SCHOFIELD, KATHERINE L.1; REHRER, NANCY J.1,2; PERRY, TRACY L.2; ROSS, ANGUS3; ANDERSEN, JESPER L.4; OSBORNE, HAMISH5
Insulin resistance is a known precursor for developing type 2 diabetes mellitus (T2DM) (12); therefore, the study of individuals who may have a genetic predisposition to T2DM is warranted. Resistance training has been reported to enhance insulin sensitivity, although there is limited literature to support this in type 2 diabetic populations (7,13,14), and no literature in those who may have a genetic predisposition. Resistance training causes a decrease in the proportion of myosin heavy chain (MHC) IIX isoform, with a corresponding increase in MHC IIA (1,2,10). In contrast, with detraining, a shift toward MHC IIX has been observed (2,28). GLUT-4 protein expression is reduced in Type I fibers of type 2 diabetic individuals (16) and appears to increase with resistance training (29). Therefore, changes in fiber type composition could influence glucose tolerance.
Within healthy populations, there is a trend toward significantly reduced insulin sensitivity (6,20,22) or glucose tolerance (17) with periods of 5 to 15 d of inactivity after previous endurance training. This effect has also been observed after 10 d of bed rest in individuals that have a parent(s) with type 2 diabetes (26). The longest detraining period conducted was that reported by Andersen et al. (3) who observed a significant reduction in insulin-mediated glucose uptake after 3 months of detraining. The extent of the change in insulin response with resistance training and subsequent detraining (for more than 10 d) in individuals who have a genetic predisposition to T2DM is unknown.
There is sufficient evidence to indicate that hyperinsulinemia causes changes in muscle fiber type from slow-twitch oxidative (Type I) to fast-twitch oxidative and glycolytic (Type IIA and Type IIX, respectively) (19). Individuals with T2DM, as well as the offspring of those with T2DM, typically have proportionally more Type II fibers (24,25) compared with those without T2DM. Those with a preponderance of Type II fibers would be expected to have enhanced performance in strength- and power-based sports; therefore, there may be positive ramifications of hyperinsulinemia on human athletic power production (11).
Therefore, the aim of the present study was to examine muscle fiber type composition, glucose tolerance, and strength and power in the offspring of those with T2DM and control subjects and their responses to a period of resistance training and subsequent detraining. We hypothesized that the offspring of T2DM subjects would have a greater proportion of MHC IIX fibers at baseline because Nyholm et al. (24) observed a greater proportion of MHC IIB (X) fibers in the first-degree relatives of T2DM subjects and because chronic elevation in insulin is known to convert Type I fibers to Type II fibers (9). We also hypothesized that the offspring of those with T2DM would have enhanced strength and power at baseline, on the basis of the greater proportion of Type II fibers and in particular IIX. We also hypothesized that both groups would increase performance with training and reduce performance with detraining. We further hypothesized that the offspring of those with T2DM would demonstrate a greater reduction in insulin response to a glucose load and a greater increase in insulin response with a period of detraining, because it was expected that familial insulin-resistant (FIR) participants would have had elevated initial insulin concentrations and as such would be more likely to respond to exercise. We hypothesized that a conversion of Type IIX to IIA fibers would be observed in both groups in response to resistance training and a reversal with detraining, and that a greater transition from IIX to IIA with training and a greater rebound of MHC IIX with detraining would be observed in the offspring of T2DM subjects. We also expected a positive correlation of insulin response, as well as performance, with MHC II isoforms.
Participants completed 9 wk of resistance training followed by 9 wk of detraining. Anthropometry (body mass; skinfolds (to determine percentage body fat); and waist, hip, and thigh girth), performance tests, an oral glucose tolerance test (OGTT), and a muscle biopsy were conducted before the training program (T1), at the end of training (T2), and after detraining (T3). The OGTT and muscle biopsy were performed 3 d (72 h) and within 1 wk, respectively, after the final training session. Physical activity records with type, duration, and intensity of exercise, outside of training, were completed by each participant at T1 and T3.
