Journal Logo

BASIC SCIENCES

Impact of Exercise and Activity on Weight Regain and Musculoskeletal Health Post-Ovariectomy

SHERK, VANESSA D.1; JACKMAN, MATTHEW R.1; HIGGINS, JANINE A.2; GILES, ERIN D.1,3; FORIGHT, REBECCA M.1; PRESBY, DAVID M.1; CARPENTER, R. DANA4; JOHNSON, GINGER C.1; OLJIRA, ROBERA1; HOUCK, JULIE A.1; MACLEAN, PAUL S.1

Author Information
Medicine & Science in Sports & Exercise: December 2019 - Volume 51 - Issue 12 - p 2465-2473
doi: 10.1249/MSS.0000000000002082

Abstract

The high prevalence of obesity is accompanied by a high incidence of individuals who attempt to lose weight. Most weight loss strategies are only transiently effective, and a strong biological drive to regain weight underlies the high recidivism rates with obesity (1,2). Adaptations to weight loss create a homeostatic system that encourages weight regain, such that maintaining a given weight is more difficult in those who were previously at a heavier weight compared with those individuals who have not lost weight. Some adaptations to weight loss include reductions in resting metabolic rate relative to body size and a reduced ability to modify substrate utilization in response to metabolic stress (3,4).

A particularly vulnerable time for weight loss maintenance in women is during menopause, and it is very common for women to enter the menopausal transition with preexisting obesity and a history of weight loss. Further, the effects of obesity and weight loss on bone health are both controversial, but could result in women entering menopause with suboptimal bone quality. The transition through menopause is associated with weight gain, increased visceral adiposity, decreased bone mass, and a general decline in metabolic health (5,6). In young women, the acute suppression of ovarian function results in a small, but significant, decline in total and resting energy expenditure (REE) (7). When food intake is not reduced to a sufficient extent, the decreased energy expenditure (EE) results in a positive energy imbalance and subsequent weight gain. We previously demonstrated that the combination of a history of weight loss and the loss of ovarian function have independent influences on weight gain in Wistar rats, such that there was a decrease in spontaneous physical activity (SPA) and an increase in food intake after loss of ovarian function (8,9). These bioenergetic changes may exacerbate existing weight problems in overweight women and lead to the onset of obesity in lean women after menopause.

Exercise is often used to assist in weight control. It is unknown, however, how well exercise interventions affect food intake and counter weight regain when the loss of ovarian function occurs during weight loss maintenance. Weight loss maintenance trials in humans have failed to demonstrate the efficacy of exercise for preventing weight regain, partly because maintaining a high exercise compliance and adherence is more difficult in clinical trials than in preclinical trials (10,11). However, findings from the National Weight Control Registry consistently show that those who are successful at weight loss maintenance exercise for more than 1 h·d−1 (12). Moreover, exercise reduces appetite and alters nutrient trafficking in male rats, which results in an attenuated drive to regain weight (13,14). We found that exercise does not attenuate weight regain in ovary-intact female rodents, but there is evidence that exercise attenuates weight regain in ovariectomized (OVX) rats that lost weight after OVX as previously reviewed (11,15). This dependency on sex hormones for the effect of exercise on weight regain highlights the need to understand the role of exercise in preventing weight regain when endogenous sex hormones change.

In humans, controlling for the timing of the onset of obesity, lifetime physical activity habits and history of weight loss attempts, while also separating the effect of aging from the effect of ovarian hormone loss due to the menopausal transition, is extremely difficult. Therefore, we recreated this phenotypic milieu in rodents to gain novel insight into how exercise can counteract the energetic challenges of menopause. The purpose of this study was to determine whether exercise training affects the ability to maintain weight loss after OVX in those with preexisting obesity. We hypothesized that regular exercise would counteract the biological drive to regain weight by preserving insulin sensitivity, mitochondrial biogenesis, bone mass, and fat oxidation in muscle, bone, and liver, and the resulting EE that might otherwise decrease with the removal of ovarian hormones.

METHODS

All procedures were approved by the University of Colorado Anschutz Medical Campus Institutional Animal Care and Use Committee. The study timeline with weight trajectories is shown in Figure 1A. Five-week-old female Wistar rats (Charles River Laboratories, Wilmington, MA) were individually housed on a 14:10 h light–dark cycle and provided free access to water. Obesity-prone (OP) (upper tertile; n = 22) and obesity-resistant (OR) (lower tertile; n = 22) rats were identified after 6 wk of high-fat diet (HFD) (46% kcal fat; Research Diets D12344, New Brunswick, NJ) feeding based on weight gain and percentage body fat responses, and allowed to continue HFD feeding until 17 to 18 wk of age, as previously described (Fig. 1A, B) (3,16,17). Rats in the middle tertile were removed from the study. At 17 to 18 wk of age, rats were further assigned within each tertile into sedentary (SED) or exercising (EX), balanced for weight and body fat percentage. All groups were switched to a medium-fat diet (MFD) (25% kcal fat, Research Diets D07091301; see Table, Supplemental Digital Content 1, Dietary formula of MFD provided by Research Diets Inc., http://links.lww.com/MSS/B670) and energy restricted to approximately 55% of the calories eaten ad libitum (AL) to induce a 15% weight loss over 2 wk, and then kept in energy balance (EB) at the reduced weight for 8 wk (Fig. 1B). Energy balance was maintained by providing a limited portion of MFD just before the start of the dark cycle.

