J Appl Physiol 111: 1554–1560, 2011.

First published August 25, 2011; doi:10.1152/japplphysiol.00921.2011.

Low-volume high-intensity interval training reduces hyperglycemia and

increases muscle mitochondrial capacity in patients with type 2 diabetes
Jonathan P. Little,1 Jenna B. Gillen,1 Michael E. Percival,1 Adeel Safdar,1,2 Mark A. Tarnopolsky,2
Zubin Punthakee,2 Mary E. Jung,3 and Martin J. Gibala1
1
Departments of Kinesiology and of 2Pediatrics and Medicine, McMaster University, Hamilton, Ontario; and 3School
of Health and Exercise Sciences, University of British Columbia Okanagan, Kelowna, British Columbia, Canada
Submitted 22 July 2011; accepted in final form 24 August 2011

Little JP, Gillen JB, Percival ME, Safdar A, Tarnopolsky MA, exercise training in T2D. The utility of HIT for improving
Punthakee Z, Jung ME, Gibala MJ. Low-volume high-intensity disease outcomes has been demonstrated in patients with met-
interval training reduces hyperglycemia and increases muscle mito- abolic syndrome, heart failure, and chronic obstructive pulmo-
chondrial capacity in patients with type 2 diabetes. J Appl Phys- nary disease (reviewed in Ref. 7). However, the potential
iol 111: 1554 –1560, 2011. First published August 25, 2011;
benefits of HIT on disease parameters in T2D have yet to be
doi:10.1152/japplphysiol.00921.2011.—Low-volume high-intensity
interval training (HIT) is emerging as a time-efficient exercise strat- established.
egy for improving health and fitness. This form of exercise has not We (8, 14, 16) and others (1, 21) have shown that HIT elicits
been tested in type 2 diabetes and thus we examined the effects of physiological remodeling comparable to moderate-intensity
low-volume HIT on glucose regulation and skeletal muscle metabolic continuous training in healthy adults, despite a substantially
capacity in patients with type 2 diabetes. Eight patients with type 2 lower time commitment and reduced total exercise volume. As
diabetes (63 ⫾ 8 yr, body mass index 32 ⫾ 6 kg/m2, HbA1C 6.9 ⫾ little as six sessions of low-volume HIT over 2 wk increases
0.7%) volunteered to participate in this study. Participants performed skeletal muscle mitochondrial capacity (8), which may be of
six sessions of HIT (10 ⫻ 60-s cycling bouts eliciting ⬃90% maximal clinical relevance for T2D given that reduced content (22) or
heart rate, interspersed with 60 s rest) over 2 wk. Before training and biogenesis (18) of mitochondria have been implicated in insu-
from ⬃48 to 72 h after the last training bout, glucose regulation was
lin resistance and T2D. Two weeks of low-volume HIT has
assessed using 24-h continuous glucose monitoring under standard-
ized dietary conditions. Markers of skeletal muscle metabolic capacity also been shown to improve glucose tolerance (1) and enhance
were measured in biopsy samples (vastus lateralis) before and after insulin sensitivity (21) in healthy adults. These findings are
(72 h) training. Average 24-h blood glucose concentration was re- intriguing because they suggest that low-volume HIT may
duced after training (7.6 ⫾ 1.0 vs. 6.6 ⫾ 0.7 mmol/l) as was the sum result in many of the same health benefits as traditional
of the 3-h postprandial areas under the glucose curve for breakfast, exercise training with substantially reduced exercise volume
lunch, and dinner (both P ⬍ 0.05). Training increased muscle mito- and time commitment. Low-volume HIT may, therefore, rep-
chondrial capacity as evidenced by higher citrate synthase maximal resent a potent, time-efficient exercise strategy to improve
activity (⬃20%) and protein content of Complex II 70 kDa subunit skeletal muscle metabolic control and glycemic regulation in
(⬃37%), Complex III Core 2 protein (⬃51%), and Complex IV patients with T2D.
subunit IV (⬃68%, all P ⬍ 0.05). Mitofusin 2 (⬃71%) and GLUT4
The primary purpose of this pilot investigation was to
(⬃369%) protein content were also higher after training (both P ⬍
0.05). Our findings indicate that low-volume HIT can rapidly improve examine the effects of low-volume HIT on glucose regulation
glucose control and induce adaptations in skeletal muscle that are and skeletal muscle metabolic capacity in individuals with
linked to improved metabolic health in patients with type 2 diabetes. T2D. On the basis of accumulating evidence indicating that
postprandial hyperglycemia plays a predominant contributing
exercise; continuous glucose monitoring; mitochondria; GLUT4; gly- role in diabetic complications (5), we used continuous glucose
cemic control
monitoring (CGM) to examine the effects of HIT on overall
glycemic exposure and postprandial glucose fluctuations. We
REGULAR EXERCISE IS AN EFFECTIVE strategy for the prevention hypothesized that 2 wk of low-volume HIT would reduce
and treatment of type 2 diabetes (T2D; 6). Most studies hyperglycemia and increase mitochondrial capacity and
examining the therapeutic effects of exercise in T2D involve GLUT4 content measured in skeletal muscle biopsy samples.