Seventeen healthy participants volunteered and gave written informed consent to participate in the study, which was approved by the University of Otago Human Ethics Committee (08/150). All participants were nonsmoking, not on medication, and free of type 1 and type 2 diabetes (normal fasted venous blood plasma glucose level of <7.0 mmol·L−1 and a 2-h postglucose load of <11.1 mmol·L−1) as defined by the World Health Organization (30). They were required to be familiar with, but not involved in, resistance training in the last 6 months, without any musculoskeletal injury in the last 6 wk. Participants were divided into FIR (two men and five women) and control (C: four men and six women) groups. FIR had to have one or both parents with clinically diagnosed T2DM (three participants had a father with T2DM, four participants had a mother with T2DM, and one participant had both parents with T2DM) and C had no familial link to T2DM. One male FIR participant dropped out midway through the training program because of personal reasons; therefore, his data were removed from analyses. The characteristics of the groups at baseline are presented in Table 1.
Resistance training was conducted three times a week (Monday, Wednesday, and Friday) for 9 wk (total of 27 sessions) and was supervised by one of the researchers. Four lower body exercises were completed bilaterally on gym equipment and consisted of squat, leg press, knee extension, and hamstring curl exercises. The load for each exercise was calculated from an estimation of the participants’ one-repetition maximum (1 RM). The goal of the resistance training was to increase strength increase hypertrophy and increase power. The training was progressive in that the load was increased throughout the program, e.g., 10–15 repetitions, two to four sets at 50%–60% 1RM in week 1, to 5–10 repetitions, four to seven sets at 65%–85% 1RM in week 9. In addition, the load was monitored and adjusted in weeks 4 and 6 (sessions 10 and 17, respectively), as 3RM testing was repeated to determine progression. From week 3, participants completed an additional exercise, a box jump, once per week (one to four repetitions, three to six sets). The box jump was included to improve power production. The recovery between each set of all exercises was 3 min.
After the resistance training program (T1–T2, week 1–9), participants entered a 9-wk detraining phase (T2–T3, week 10–18). Participants were instructed to return to their baseline physical activity levels. To ensure physical activity levels returned to baseline in the detraining period (T2–T3), physical activity records were completed by each participant in weekly meetings or weekly e-mails. Participants were reminded of their baseline levels and advised not to increase their physical activity.
Blood Sampling and Analyses
An OGTT was performed in the morning (0700–0730 h) after a 10-h overnight fast and no strenuous exercise for 24 h prior. A cannula was inserted in an antecubital vein. Blood samples of 3.5 mL were drawn and transferred to ethylenediaminetetraacetic acid tubes for analysis of plasma glucose and insulin concentration. After a fasting blood sample was drawn, participants consumed a 75-g glucose drink (time = 0) (CarboTest®, Auckland, New Zealand) within 5 min. Additional blood samples were drawn at 15, 25, 30, 60, 90, and 120 min, and cannula patency was maintained with a saline flush (0.9%) between each sample. The samples were centrifuged at 3000g (4°C) for 10 min, and blood plasma was removed and stored at −80°C until analysis.
Plasma glucose concentration was determined by automated spectrophotometric analysis (Cobas c 111; Roche Instrument Centre, Switzerland) using the hexokinase reaction (GLUC2; Roche Diagnostics GmbH, Indianapolis, IN). Plasma insulin concentration was determined by an electrochemiluminescence immunoassay (ECLIA Insulin, Elecsys Insulin kit, Roche Diagnostics). Coefficients of variation for blood glucose and blood insulin were 2.0% and 1.7%, respectively.
Muscle Sampling and Analysis
Needle biopsy samples were taken from the vastus lateralis of the right thigh, under local anesthesia (Xylocaine®, lignocaine hydrochloride, 50 mg in 5 mL (1%) (AstraZeneca Pty Ltd, North Ryde, New South Wales)). The muscle sample was covered with mounting medium (OCT compound—TissueTek®; Sakura Finetek, Inc., Torrance CA) and frozen in isopentane, cooled in liquid nitrogen. Samples were stored at −80°C until analysis. Biopsies at T2 and at T3 were taken 1 cm distally or proximally (random allocation) from the T1 biopsy. Half of the muscle sample was used for immunohistochemistry (data not reported) and the other half for gel electrophoresis. MHC analyses of MHC I, IIA, and IIX isoforms were performed on sodium dodecyl sulfate polyacrylamide gel electrophoresis as described by Andersen and Aagaard (2).