FIGURE 1
FIGURE 1:
OP and OR phenotypes were revealed with 13 wk of HFD. Rats were switched to MFD for the weight loss, WLM, and AL refeeding phases. The solid black line is a general weight trajectory. C, Calorimetry data collection before OVX, after recovery from OVX but before AL feeding began (rats were maintained in EB), during the first week of post-OVX AL feeding, and just before sac (A). Obesity development and weight loss (B) before OVX, and weight regain (C). The rate of regain was highest in week 1 of refeeding (C, D). Fat mass (E) and lean mass (F) before and after weight loss and regain. †P < 0.05, phenotype–activity interaction; *P < 0.01, main effect of phenotype (OP > OR). WLM, weight loss maintenance.

At the end of the 8-wk caloric restriction/maintenance period, rats were surgically ovariectomized (OVX) using dorsal entry under isoflurane anesthesia (18). After OVX, we aimed to observe changes in EE and SPA, independent from refeeding. To do this, rats were provided a calorically limited (CL) amount of MFD during the 6 to 10 d of recovery after OVX, the last 3 d of which were in the metabolic monitoring system (described below). The CL provisions were based on anticipated calories needed to maintain EB, which was initially the same as their pre-OVX intake. Adjustments to the food allotment were made if a 2- to 3-d trend of weight loss or gain occurred. After 3 d of metabolic monitoring in the CL phase, animals were allowed to eat MFD AL for the remainder of the study. The EX rats continued with their exercise program. Rats were euthanized by exsanguination under anesthesia once their weight plateaued, which occurred 9 to 11 wk post-OVX. Body composition, EB, SPA, and blood insulin and glucose data from the SED groups were compared with rats that did not lose weight (9) to test for additive effects of OVX and prior weight loss on post-OVX weight gain as a distinct research question.

Programmed Exercise

At the start of weight loss, rats randomized to the EX groups were acclimated to treadmill exercise (Exer-6M Treadmill; Columbus Instruments, Columbus, OH). The first 4 wk were a ramp up phase where time and speed were gradually increased to the final prescribed dose of exercise, which was 15 m·min−1 for 60 min·d−1, 6 d·wk−1. Exercise bouts occurred during the light cycle within 2 h of the start of the dark cycle. Rats reached the full speed and time while they were in the weight maintenance phase and remained at the same full speed and time until euthanasia. Our pilot data indicated that the energy expended during treadmill exercise was very similar (<0.5 kcal difference) between OP and OR rats (data not shown). Rats were scored from 1 (worst) to 10 (best) during each bout of exercise based on the quality of running and amount of encouragement required to complete the exercise bout, as previously described (19).

Body Composition

Fat mass and fat-free mass were measured using quantitative magnetic resonance (EchoMRI, Houston, TX) at three time points: (1) before the initiation of weight loss, (2) just before OVX, and (3) on the day of euthanasia.

Metabolic Monitoring

Energy intake (EI), SPA, and total EE (TEE) and REE were measured in metabolic monitoring systems with indirect calorimetry as previously described (9). Each metabolic cage was equipped with an animal activity meter (Opto-Max; Columbus Instruments), and total, ambulatory, and nonambulatory activity were monitored for each 24 h period. Within the week before OVX, daily vaginal lavages were performed 4 to 5 h before the start of the dark cycle to determine the timing of the estrus phase within the estrous cycle to time the start of metabolic monitoring. Pre-OVX metabolic monitoring lasted for 4 d. After recovery from OVX surgery (3–7 d), rats were monitored in the calorimeters for 3 d while on calorically limited (CL) provisions to remain in EB, plus 3 d of AL refeeding. Finally, rats were monitored for 4 d during AL feeding within a week of euthanasia (~ 9–11 wk post-OVX, Fig. 1A).

Metabolic rate (MR) was calculated from gas exchange measurements acquired every 14 min using the Weir equation (20): MR = 3.941 V˙O2 + 1.106 V˙CO2 – 2.17 N, where N is urinary nitrogen. The MR was averaged and extrapolated over 24 h to estimate TEE. Energy balance was calculated as the difference between EI and TEE. Metabolic data were derived as mean values from 4 d in pre-OVX, 1 to 3 d in post-OVX-CL, 1 to 3 d in post-OVX-AL, and 1 to 3 d in post-OVX final.

Skeletal Outcomes

Areal bone mineral density and bone mineral content

Immediately before both OVX and sacrifice, total body areal bone mineral density (aBMD) and bone mineral content were assessed using dual-energy X-ray absorptiometry (Lunar Corp., Madison, WI) while under isoflurane anesthesia.