continuous, low- to moderate-intensity exercise such as walk-
ing, jogging, or cycling for ⱖ30 min/session (reviewed in Ref. METHODS
6). Although the optimal strategy has not been established,
Participants
higher intensity exercise may be more effective for improving
glycemic control in patients with T2D (2, 25). Recently revised Participants were recruited through local diabetes clinics, commu-
guidelines from the American Diabetes Association advocate nity diabetes information sessions, and poster advertisement. All
at least 150 min of moderate to vigorous exercise per week (6). participants were diagnosed with T2D at least 3 mo prior by a
High-intensity interval training (HIT), which involves repeated clinician according to standard criteria, including a fasting glucose
bursts of vigorous exercise interspersed with periods of rest, ⱖ7.0 mmol/l and/or 2-h oral glucose tolerance test blood glucose
may be an attractive option to implement higher intensity concentration ⱖ11.1 mmol/l, were not taking insulin, and had no
history of end-stage liver or kidney disease, neuropathy, retinopathy,
hypertension that could not be controlled by standard medication,
Address for reprint requests and other correspondence: M. J. Gibala, Dept. cardiovascular disease, or other contraindication to exercise. Eight
of Kinesiology, Ivor Wynne Centre, Rm 219, McMaster Univ., 1280 Main St. individuals [mean age 62.5 ⫾ 7.6 yr, body mass index 31.7 ⫾ 5.8
West, Hamilton, ON L8S 4K1, Canada (e-mail: gibalam@mcmaster.ca). kg/m2, hemoglobin A1C (HbA1C) 6.9 ⫾ 0.7% (range 6.4 – 8.5%)]
1554 8750-7587/11 Copyright © 2011 the American Physiological Society http://www.jap.org
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (125.166.116.202) on November 2, 2018.
Copyright © 2011 American Physiological Society. All rights reserved.
 INTERVAL TRAINING IN TYPE 2 DIABETES 1555
volunteered to participate in this study. Six participants were seden- within a total time commitment of 75 min/wk, including warm-up,
tary, which was defined as less than or equal to two exercise sessions cool-down, and the recovery interval between high-intensity efforts.
of 30 min/wk. Two participants reported engaging in ⬃30 min of Posttesting. CGM data were collected for a 24-h period starting
low-intensity walking exercise on 3–5 days/wk, in accordance with ⬃48 h after the final training session. Diet was controlled to be the
guidelines provided from their diabetes care team. Six subjects were same as pretraining. Resting muscle biopsy samples were obtained
taking blood glucose lowering medications but had HbA1C values ⬃72 h following the final training session. Approximately 2– 4 days
ⱕ8.5% and were not on exogenous insulin therapy. Four patients were after the biopsy, the walk test (performed at the same speed as
treated with metformin only, one patient with gliclazide only, and one pretraining) and maximal exercise test were performed using the same
patient with a combination of metformin, pioglitazone, sitagliptin, and procedures as baseline testing. Perceived enjoyment of low-volume
repaglinide. Due to the short duration of the intervention, participants HIT was assessed by asking participants how enjoyable they would
did not adjust their medications and were instructed to maintain their find engaging in 1) a single bout of HIT (10 ⫻ 1 min) and 2) HIT at
typical dietary and activity patterns throughout. All participants pro- least 3 times/wk for the next 4 wk using a 9-point Likert scale ranging
vided written informed consent. The study protocol was approved by from 1 (not enjoyable at all) to 9 (very enjoyable).
the Hamilton Health Sciences/McMaster University Faculty of Health
Sciences Research Ethics Board. Continuous Glucose Monitoring
Average blood glucose concentration and area under the glucose
Experimental Design curve were calculated from CGM data for a 24-h period before and
after training. CGM data were also used to analyze the 3 h postpran-
The experimental design consisted of 1) medical clearance and dial areas under the glucose curve for breakfast, lunch, and dinner. On
familiarization, 2) baseline testing, 3) a 2-wk training intervention, CGM data collection days, participants consumed their habitual
and 4) posttesting. breakfast but were provided with standardized snacks (e.g., almonds,
Medical clearance and familiarization. Height and weight were fruit, vegetables) based on their personal preferences and habits.
recorded, and a maximal exercise test on a recumbent cycle ergometer Lunch and dinner were standardized for all participants by providing
(Corival, Lode BV, Groningen, The Netherlands) was performed with vouchers to a local sandwich restaurant. Subjects completed detailed
pre- and postexercise 12-lead electrocardiogram (ECG) collection to dietary logs to record the timing and quantity of all food consumed
confirm the absence of any underlying contraindications to vigorous during the pretraining day. For posttesting, these dietary logs, along