The performance tests were conducted in one session after the anthropometric measures were recorded, and these consisted of a standing vertical jump, a 10-s Wingate test, and a strength test. Before the tests, participants completed a 10-min warm-up on a cycle ergometer (Monark Cycle Ergometer, model 828E; Monark Exercise AB, Vansbro, Sweden) using a light workload to elicit approximately 75–85 rpm.
Standing vertical jump test.
Participants performed three squat jumps with no arm countermovement, with 3 min of recovery between each jump. Before testing, a reflective marker was placed on the right greater trochanter (hip) of the participant. Participants were instructed to jump as high as they could (with hands on hips to prevent arm movement) and to prevent their knees from bending while in midair. One practice jump was allowed before the test jumps were recorded. The height of the standing vertical jump was measured using SIMI motion analysis system (SIMI Reality Motion Systems, Postfach 1518, 85705 Unterschleissheim, Germany) with a Canon MVX200i Digital Video Camcorder (Canon Europa N.V., Amstelveen, The Netherlands) at 100 Hz. The vertical distance jumped was measured from the hip marker at the starting (standing) position to the hip marker at the highest point of displacement. An average of the three trials was calculated for statistical comparison.
Wingate (10-s maximal power cycling test).
Participants completed two 10-s maximal effort Wingate tests on a modified mechanically braked ergometer (Monark Cycle Ergometer, model 828E, Monark Exercise AB) with a load of 75 g·kg−1 body mass. Participants performed two tests with 5 min of recovery between each test. Peak power output was recorded using Wingate Computer Software (School of Physical Education, University of Otago, Dunedin, New Zealand) and averaged for statistical comparison.
The strength test involved a 3RM. Three-repetition maximum testing was performed on all four lower body exercises, bilaterally, which were used for resistance training. For T1 measurements, a warm-up set of 10 repetitions was completed by each participant at a weight they thought they could complete. The weight was increased conservatively to complete five repetitions. Thereafter, the load was increased incrementally (2.5 kg) until a maximal load was obtained. This was determined as the weight that could be successfully lifted with correct technique for three, but not four, complete repetitions. In weeks 4 and 6 of the training program, as well as at T2 and T3, 3RM testing followed a protocol on the basis of the latest 3RM values completed: 10 repetitions of 50% 3RM, 5 repetitions of 75% 3RM, and 3 repetitions of 90% 3RM. Thereafter, the load was increased incrementally again to determine the new 3RM value.
Calculations and Statistical Analyses
Results are expressed as means ± SD. The level of significance was set at P ≤ 0.05. Area under the curve (AUC) for plasma glucose and insulin concentration (based on the OGTT) were computed by MatLab® (version 2009a, The MathWorks™ Inc., Natick, MA) via a trapezoid method, using fasting concentrations as baseline. Values that fell below the fasting concentration were not included in the AUC. Insulin sensitivity was determined by using the homeostasis model of insulin resistance (HOMA-IR), which was calculated from the formula: HOMA-IR = fasting glucose (mmol·L−1) × fasting insulin (μU·mL−1) / 22.5 (21).
Baseline comparisons between groups for anthropometric measures, performance, MHC isoform distribution, and fasting glucose and insulin concentrations, HOMA-IR and insulin AUC, as well as pretraining to posttraining differences in physical activity were determined using two-tailed Student’s t-tests. Two mixed model analyses (SPSS 19.0 for Windows; SPSS Inc., Chicago, IL) were conducted, one for the effects of training and one for detraining. Each had four factors: condition (two levels, C and FIR), time point (two levels, either T1 and T2 or T2 and T3), sex (two levels, male and female), and participant (16 levels, random effect factor). These were conducted for anthropometric measures, vertical distance jumped, peak power, 3RM strength values, fasting glucose and insulin concentration, HOMA-IR, glucose AUC and insulin AUC, MHC I, MHC IIA, and MHC IIX. When sex effects were significant (i.e., anthropometric and performance measures), the analyses were rerun using sex as a covariate. When sex effects were not significant, the analyses were rerun with sex not included. At time point T3, there was inadequate muscle tissue to conduct MHC analyses in two control muscle biopsy samples, so statistical analysis and results are based on n = 8 for controls. Pearson correlations were computed between MHC isoform composition and blood parameter measures.