Long Bone Morphometry

Microcomputed tomography (Siemens Inveon; Erlangen, Germany) scans were performed on whole femora and tibiae at a resolution of 30 and 47 μm, respectively. Cortical cross-sectional area (CSA), thickness, second moments of area (Imax, Imin, measures of bending resistance in the directions of maximum and minimum stiffness), and section moduli (Zmax, Zmin, measures of bending strength in the directions of maximum and minimum strength) were measured along the diaphyseal shafts of tibiae and femurs using the BoneJ plugin for ImageJ (21). The volume of interest for the tibial shaft was from the tibia-fibula junction to 14.1 mm proximal, averaged over 150 slices spaced 47 μm apart. The volume of interest for the femoral shaft was 6 mm in each direction from the femoral midshaft, averaged over 200 slices spaced 30 μm apart. Mechanical testing of hindlimb bones was performed as previously described (19).

Mitochondrial Respiration

Gastrocnemius skeletal muscle was rapidly excised from each hindlimb under anesthesia. Mitochondrial isolations were performed at 0°C to 4°C according to the methods of Makinen and Lee (22). Resting (state 4) and maximal coupled (state 3) mitochondrial oxygen consumption was measured in a respiration chamber maintained at 37°C (Strathkelvin Instruments, North Lanarkshire, Scotland). Incubations were carried out at 37°C in a 0.5-mL final volume containing 100 mM KCl, 50 mM MOPS, 10 mM K2PO4, 10 mM MgCl2, 0.5 mM ethylene glycol-bis(β-aminoethyl ether)-N,N,N′,N′-tetraacetic acid, 20 mM glucose, and 0.2% bovine serum albumin, pH 7.4. Mitochondrial respiration was monitored at the following concentrations: 1 mM malate, 10 mM glutamate, 1 mM pyruvate, 10 μM l-palmitoylcarnitine, and 10 mM succinate. Mitochondria and substrates were added, and the coupled maximal respiration rate was initiated with the addition of ADP (100 μM).

Serum Hormones and Substrates

Blood was collected from the inferior vena cava at the time of euthanasia, and plasma was stored at −80°C until analysis. All analyses were performed in duplicate. Plasma insulin and undercarboxylated osteocalcin were measured by ELISA (80-INSRT-E01 ALPCO; Salem, NH and MK118 Takara Bio USA, Mountain View, CA, respectively); plasma glucose was measured using a colorimetric assay (TR15421; Thermo Fisher Scientific, Waltham, MA). Follicle-stimulating hormone (FSH) was measured by MILLIPLEX map Rat Pituitary Panel kit, EMD Millipore product RPTMAG-86K.

Statistical Analyses

Primary analyses using phenotype and exercise status as factors

Data were analyzed with SAS version 9.4. This study was designed to test a difference in the expected change in weight and fat mass between EX and SED rats (23). Based on our previous work in males, a sample size of 6 was required to detect a 35‐g difference in weight gain between SED and EX rats at 80% power and a P value of 0.05. Two-way (phenotype, activity) ANOVA were used to test the hypothesis that exercise will attenuate the increases in weight and fat mass after OVX in OP and OR rats. In the case of a significant interaction, a Tukey post hoc comparison was made to test for differences between EX and SED. If the interaction was not significant, then main effects (phenotype, activity) were interpreted.

We performed similar two-way ANOVA on changes in EI, EE, EB, and SPA (secondary outcomes) as the mechanistic underpinnings for weight regain. Two-way ANOVA at each time point for these outcomes are used for descriptive purposes to aid in the interpretation of change comparisons. End-of-study outcomes (e.g., bone and mitochondria outcomes) were also compared with two-way ANOVA. Finally, we performed an ANCOVA with TEE as the dependent variable and lean mass as the covariate to explore group differences in metabolic efficiency. Pearson’s product moment correlations were used to examine relationships among secondary outcomes. The level of significance was set at P < 0.05.

Exploring the role of changes in spontaneous physical activity on post-OVX outcomes

Because we observed a very large weight regain and a very large reduction in SPA during the first week of AL refeeding, we performed a hypothesis-generating secondary analysis to explore whether the decrease in SPA after OVX was counteracting our expected effects of exercise. We split groups based on the median relative change in total SPA from pre-OVX to the early regain (first 4 d) post-OVX, designating them as “sustained SPA (S-SPA)” and “declining SPA (D-SPA),” and performed a two-way ANOVA with total SPA and Exercise status as the factors. Five of 12 OP rats were S-SPA, and 7 of 13 OR rats were S-SPA. Rats that did not have SPA data during the early refeeding phase were not included in the secondary analysis. The level of significance was set at P < 0.05.