exercise participation. The test started at 30 W and increased by 15 with all snacks and vouchers for lunch and dinner were provided with
W/min until volitional exhaustion. Peak power output (Wmax) and instructions for diet replication under free-living conditions over the
maximal heart rate (HRmax) were recorded. Following ECG clearance 24-h CGM data collection period. As per manufacturer’s recommen-
by a study physician, participants completed one to two familiariza- dations, capillary blood glucose samples were obtained at four points
tion sessions to become acquainted with low-volume HIT. These during the day at a time when blood glucose would be expected to be
sessions were also used to determine the interval power output that stable (i.e., upon awakening, before lunch, before dinner, and before
elicited ⬃90% HRmax. bed) and were automatically stored in the glucose meters provided to
Baseline testing. Prior to training, participants performed a 15-min participants. These four values were used during CGM downloading
walking test to examine the cardiovascular response and ratings of to construct 24-h blood glucose curves based on interstitial glucose
perceived exertion (RPE) during exercise. Speed was self-selected by recordings averaged every 5 min by the CGM device using the
each participant during an initial 5-min warm-up. Heart rate was associated software algorithm (Solutions Software, Medtronic,
measured by telemetry (Polar), and RPE was measured using the 0 –10 Northridge, CA). CGM data were exported and analyzed using Sigma-
continuous/interval scale. Plot (Statsoft, Chicago, IL). Reproducibility of the CGM device in our
At least 2 days after the walk test, participants reported to the lab was verified in five volunteers who wore the monitor on two
laboratory for CGM device insertion (CGMS iPro, Medtronic, occasions separated by 1 wk under identical dietary conditions. The
Northridge, CA). Participants were given a glucose meter (OneTouch coefficient of variation for the 24 h blood glucose measurements was
UltraMini, Lifescan, Milpitas, CA) with instructions for both calibra- 2.8% (data not shown).
tion and capillary blood sampling and individualized control diets.
The following day served as a dietary control day for 24-h CGM data Muscle Analyses
collection. Subjects returned to the laboratory 2 days later for removal Citrate synthase enzyme activity. One piece of muscle (⬃20 mg)
of the CGM device and collection of a resting skeletal muscle biopsy was homogenized using a glass tissue grinder (Kimble/Kontes
sample as we previously described (8). Briefly, muscle samples were 885300 – 0002) in 10 volumes of buffer containing (in mM) 70
obtained under local anesthesia (1% Lidocaine) from the vastus sucrose, 220 mannitol, 10 HEPES (pH 7.4) supplemented with pro-
lateralis using a Bergstrom needle adapted with suction. Muscle tease inhibitors (Complete Mini, Roche Applied Science, Laval, PQ,
samples were quickly blotted to remove excess blood, sectioned into Canada) and used to determine the maximal activity of citrate syn-
several pieces, and placed in separate vials before snap freezing in thase (CS) as we previously described (8, 16). Protein concentration
liquid nitrogen for subsequent analyses. of homogenates was determined using a commercial assay (BCA
Training. Approximately 5 days after the muscle biopsy procedure, Protein Assay, Pierce, Rockford, IL), and enzyme activity is ex-
subjects commenced training. The HIT protocol involved a total of six pressed as millimoles per kilogram of protein per hour wet weight.
supervised sessions over 2 wk (Monday, Wednesday, Friday each Western blotting. A second piece of muscle (⬃30 mg) was homog-
week). Each session consisted of 10 ⫻ 60-s cycling intervals inter- enized in RIPA buffer for Western blot analyses using techniques
spersed with 60 s of recovery based on our recent work (14). Training described previously (8, 16). Briefly, protein concentration of homog-
was performed on a cycle ergometer (LifeCycle C1 or R1, Life enates were determined as above and equal amounts of protein (5–20
Fitness, Schiller Park, IL) set in constant watt mode at a pedal cadence ␮g) were prepared in 4⫻ Laemmli’s buffer and heated to 95°C before
of 80 –100 revolutions/min. Individual workloads were selected to being separated by 10 –12.5% SDS-PAGE and electrotransferred to
elicit a heart rate of ⬃90% HRmax during the intervals. During nitrocellulose membranes. Ponceau S staining was performed follow-
recovery, participants were allowed to rest or pedal slowly against a ing transfer to visualize equal loading and transfer. Following 1 h
resistance of 50 W. Each training session included a 3-min warm-up blocking in 5% fat-free milk Tris-buffered saline 0.1% Tween 20
and 2-min cool-down at 50 W, for a total of 25 min. Therefore, the (TBS-T), membranes were incubated in primary antibodies overnight
training protocol involved a total of 30 min of high-intensity exercise at 4°C or at room temperature for 2 h in 3% fat-free milk TBS-T or