Most of the participants (94%) were involved in recreational physical activity (e.g., social sport and commuting) and exercised 3.2 ± 1.6 times per week for 1.1 ± 0.5 h per session before recruitment into the study. There were no baseline differences observed between groups in the intensity, frequency, or duration of physical activity (Table 1). There were no differences (P > 0.050) in anthropometric measures, peak power or vertical distance jumped (Table 1), strength (Fig. 1), or MHC composition (Fig. 2) between FIR and C at baseline, nor were there significant differences in HOMA-IR (P = 0.28), fasting glucose (P = 0.07), fasting insulin (P = 0.37) (Table 2), glucose AUC (P = 0.38), or insulin AUC (P = 0.12) at baseline between FIR and C (Fig. 3). There was a large degree of variability in HOMA-IR (T1 range, C: 0.62–3.06, FIR: 0.94–3.28) and fasting insulin (T1 range, C: 20.8–97.6, FIR: 33.2–102.6) (Fig. 4). It is noteworthy that the clinical (75th percentile) cut-off for HOMA-IR is 2.6 and for fasting insulin is 83 pmol·L−1 (5).
Training Effects (T1–T2)
There was no significant main effect of training (P = 0.724) or condition (P = 0.072) on fasting glucose concentration (Table 2). There was also no significant main effect of training (P = 0.434) or condition (P = 0.292) on fasting insulin concentration (Table 2). There was also no main effect of training (P = 0.814) or condition (P = 0.925) on glucose AUC (Fig. 3A). There was, however, a trend for a main effect of training to decrease insulin AUC (P = 0.054) and a trend (P =0.052) for a main effect of condition, with FIR tending to have a greater insulin AUC (Fig. 3B). There was also a significant interaction of condition × time (P = 0.050), that is, with training, FIR reduced insulin AUC to a greater extent than C. There were neither significant effects of training (P = 0.496) nor condition (P = 0.464) on HOMA-IR (Table 2).
No significant main effects of training on the proportion of MHC I (P = 0.153), MHC IIA (P = 0.144), or MHC IIX (P = 0.369) were observed (Fig. 2). There were also no main condition effects on MHC isoform distribution after training (P > 0.050). There was a negative correlation between OGTT glucose AUC and MHC I content after training, that is, the greater the glucose AUC, the lower the proportion of MHC I (R2 = 0.268, P = 0.040). There was a positive correlation between fasting blood insulin concentration and MHC IIA content after training (P = 0.046), that is, the greater the fasting blood insulin concentration, the greater the amount of MHC IIA (R2 =0.254). Glucose AUC after training was also positively correlated with MHC IIX content (R2 = 0.264, P = 0.041).
Anthropometric measures did not change in response to training (P > 0.05), except for thigh girth, which increased by 1.2 cm (P ≤ 0.001, Table 2). There were no differences between FIR and C in anthropometric measures in response to training.
There was a significant main effect of training to increase vertical distance jumped (P = 0.001); however, there was no significant (P = 0.342) main effect of condition (Table 2). There was a significant (P = 0.005) main effect of training to improve peak power, with no significant (P = 0.182) main effect of condition (Table 2). There was a significant main effect of training on 3RM with increases in the leg press, leg extension, leg curl, and squat exercises (all P ≤ 0.001, Fig. 1). No significant main effects of condition on leg press, leg extension, leg curl, or squat exercises were observed (P = 0.458, P = 0.125, P = 0.154, and P = 0.192, respectively; Fig. 1)
Detraining Effects (T2–T3)
The mean duration of detraining until retesting was 9.2 wk (64.5 d; range, 61–67 d). All participants returned to similar baseline physical activity levels as before the beginning of the study, determined by physical activity records (P = 0.140).