RESULTS

Before the loss of ovarian function

Before OVX, all groups had similar lean mass, and the difference in body weight between OP and OR rats was driven by a higher fat mass in OP rats before (102 ± 5 g vs 61 ± 2 g) and after weight loss (43 ± 4 g vs 22 ± 2 g) (Fig. 1B, E and F). There were no group differences in EI (Table 1) or SPA [see Figure, Supplemental Digital Content 2, total, ambulatory, and nonambulatory spontaneous physical activity (SPA) progressively decreased (all P < 0.01) from before OVX to the calorically limited phase (before regain) after OVX, to the early AL refeeding phase after OVX, and to just before sacrifice; P > 0.36 for both, http://links.lww.com/MSS/B671] in the pre-OVX period. However, TEE was higher in both OP rats (P = 0.032) and EX rats (P = 0.012) compared to their OR and SED counterparts, and the increase in REE and EB in OP rats neared significance (P = 0.090, 0.091; Table 1). Running compliance before OVX was similar between OR and OP rats (running scores, 8.1 ± 0.2 and 7.7 ± 0.3; P > 0.29) and remained similar between OR and OP rats after OVX (running scores, 6.6 ± 0.4 and 7.1 ± 0.2; P > 0.28).

TABLE 1
TABLE 1:
EB, EE, EI before OVX, after OVX during CL, and early AL feeding, and at end of study.

Roles of obesity and exercise after OVX in rats with prior weight loss

The EX rats met or surpassed their preweight loss weight within 2 wk of OVX under AL feeding conditions (13.9 ± 1.5 d, Fig. 1C). Although the rate of weight regain in the first 3 wk of AL feeding was similar (P > 0.40) among all groups, exercise resulted in an eventual attenuation of weight regain in OP rats, which was still greater than in OR rats (interaction; P = 0.01, Fig. 1C-D). The divergence in weight regain trajectories emerged in week 5 (interaction; P = 0.096; phenotype; P < 0.01) and week 6 (interaction; P = 0.064). The reduction in weight regain in OP-EX compared with OP-SED was explained by smaller increases in both fat (interaction; P = 0.09; phenotype; P < 0.001) and lean mass (interaction; P = 0.03), whereas OR-EX increased lean mass compared to OR-SED (Figs. 1E–F).

There were no group differences in the increases in EB nor were there group differences in EB during early regain or end of study (P > 0.13 for group differences, Fig. 2A). There were also no group differences in EI (Fig. 2B). Energy intake increased (P < 0.01) similarly in all groups after OVX (P > 0.35 for group difference in change), although EI at the end of regain tended to be higher in OP (P = 0.072). After OVX, during the calorically limited phase, TEE remained higher in OP compared with OR and EX compared to SED rats (P = 0.022, 0.028, Table 1). During early refeeding, TEE (Fig. 2C) and REE increased similarly in all groups (all P > 0.24). At the end of weight regain, the increases in TEE and REE remained similar among OR and OP groups. The SED rats had a larger increase (P < 0.040) in TEE and REE from pre-OVX to the end of study, compared with EX rats (Table 1; Fig. 2).

FIGURE 2
FIGURE 2:
Increases in positive EB (A) after OVX were driven by the large increase in EI (B), especially during the early (first week) regain phase. EI, and the resulting EB were still higher than pre-OVX levels at the end of the study when weights plateaued, but to a lesser extent. TEE (C) also increased during the post-OVX early regaining phase, but remained higher than pre-OVX, largely because of the heavier body weight. *P < 0.05, main effect of activity (EX < SED).

Spontaneous physical activity decreased (P < 0.01) similarly in all groups after OVX (group differences in change all P > 0.38, Fig. 3; Supplemental Digital Content 2—Supplemental Figure 1, http://links.lww.com/MSS/B671), and there were no group differences in total or ambulatory SPA counts at any time point (all P > 0.20). The decrease (−13.2% ± 5.1%) in total SPA after OVX was observed before refeeding, worsened during refeeding (−34.5% ± 5.1%), and continued to decrease until the end of the study (−45.1% ± 2.9%). Additionally, ambulatory SPA decreased (−14.8% ± 6.5% before refeeding, −37.6% ± 6.6% during early refeeding, −56.4% ± 3.1% at the end of the study) to a greater extent than total or nonambulatory SPA. As a result, ambulatory SPA became less of a contributor to total SPA over time (51.6% ± 1.0% of total SPA at pre-OVX to 39.6% ± 0.7% of total SPA at the end of the study).

FIGURE 3
FIGURE 3:
Total SPA pre-OVX, post-OVX before regain (CL), post-OVX during early regain (AL), and at the end of the study (final) consists of ambulatory (solid bars) and nonambulatory (striped bars) SPA. Both ambulatory and nonambulatory SPA decrease over time, but ambulatory SPA is lost to a greater extent. All post-OVX time points are significantly lower than pre-OVX values.

When adjusting for lean mass, EX rats had a higher TEE before OVX (P = 0.001) and during the calorically limited (P = 0.007) phase [see Figure, Supplemental Digital Content 3, Total EE, adjusted for body weight, lean mass, or fat mass before OVX, and after OVX during caloric limitation, during the first 3 d of regain, and just before sacrifice, http://links.lww.com/MSS/B672]. During the refeeding period, the difference in TEE between OP-EX (higher) and OP-SED was larger than the difference between the OR groups (interaction P = 0.021; Supplemental Digital Content 3—Supplemental Figure 2, http://links.lww.com/MSS/B672). However, TEE at the end of the study was similar among groups after adjusting for lean mass (P > 0.69). Similar patterns were found when TEE was adjusted by fat mass or total body weight, but the interaction was not significant during refeeding.