J Appl Physiol • VOL 111 • DECEMBER 2011 • www.jap.org

Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (125.166.116.202) on November 2, 2018.
Copyright © 2011 American Physiological Society. All rights reserved.
1556 INTERVAL TRAINING IN TYPE 2 DIABETES

0.01). Area under the 24-h blood glucose curve was also lower
following HIT (pre: 11,066 ⫾ 1,703 vs. post: 9,572 ⫾ 995
mmol·l⫺1·day⫺1, P ⫽ 0.02). The sum of the 3-h postprandial
area under the glucose curves for breakfast, lunch, and dinner
was significantly lower posttraining (pre: 965 ⫾ 483 vs. post:
679 ⫾ 437 mmol·l⫺1·9 h⫺1, P ⫽ 0.01). Pre- and posttraining
24-h blood glucose curves for a representative subject are
shown in Fig. 2B.
Adaptations in Skeletal Muscle
The maximal activity of CS was elevated following training
(Fig. 3A, P ⫽ 0.04). Training also increased skeletal muscle
mitochondrial protein content as evidenced by changes in
Complex II 70 kDa subunit (P ⫽ 0.03), Complex III Core 2
protein (P ⫽ 0.04), and COX subunit IV (P ⫽ 0.02) measured
by Western blotting (Fig. 3B). The protein content of CS
Fig. 1. Characterization of high-intensity interval training protocol. Training (⬃57%; data not shown), NDUFA9, COX subunit II (⬃53%;
intensity expressed as a percentage of peak workload (bars), peak heart rate data not shown), and ATP synthase ␣-subunit also increased,
(solid line), and rating of perceived exertion (dashed line) averaged across all but did not reach statistical significance (P ⫽ 0.06 – 0.12; Fig.
6 sessions for all subjects. Values are means ⫾ SD (N ⫽ 8). RPE, ratings of
perceived exertion.