There were no significant main effects of detraining or condition on fasting glucose concentration (P = 0.240 and P = 0.236, respectively; Table 2). There were also no significant main effects of detraining or condition on fasting insulin concentration (P = 0.877 and P = 0.163, respectively; Table 2). There were also no significant main effects of detraining or condition on glucose AUC (P = 0.817 and P = 0.940, respectively, Fig. 3A). However, there were significant main effects of detraining and condition on insulin AUC (P = 0.031 and P = 0.018, respectively; Fig. 3B). There was also a significant condition × time interaction (P = 0.023), that is, insulin AUC increased in FIR but not in C as a result of detraining (Fig. 3B). There were no significant main effects of detraining or condition on HOMA-IR (P = 0.900 and P = 0.232, respectively; Table 2).
There were no significant main effects of detraining on MHC I or MHC IIA isoform distributions (P = 0.213 and P = 0.100, respectively); however, there was a significant main effect of detraining to increase percentage MHC IIX (P = 0.026, Fig. 2). There were no significant main effects of condition on MHC I (P = 0.854), MHC IIA (P = 0.310), or MHC IIX (P = 0.535) proportions. An inverse correlation between fasting blood glucose concentration and MHC IIX was observed at T3, that is, the greater the fasting blood glucose concentration, the less the MHC IIX content after detraining (R2 = 0.345, P = 0.027).
There were no differences in anthropometric measures between FIR and C as a result of detraining (Table 1). The body mass, percentage body fat, or thigh girth did not change with detraining (P = 0.359, P = 0.075, and P = 0.078, respectively). There was a trend to increase waist/hip ratio with detraining (P = 0.052).
There was no significant main effect of detraining on vertical distance jumped (P = 0.594), nor was there a significant main effect of condition (P = 0.342) (Table 2). There was also no main effect of detraining on peak power (P = 0.334), with no significant main effect of condition (P = 0.174) (Table 2). There was a significant main effect of detraining to decrease 3RM in the leg press (P = 0.002), leg extension (P < 0.001), leg curl (P < 0.001), and squat exercises (P = 0.001, Fig. 1). No significant main effects of condition for leg press, leg extension, leg curl, or squat exercises were observed (P = 0.481, P = 0.159, P = 0.168, and P = 0.227, respectively).
A key finding of the current study is a rise in insulin AUC when abstaining from resistance training, after a period of training, in those with a familial link to T2DM, despite continuation of other activities. It also appears that resistance training elicits a greater, and metabolically significant, response in reducing insulin AUC in those with a familial link to T2DM compared with those without this link. Confirming our hypotheses, these findings demonstrate that individuals with parents of T2DM may be more responsive to this exercise mode and may be more metabolically sensitive to detraining than C. This highlights the effect of a resistance exercise intervention and importance of maintenance, particularly in individuals who may have a genetic predisposition to T2DM.
A reduction in insulin AUC without altered glucose response has been observed by others (13,23). In those studies, a 75- to 100-g OGTT was used to assess changes in insulin response before and after a period of 8–10 wk of resistance training in type 2 diabetic participants (13) and healthy individuals (23). Insulin AUC decreased by 10% in those with T2DM (13), 18.9% in healthy individuals (23). In the current study, the mean reduction in insulin AUC as a result of training just failed to reach significance (P = 0.054). The time of sampling, 72 h after exercise, was selected to avoid the acute effects of the last exercise bout. This interval may explain the lack of a significant change in insulin AUC with training, relative to other studies in which sampling was conducted 48 h after exercise (13,23). Although we were looking for chronic rather than acute effects of exercise, when designing this type of study, one may want to consider the timing of the measurement with respect to how frequently the individuals in question typically train and alter sampling time accordingly. The apparent decrease in training effect on insulin response with increasing time since the last exercise bout indicates that the frequency of exercise is important and suggests that to optimize the training response, exercise may need to be conducted at least every other day.