Tibia size (CSA, periosteal circumference, cortical thickness) and strength-related outcomes (Imax, Imin, Zmax, Zmin) were not different between OR and OP rats, but were improved with exercise, such that improvements in size led to improvements in moduli and bending resistance (Fig. 4). However, the effect of exercise on femur outcomes was not significant, due to the lack of an increase in EX OP rats. We observed this pattern for EX to improve gastrocnemius mitochondrial respiration in OR rats, but not in OP rats, but it did not reach statistical significance (see Table, Supplemental Digital Content 4, State 3 and state 4 respiration of gastrocnemius muscle, http://links.lww.com/MSS/B673). Resting (state 4) respiration in the presence of glutamate+malate or pyruvate+palmotylcarnitine was correlated with tibia and femur size (CSA), stiffness, and strength (Imax, Imin, experimental failure force) outcomes (r = 0.40–0.56, P < 0.05). Resting respiration in the presence of glutamate+malate was correlated with change in lean mass (r = 0.39, P < 0.03).

FIGURE 4
FIGURE 4:
Skeletal outcomes. A, tibia CSA, (B) femur CSA, (C) tibia maximum and minimum second moment of area (Imax, Imin), (D) femur Imax, Imin.

Corticosterone levels demonstrated an increase (P < 0.001) over time, but an interaction (P = 0.021) only existed at the end of the study, such that corticosterone was higher in OR-EX compared to OR-SED rats. In contrast, OP-EX had lower corticosterone than OP-SED. Insulin levels at the end of the study were higher in OP rats (P < 0.05, Supplemental Digital Content 1—Supplementary Table 1, http://links.lww.com/MSS/B670). The FSH concentrations at the end of the study were lower in OP rats (P < 0.05, see Table, Supplemental Digital Content 5, Plasma FSH, insulin, and osteocalcin, and urinary corticosterone, http://links.lww.com/MSS/B674).

Exploring the role of changes in spontaneous physical activity on post-OVX outcomes

The SPA change was independent of exercise, as SPA levels did not show signs of compensation for exercise status. Early change in SPA was correlated (P < 0.05) with almost every major outcome, including weight and fat mass gain (r = −0.45 to −0.47), changes in EI and TEE (r = −0.41 to −0.66), Imin (r = 0.45–0.55), glucose (r = 0.39–0.53), and osteocalcin concentrations (r = 0.44). Decreases in SPA were correlated with a lower mitochondrial respiration of pyruvate, glutamate, and glutamate+malate (r = 0.45–0.51).

Our exploratory analysis uncovered that rats with higher levels of SPA before OVX had greater decreases in both total and ambulatory SPA after OVX, and were designated as D-SPA, compared to rats with lower pre-OVX levels of SPA. Rats that exhibited a smaller decrease in SPA post-OVX during early refeeding were defined as S-SPA because they better sustained activity throughout the regain phase. S-SPA rats had a continued reduction in SPA throughout regain, whereas the decline in SPA in D-SPA rats slowed after the early refeeding phase. At the end of the regain period, S-SPA rats still better maintained SPA (P < 0.05, Fig. 5). The magnitude of decreases in SPA had an independent effect on rate of weight regain and the total weight regain through the first week of refeeding, such that D-SPA rats gained approximately 27% more weight than S-SPA rats during the first week (67.5 ± 3.7 vs 53.4 ± 7.1 g). This was associated with a greater increase in EI (P < 0.05) in the D-SPA group during early refeeding. This, in turn, resulted in a greater total weight gain in D-SPA rats that neared significance (P = 0.062). Independent of exercise or phenotype, S-SPA rats had greater strength in the tibia (failure load, Imin, Zmin, P < 0.03 and Zpolar P = 0.08), due to periosteal apposition (P = 0.06). Within the exploratory analysis, the main effect of activity (EX vs SED) did not reach significance for any outcome [see Figure, Supplemental Digital Content 6, Secondary analysis with SED and EX groups separated based on the median relative change (%) in spontaneous physical activity (SPA) from pre-OVX to the first week of post-OVX AL refeeding demonstrated two distinct patterns of loss of SPA, http://links.lww.com/MSS/B675].

FIGURE 5
FIGURE 5:
A, Secondary analysis based on the median relative change (%) in SPA from pre-OVX to the first week of post-OVX AL refeeding demonstrated two distinct patterns of loss of SPA. B, Those who had the largest decreases in SPA after OVX were more active before OVX. C, Groups who remained closer to pre-OVX SPA levels (S-SPA) during early post-OVX refeeding had lower early increases in EI and TEE, and (D) slower weight regain over the first week of post-OVX AL refeeding. E, At the end of study, S-SPA rats had higher tibia resistance to bending. F, S-SPA groups tended to have less weight regain (P < 0.06), which was primarily fat (N = 5–8 per group). *P < 0.05, significantly different from their D-SPA counterparts.

