD (n = 8) starting at 6 weeks until 26 weeks of age. Three cohorts of HFD mice were fed the HFD until 18 weeks of age and then placed under CR (n = 8), on HIIT exercise regimen (n = 8), or under CR-HIIT exercise regimen (n = 8) for the remaining 7 (HIIT and HIIT-CR) or 8 (CR) weeks. Baseline metabolic and neuropathy phenotyping and VO_2max determination were performed at 15 weeks of age. Terminal metabolic and neuropathy phenotyping and SCN isolation were performed at 25 weeks (HIIT and HIIT-CR) and 26 weeks (SD, HFD, and CR) of age. B: HIIT treadmill exercise paradigm. Before training, animals were allowed 5 min of rest on the treadmill. Training was composed of four intervals. Each interval consisted of 3 min of low-intensity exercise, followed by 4 min of high-intensity exercise. Mice completed a final 3 min of low-intensity exercise after the last interval. Low-intensity exercise was defined as 40% VO_2max and high intensity exercise as 70–75% VO_2max.

Figure 5

CR or CR-HIIT after 12 weeks of HFD improves metabolic health. Baseline (15 weeks of age) and terminal (25–26 weeks of age) body weight (A), percentage of fat mass (B), percentage of lean mass (C), terminal gastrocnemius muscle (gastroc) weight relative to body weight (D), absolute gastroc weight (E), and glucose tolerance testing (GTT) (F–H) were measured in SD, HFD, CR, HIIT, and HIIT-CR mice. AUC, area under the curve. G: For terminal GTT, *represents significance of HFD vs. SD and #represents significance of HIIT vs. SD. Data are represented as least square means ± SD, with each data point representing an individual animal (n = 8 mice/group). PROC MIXED with animal ID set as random effect and time point (baseline vs. terminal) set as repeated measures for longitudinal data.*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Figure 6

CR, HIIT regimen, or HIIT-CR regimen after 12 weeks of HFD improves nerve function. Baseline (15 weeks of age) and terminal (25–26 weeks of age) sural sensory (A) and sciatic motor nerve conduction velocities (NCVs) (B) and IENFDs (C) were measured in SD, HFD, CR, HIIT, and HIIT-CR mice. Data are represented as least square means ± SD, with each data point representing an individual animal (n = 8 mice/group). PROC MIXED with animal ID set as random effect and time point (baseline vs. terminal) set as repeated measures for longitudinal data. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Figure 7

CR or CR-HIIT regimen after 12 weeks of HFD promotes AMPK activation in SCN. A: Representative Western blots of SCN from SD, HFD, CR, HIIT, and HIIT-CR mice. B: Densitometry analysis of AMPK phosphorylation (pAMPK) at threonine 172 (pT172) in SCN of SD, HFD, CR, HIIT, and HIIT-CR mice. Protein expression normalized to tubulin and relative to control (SD). Data are least square means ± SD, with each data point representing one SCN from an individual animal (SD, HFD, CR, HIIT-CR, n = 7 mice/group; HIIT, n = 6 mice/group). PROC MIXED with animal ID set as random effect. **P < 0.01, ***P < 0.001.

Figure 8

AMPK activation prevents palmitate-induced insulin resistance in rSCs. A: Treatment timeline. rSCs were treated with BSA or 250 µmol/L palmitate for 48 h. Media was switched to fresh media containing 1 mmol/L AICAR for 1 h and then stimulated with 20 nmol/L insulin for 30 min. B: Representative Western blots. C: Densitometry analysis of Akt phosphorylation (pAkt) at serine (S473) and ERK1/2 phosphorylation (pERK1/2) at threonine 202 (pT202) and tyrosine 204 (pY204) in rSCs treated with palmitate and AICAR and stimulated with insulin. Protein expression normalized to tubulin. Data are mean ± SD from three independent cell culture experiments with duplicate treatments in each experiment. One-way ANOVA with the Tukey post hoc test for multiple comparisons, ***P < 0.001, ****P < 0.0001.