The majority of studies in which comparable measures (i.e., an OGTT) were taken have had detraining periods (returning to normal activities of daily living) of 15 d or less (6,17,22,26) and consisted of the removal of the training (training cessation) (6,17,22) or bed rest (26), with the exception of Andersen et al. (3) who used a detraining period (training cessation) of 90 d. Despite no changes in HOMA-IR with detraining, a significant difference between our two groups in the insulin response to an oral glucose load was observed. This difference in insulin response was evident despite no differences in percentage body fat, thigh girth (indicating muscle mass), or waist/hip ratio. It can therefore be concluded that cessation of resistance training of this length is sufficient to increase the insulin response to glucose ingestion in those who have a familial link to T2DM, even with continuation of other forms of physical activity.
Although no baseline differences in glucose or insulin measures were observed, it is plausible that the differential response of insulin to a glucose load with training and detraining may be related to initial differences in insulin responsiveness but, because of low statistical power, were not detected. Nevertheless, the observed differences in insulin AUC in those with a familial link to type 2 diabetes indicate that this group may be particularly advantaged by undertaking a resistance training regimen and particularly disadvantaged if the training stimulus is removed.
Our results contrast with an earlier finding of an increased number of Type IIB fibers in the offspring of those with T2DM (24). It has been observed in rodents that Type II fiber proportion increases when exposed to insulin (18); therefore, it is likely that among our FIR participants, the metabolic environment for this transition was not sufficient to elicit differences in MHC IIX at baseline; this is supported by no observable large differences between groups in baseline fasting insulin or glucose.
With resistance training, we expected an increase in MHC IIA with a corresponding decrease in MHC IIX; however, this was not observed. This is in contrast with results from the resistance training literature in those without T2DM (1,2,4,27). It is possible that in our participants, prior alternative training had already resulted in this shift or that the duration of the training period was not long enough to result in these changes. Typically, these changes are observed after 12 wk of training (35–45 sessions) (2), although some have seen directional changes at 8 wk (8). The high individual variability and low power would also have prevented detection of small changes.
Typically, if a period of detraining occurs after training, MHC IIX expression is increased (2,4). In the current study, this increase in IIX was observed with detraining, despite no change in the opposite direction previously with training. It may also have been that we missed a possible change in fiber type to more IIA in between when the training ended and at the end of detraining. It is unknown what happened in between these two time points, and adaptation to the training may still have been taking place into recovery. Most of our subjects admitted to having muscle soreness from a previous session when showing up for each subsequent training. This supports the idea that adaptation may not have fully occurred until the early part of the detraining period. The fact that our subjects trained three times per week (in contrast to most other studies in which subjects trained twice per week) may explain the soreness and possible delayed adaptation.
The effect of “downtime” on fiber type adaptations, and particularly MHC IIX overshoot, has ramifications, particularly for the training athlete, and may be positive or negative depending on whether the sport is endurance or strength/speed based.
GLUT-4 expression is less in fast (Type II) muscle fibers than that in slow (Type I) fibers (15). Therefore, Type II fibers have a reduced capacity to take up glucose than Type I fibers. This may account for the positive correlation observed between fasting blood insulin and MHC IIA content after training and the negative correlation between glucose AUC and MCH1 content.
Nine weeks of resistance training improved strength and power in both groups.
We can confirm that the physical activity completed during detraining was not different from baseline, which is also confirmed by the decrease in strength in all 3RM tests. However, despite decreases in strength, participants were able to maintain power after 9 wk of detraining. Maintenance of power in this circumstance could have resulted from increased proportion of MHC IIX fibers observed as a result of detraining.
This study was limited by small numbers, particularly of FIR for which there were strict criteria. Another limitation is the lack of a control group, making it difficult to ascertain if there was a seasonal effect. Furthermore, training intervention studies of this duration are notoriously difficult in terms of recruitment as well as maintenance. That all except one participant completed the intervention is positive; unfortunately, the one who dropped out was FIR.
Because of the small numbers and large variability, the lack of significance, particularly for fasting insulin and HOMA-IR, may be related to low power. On the basis of n = 6 (FIR), the power to detect a training effect in fasting insulin was 0.07, and for n = 10 (C), the power was 0.10, with α = 0.05 and mean SD of 25.9. Even if we had increased to n = 20, the power would only have been 0.16. For HOMA-IR for n = 6, a power of 0.07 was obtained, and for n = 10, a power of 0.09 was obtained, with α = 0.05 and mean SD of 0.88.