DISCUSSION

The most significant observation of this study was that after OVX, regular exercise attenuated weight regain in OP, but not OR, rats. A unique observation that stemmed from our secondary analysis was that the rate of decline in SPA after OVX independently affected food intake, weight regain, and bone outcomes, and this relationship was not diminished with regular exercise. Taken together, these data would suggest that leanness and exercise do not necessarily protect from biological drive to regain weight after the loss of ovarian function. Furthermore, adiposity and physical activity levels may modulate whether exercise has any effect on body weight regulation in this context. Therefore, it will be important to elucidate the underlying mechanisms of the interplay between SPA, the loss of ovarian hormones, and obesity and their effects on menopause-associated metabolic dysfunction.

It is very common for women to enter the menopausal transition with preexisting obesity and a history of weight loss, and exercise is often used to assist in weight control. We recreated this phenotypic milieu to gain novel insight into how exercise can counteract the energetic challenges of menopause. Combining a history of prolonged caloric restriction with loss of ovarian function resulted in rapid weight gain after OVX, as described in our recent article (9). We expected exercise to counteract the biological drive to regain weight by preserving insulin sensitivity, mitochondrial biogenesis, and fat oxidation in muscle, bone, and liver, and the resulting EE that might otherwise decrease with the removal of ovarian hormones (8,24–26). Exercise reduces overfeeding during weight regain in males, which can then prevent weight from returning to preweight loss levels (13). Prior work suggested that females under the same conditions would compensate for the energy deficit of exercise with increased food intake (27). This appears to be the case for female OR OVX rats in the current study and in intact females in our prior work (11). At present, it is unclear as to why adiposity levels modulate the effect of exercise on weight regain after OVX, but it appears that obesity is modulating how exercise and ovarian hormones interact in homeostatic regulation.

Considering that SPA markedly decreased after OVX, the increase in TEE during early regain was clearly a reflection of increased food intake and the resulting thermic effect of feeding. The changes in absolute TEE levels did not provide a clear explanation for the differential effects of exercise on weight regain in OR and OP rats. The adjusted (lean mass) analysis would, however, suggest that regular exercise reduced metabolic efficiency in OP rats. This effect on metabolic efficiency may have contributed to the attenuated weight regain over the subsequent 8 wk of refeeding. Regardless of lean mass differences, the modest (~2 kcal·d−1) difference in EB between OP-EX and OP-SED could reasonably explain the differential weight gain when extrapolated over time.

Another factor that may have contributed to this differential weight regain is how the ingested energy is utilized and stored. We previously observed that OR female rats inherently have a greater capacity to oxidize dietary fat during overfeeding (3). Exercise training is known to induce a glycogen-sparing effect and increases the capacity to oxidize fat (28). In male OP rats, we observed that regular exercise increased the oxidation of dietary fat and trafficked excess carbohydrate through more energetically expensive pathways of deposition (de novo lipogenesis) (23). This differential fuel trafficking and use was associated with an increase in TEE and attenuated weight regain. In the present study, we speculated that exercise may have been less effective in OR females because of their inherent ability to take up and oxidize dietary fat. Contrary to this hypothesis, regular exercise tended to improve metabolic capacity of isolated mitochondria in OR, but not OP rats. Taken together, these data would suggest that the exercise-induced attenuation of weight regain in OP rats could not be linked to a preferential induction of metabolic capacity. To our surprise, it was the decline in SPA, and not the exercise intervention, that was associated with the metabolic characteristics of muscle mitochondrial function.

The bone formation response to mechanical loading is strongest when the loads (and resulting strains) are dynamic, rapid or large, and non-routine. Favorable skeletal adaptations with exercise were not surprising, because exercise provides dynamic loading (29). The effect of exercise was more evident in the tibia, which experiences greater mechanical strain during running. Our results clearly indicate that adiposity, which is static loading, did not induce favorable bone adaptations. We did, however, find an independent effect of SPA on skeletal outcomes in the secondary analysis. SPA generally consists of numerous low-level routine loads (30), which could protect osteocyte and osteoblast viability and potentiate the response to higher magnitude mechanical loading that comes with exercise (31).

Given that the changes in SPA correlated with so many of the outcomes in this study, it may be that we need to direct more attention to this critical variable and its impact on the metabolic effects of OVX. The fact that these rats had large individual variability in the change in SPA in response to OVX allowed us to perform a hypothesis-generating secondary analysis. The decline in SPA that occurs with OVX is thought to be estrogen-dependent (8,32), but more recent evidence suggest that the compensatory increase in FSH may also be involved (33). We do not know what underlies the variability in the decrease in SPA or why those that had higher pre-OVX SPA lost more SPA after OVX, although central regulation is certainly involved (34). What we show here and in our previous report is that the decline in SPA precedes the weight regain (9). In this article, we go further to show that these changes predict food intake, EB, and subsequent decline in metabolic health. Whether this relationship is mediated by the capacity to oxidize fat (35,36), changes in insulin or leptin sensitivity (37), or some other mechanism, will need to be determined in future studies.