Thus, increasing subject numbers may not have altered the result to a trend for a significant difference, unless variability within each group changed. The control group had similar baseline variability as FIR, including some subjects who were beyond the clinical cut-offs for fasting insulin and HOMA-IR. It appears that fasting values, and the calculated indices derived from them, do not have the sensitivity that dynamic responses to a glucose load do.
In summary, this is the first study to measure metabolic adaptations, along with MHC isoform distributions, in young adult individuals (with and without a familial link to T2DM) before and after 9 wk of resistance training and 9 wk of detraining. Insulin AUC decreased considerably with training, and increased in response to detraining, in those with a possible genetic link to T2DM. Clinically, this highlights the positive effect a resistance training program may have and the quick return to pretraining state when training is discontinued. A moderate level of resistance training may play a role in the prevention of T2DM, particularly in those individuals who have a familial link and may be more at risk of developing T2DM.
Funding for insulin and glucose analyses was provided by the Sports Nutrition and Exercise Metabolism Research Group Discretionary Fund, Department of Human Nutrition, Otago University.
We are very appreciative of the dedication of our participants during training and testing. We acknowledge the assistance of Margaret Waldron (University of Otago, New Zealand) with cannulations, Dr. David Gerrard (University of Otago, New Zealand) for assisting with muscle biopsy sampling, and Brian Niven for statistical analysis advice.
No conflict of interest is declared for KL Schofield, NJ Rehrer, TL Perry, A Ross, JL Andersen, and H Osborne.
The results of the present study do not constitute endorsement by the American College of Sports Medicine.
1. Adams GR, Hather BM, Baldwin KM, Dudley GA. Skeletal muscle myosin heavy chain composition and resistance training. J Appl Physiol
. 1993; 74 (2): 911–5.
2. Andersen JL, Aagaard P. Myosin heavy chain IIX overshoot in human skeletal muscle. Muscle Nerve
. 2000; 23: 1095–104.
3. Andersen JL, Schjerling P, Andersen LL, Dela F. Resistance training and insulin action in humans: effects of detraining. J Physiol
. 2003; 551 (3): 1049–58.
4. Andersen LL, Andersen JL, Magnusson SP, et al.. Changes in the human muscle force–velocity relationship in response to resistane training and subsequent detraining. J Appl Physiol
. 2005; 99: 87–94.
5. Ascaso JF, Pardo S, Real JT, Lorente RI, Priego A, Carmena R. Diagnosing insulin resistance by simple quantitative methods in subjects with normal glucose metabolism. Diabetes Care
. 2003; 26 (12): 3320–5.
6. Bajpeyi S, Tanner CJ, Slentz CA, et al.. Effect of exercise intensity and volume on perstistence of insulin sensitivity during training cessation. J Appl Physiol
. 2009; 106: 1079–85.
7. Baldi JC, Snowling N. Resistance training improves glycaemic control in obese type 2 diabetic men. Int J Med Sci
. 2003; 24 (6): 419–23.
8. Campos G, Luecke T, Wendeln H, et al.. Muscular adaptations in response to three different resistance-training regimens: specificity of repetition maximum training zones. Eur J Appl Physiol
. 2002; 88 (1): 50–60.
9. Cederholm T, Sylvén C, Esbjörnsson-Liljedahl M, Jansson E. Insulin treatment increases skeletal muscle fibre area in patients with diabetes mellitus type 2. Clin Physiol
. 2000; 20 (5): 354–9.
10. Charette S, McEvoy L, Pyka G, et al.. Muscle hypertrophy response to resistance training in older women. J Appl Physiol
. 1991; 70 (5): 1912–6.
11. Chou SW, Lai CH, Hsu TH, et al.. Characteristics of glycaemic control in elite power and endurance athletes. Prev Med
. 2005; 40: 564–9.
12. DeFronzo RA, Ferrannini E. Insulin resistance. A multifaceted syndrome responsible for NIDDM, obesity, hypertension, dyslipidemia, and atherosclerotic cardiovascular disease. Diabetes Care
. 1991; 14 (3): 173–94.