Surgical OVX, as used in the present study, imparts an abrupt change in the hypothalamic–pituitary–gonadal axis, whereas menopause is a longer-term transition in which the function of the ovaries is highly variable. This limitation in modeling the human condition needs to be considered when translating our observations. Although the timing may be different, both OVX and menopause lead to a compensatory increase in FSH, which has been implicated in having an independent effect on adiposity, bone density, and other aspects of metabolic health (33). In the present study, end-of-study FSH concentrations were lower in OP rats. It is unclear if FSH is a moderator of the effects of exercise on weight regain after the loss of ovarian function, but it is FSH has been proposed to have numerous effects on peripheral tissues through its putative receptor (38). Future studies that dissect out the specific effects of each component of the hypothalamic-pituitary-gonadal axis are warranted, while considering the important differences between the surgical intervention and the menopausal transition.

CONCLUSIONS

In conclusion, leanness and exercise do not necessarily protect from OVX-induced weight gain. Exercise ultimately prevented weight gain in OP rats, but it was loss of SPA that was the greatest contributor to post-OVX weight gain. These observations suggest a complex perturbation in homeostatic regulation that has implications for women at the menopausal transition. Further studies are needed to understand the interplay between adiposity, SPA, and the loss of ovarian function, so that we can better prevent menopause-associated metabolic dysfunction.

The authors appreciate the assistance of the Colorado Nutrition and Obesity Research Center’s Energy Balance Laboratory and Metabolic Core [National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) Grant DK-48520].

Grants: This study was supported by the following National Institutes of Health (NIH) grants: P50 HD073063 (PSM, VDS), T32 DK007658 (VDS), TR002534 (VDS), U54 AG062319 (PSM), P30 48520 (NORC), K99/R00 CA169430 (EDG), UL1TR002535 (JAH), TL1TR002533 (RF, DP).

Disclosures: No conflicts of interest, financial or otherwise, are declared by the authors. Results of the study do not constitute endorsement by the American College of Sports Medicine.