13. Dunstan DW, Puddey IB, Beilin LJ, Burke V, Morton AR, Stanton KG. Effects of a short-term circuit weight training programe on glycaemic control in NIDDM. Diabetes Res Clin Pract
. 1998; 40: 53–61.
14. Fenicchia LM, Kanaley J, Azevedo JL Jr, et al.. Influence of resistance exercise training on glucose control in women with type 2 diabetes. Metabolism
. 2004; 53 (3): 284–9.
15. Gaster M, Poulsen P, Handberg A, Schrøder HD, Beck-Nielsen H. Direct evidence of fiber type–dependent GLUT-4 expression in human skeletal muscle. Am J Physiol Endocrinol Metab
. 2000; 278 (5): E910–6.
16. Gaster MM, Staehr PP, Beck-Nielsen HH, Schrøder HDH, Handberg AA. GLUT4 is reduced in slow muscle fibers of type 2 diabetic patients: is insulin resistance in type 2 diabetes a slow, type 1 fiber disease? Diabetes
. 2001; 50 (6): 1324–9.
17. Heath GW, Gavin JR III, Hinderliter JM, Hagberg JM, Bloomfield SA, Holloszy JO. Effects of exercise and lack of exercise on glucose tolerance and insulin sensitivity. J Appl Physiol
. 1983; 55 (2): 512–7.
18. Holmang A, Brezinska Z, Bjorntorp P. Effect of hyperinsulinemia on muscle fibre composition and capillarization in rats. Diabetes
. 1993; 42 (7): 1073–9.
19. Houmard JA, O’Neill DS, Zheng D, Hickey MS, Dohm GL. Impact of hyperinsulinemia on myosin heavy chain gene regulation. J Appl Physiol
. 1999; 86 (6): 1828–32.
20. King DS, Dalsky GP, Clutter WE, et al.. Effects of exercise and lack of exercise on insulin sensitivity and responsiveness. J Appl Physiol
. 1988; 64 (5): 1942–6.
21. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and β-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia
. 1985; 28 (7): 412–9.
22. Mikines KJ, Sonne B, Tronier B, Galbo H. Effects of training and detraining on dose–response relationship between glucose and insulin secretion. Am J Physiol Endocrinol Metab
. 1989; 256 (5): E588–96.
23. Miller WJ, Sherman WM, Ivy JL. Effect of strength training on glucose tolerance and post-glucose insulin response. Med Sci Sports Exerc
. 1984; 16 (6): 539–43.
24. Nyholm B, Qu Z, Kaal A, et al.. Evidence of an increased number of type IIb muscle fibres in insulin-resistant first-degree relatives of patients with NIDDM. Diabetes
. 1997; 46 (11): 1822–8.
25. Oberbach A, Bossenz Y, Lehmann S, et al.. Altered fibre distribution and fibre-specific glycolytic and oxidative enzyme activity in skeletal muscle of patients with type 2 diabetes. Diabetes Care
. 2006; 29 (4): 895–900.
26. Sonne MP, Alibegovic AC, Hojbjerre L, Vaag A, Stalknecht B, Dela F. Effect of 10 days of bedrest on metabolic and vascular insulin action: a study in individuals at risk of for type 2 diabetes. J Appl Physiol
. 2010; 108: 830–7.
27. Staron R, Malicky E, Leonardi M, Falkel J, Hagerman F, Dudley G. Muscle hypertrophy and fast fiber type conversions in heavy resistance–trained women. Eur J Appl Physiol Occup Physiol
. 1990; 60 (1): 71–9.
28. Staron RS, Leonardi MJ, Karapondo DL, et al.. Strength and skeletal muscle adaptations in heavy-resistance–trained women after detraining and retraining. J Appl Physiol
. 1991; 70 (2): 631–40.
29. Wang Y, Simar D, Fiatarone Singh M. Adaptations to exercise training within skeletal muscle in adults with type 2 diabetes or impaired glucose tolerance: a systematic review. Diabetes Metab Rev
. 2009; 25: 13–40.
30. World Health Organization. Definition, Diagnosis and Classification of Diabetes Mellitus and Its Complications. Part 1: Diagnosis and Classification of Diabetes Mellitus
. Geneva: World Health Organization; 1999. p. 52.