REFERENCES

1. Kraschnewski JL, Boan J, Esposito J, et al. Long-term weight loss maintenance in the united states. Int J Obes (Lond). 2010;34(11):1644–54.
2. Maclean PS, Bergouignan A, Cornier MA, Jackman MR. Biology’s response to dieting: The impetus for weight regain. Am J Physiol Regul Integr Comp Physiol. 2011;301(3):R581–600.
3. Jackman MR, Kramer RE, MacLean PS, Bessesen DH. Trafficking of dietary fat in obesity-prone and obesity-resistant rats. Am J Physiol Endocrinol Metab. 2006;291(5):E1083–91.
4. Astrup A. The relevance of increased fat oxidation for body-weight management: metabolic inflexibility in the predisposition to weight gain. Obes Rev. 2011;12(10):859–65.
5. Polotsky HN, Polotsky AJ. Metabolic implications of menopause. Semin Reprod Med. 2010;28(5):426–34.
6. Lobo RA. Metabolic syndrome after menopause and the role of hormones. Maturitas. 2008;60(1):10–8.
7. Melanson EL, Gavin KM, Shea KL, et al. Regulation of energy expenditure by estradiol in premenopausal women. J Appl Physiol (1985). 2015;119(9):975–81.
8. Giles ED, Jackman MR, Johnson GC, Schedin PJ, Houser JL, MacLean PS. Effect of the estrous cycle and surgical ovariectomy on energy balance, fuel utilization, and physical activity in lean and obese female rats. Am J Physiol Regul Integr Comp Physiol. 2010;299(6):R1634–42.
9. Sherk VD, Jackman MR, Giles ED, et al. Prior weight loss exacerbates the biological drive to gain weight after the loss of ovarian function. Physiol Rep. 2017;5(10).
10. Tate DF, Jeffery RW, Sherwood NE, Wing RR. Long-term weight losses associated with prescription of higher physical activity goals. Are higher levels of physical activity protective against weight regain? Am J Clin Nutr. 2007;85(4):954–9.
11. Foright RM, Presby DM, Sherk VD, et al. Is regular exercise an effective strategy for weight loss maintenance? Physiol Behav. 2018;188:86–93.
12. Catenacci VA, Ogden LG, Stuht J, et al. Physical activity patterns in the national weight control registry. Obesity. 2008;16(1):153–61.
13. MacLean PS, Higgins JA, Wyatt HR, et al. Regular exercise attenuates the metabolic drive to regain weight after long-term weight loss. Am J Physiol Regul Integr Comp Physiol. 2009;297(3):R793–802.
14. Giles ED, Steig AJ, Jackman MR, et al. Exercise decreases lipogenic gene expression in adipose tissue and alters adipocyte cellularity during weight regain after weight loss. Front Physiol. 2016;7:32.
15. Pighon A, Paquette A, Barsalani R, et al. Resistance training attenuates fat mass regain after weight loss in ovariectomized rats. Maturitas. 2009;64(1):52–7.
16. Jackman MR, MacLean PS, Bessesen DH. Energy expenditure in obesity-prone and obesity-resistant rats before and after the introduction of a high-fat diet. Am J Physiol Regul Integr Comp Physiol. 2010;299(4):R1097–105.
17. Giles ED, Jackman MR, MacLean PS. Modeling diet-induced obesity with obesity-prone rats: implications for studies in females. Front Nutr. 2016;3:50.
18. Park SB, Lee YJ, Chung CK. Bone mineral density changes after ovariectomy in rats as an osteopenic model: stepwise description of double dorso-lateral approach. J Korean Neurosurg Soc. 2010;48(4):309–12.
19. Sherk VD, Carpenter RD, Giles ED, et al. Ibuprofen before exercise does not prevent cortical bone adaptations to training. Med Sci Sports Exerc. 2017;49(5):888–95.
20. Weir JB. New methods for calculating metabolic rate with special reference to protein metabolism. J Physiol. 1949;109(1–2):1–9.
21. Doube M, Klosowski MM, Arganda-Carreras I, et al. Bonej: free and extensible bone image analysis in imagej. Bone. 2010;47(6):1076–9.
22. Makinen MW, Lee CP. Biochemical studies of skeletal muscle mitochondria. I. Microanalysis of cytochrome content, oxidative and phosphorylative activities of mammalian skeletal muscle mitochondria. Arch Biochem Biophys. 1968;126(1):75–82.
23. Steig AJ, Jackman MR, Giles ED, et al. Exercise reduces appetite and traffics excess nutrients away from energetically efficient pathways of lipid deposition during the early stages of weight regain. Am J Physiol Regul Integr Comp Physiol. 2011;301(3):R656–67.
24. Oosthuyse T, Bosch AN. Oestrogen’s regulation of fat metabolism during exercise and gender specific effects. Curr Opin Pharmacol. 2012;12(3):363–71.
25. Gupte AA, Pownall HJ, Hamilton DJ. Estrogen: an emerging regulator of insulin action and mitochondrial function. J Diabetes Res. 2015;2015:916585.
26. Rohling M, Herder C, Stemper T, Mussig K. Influence of acute and chronic exercise on glucose uptake. J Diabetes Res. 2016;2016:2868652.
27. Applegate EA, Upton DE, Stern JS. Food intake, body composition and blood lipids following treadmill exercise in male and female rats. Physiol Behav. 1982;28(5):917–20.
28. Fan W, Waizenegger W, Lin CS, et al. PPARdelta promotes running endurance by preserving glucose. Cell Metab. 2017;25(5):1186–93 e4.
29. Turner CH. Three rules for bone adaptation to mechanical stimuli. Bone. 1998;23(5):399–407.
30. Rubin CT, Lanyon LE. Regulation of bone formation by applied dynamic loads. J Bone Joint Surg Am. 1984;66(3):397–402.
31. Sherk VD, Rosen CJ. Senescent and apoptotic osteocytes and aging: exercise to the rescue? Bone. 2019;121:255–8.
32. Gorzek JF, Hendrickson KC, Forstner JP, Rixen JL, Moran AL, Lowe DA. Estradiol and tamoxifen reverse ovariectomy-induced physical inactivity in mice. Med Sci Sports Exerc. 2007;39(2):248–56.
33. Liu P, Ji Y, Yuen T, et al. Blocking fsh induces thermogenic adipose tissue and reduces body fat. Nature. 2017;546(7656):107–12.
34. Roberts MD, Ruegsegger GN, Brown JD, Booth FW. Mechanisms associated with physical activity behavior: Insights from rodent experiments. Exerc Sport Sci Rev. 2017;45(4):217–22.
35. Bergouignan A, Momken I, Lefai E, et al. Activity energy expenditure is a major determinant of dietary fat oxidation and trafficking, but the deleterious effect of detraining is more marked than the beneficial effect of training at current recommendations. Am J Clin Nutr. 2013;98(3):648–58.
36. Rynders CA, Bergouignan A, Kealey E, Bessesen DH. Ability to adjust nocturnal fat oxidation in response to overfeeding predicts 5-year weight gain in adults. Obesity. 2017;25(5):873–80.
37. Flores MB, Fernandes MF, Ropelle ER, et al. Exercise improves insulin and leptin sensitivity in hypothalamus of wistar rats. Diabetes. 2006;55(9):2554–61.
38. Das N, Kumar TR. Molecular regulation of follicle-stimulating hormone synthesis, secretion and action. J Mol Endocrinol. 2018;60(3):R131–55.
Keywords:

MENOPAUSE; ENERGY BALANCE; SPONTANEOUS PHYSICAL ACTIVITY; OBESITY; EXERCISE

Supplemental Digital Content

Copyright © 2019 by the American College of Sports Medicine