3% BSA TBS-T depending on previously determined optimization

conditions. After 3 ⫻ 5-min washes in TBS-T, membranes were
incubated in the appropriate species-specific secondary antibody di-
luted (1:10,000) in 3% fat-free milk TBS-T for 1 h at room temper-
ature, washed in TBS-T for 3 ⫻ 15 min, and visualized by chemilu-
minescence (SuperSignal West Dura, Pierce) using a FluorChem SP
Imaging System (Alpha Innotech, San Leandro, CA). ImageJ software
(NIH) was used to quantify the optical density of protein bands.
␣-Tubulin (Cell Signaling Technology, #2125), which did not change
following training (P ⫽ 0.91), was used as a loading control. Primary
antibodies for the following proteins of interest were used: NDUFA9
(Mitosciences, MS111), Complex II 70 kDa subunit (Mitosciences,
MS204), Complex III Core 2 protein (Mitosciences, MS304), cyto-
chrome c oxidase (COX) subunit II (MitoSciences, MS405), COX
subunit IV (Mitosciences, MS408), ATP synthase ␣-subunit (Mito-
sciences, MS507), CS (kind gift from Dr. Brian Robinson, The
Hospital for Sick Children, Toronto, Canada), mitofusin (Mfn) 2
(Sigma, M6319), and GLUT4 (Millipore, AB1345).
Statistical Analyses
All data were analyzed using paired Student’s t-tests with signifi-
cance set at P ⱕ 0.05 (Sigma Stat v3.10). Values are means ⫾ SD in
the text and on figures.
RESULTS

Descriptive Characteristics of Training

All participants completed all prescribed intervals during
training with no complications. Interval intensity averaged
across all intervals for all subjects corresponded to 95 ⫾ 14%
of Wmax, elicited 88 ⫾ 3% HRmax, and RPE was 6.4 ⫾ 1.3
(0 –10 scale). The response to each interval averaged across all
six training sessions for all subjects is depicted in Fig. 1.
Training had no effect on body mass (pre: 93 ⫾ 19 kg vs. post:
92 ⫾ 18 kg, P ⫽ 0.28). On average, perceived enjoyment of
HIT was rated high by this group of participants (single Fig. 2. Two weeks of high-intensity interval training improves glycemic
session, 8.1 ⫾ 1.0; 3 times/wk, 7.9 ⫾ 1.0). control. A: average blood glucose concentration measured by continuous
glucose monitoring (CGM) over a 24-h period before (Pre) and after (Post) 2
Continuous Glucose Monitoring wk of training. B: blood glucose concentration assessed by CGM over 24 h
before (Pre; solid line) and after (Post; dashed line) training in a representative
Average blood glucose concentration over 24 h was reduced subject. Posttraining CGM data was collected from ⬃48 –72 h following the
from 7.6 ⫾ 1.0 to 6.6 ⫾ 0.7 mmol/l after training (Fig. 2A, P ⫽ final training session. Values are means ⫾ SD (N ⫽ 7). *P ⬍ 0.05.

J Appl Physiol • VOL 111 • DECEMBER 2011 • www.jap.org

Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (125.166.116.202) on November 2, 2018.
Copyright © 2011 American Physiological Society. All rights reserved.
 INTERVAL TRAINING IN TYPE 2 DIABETES 1557

Fig. 3. Summary of metabolic adaptations in

skeletal muscle following 2 wk of high-
intensity interval training. A: maximal activ-
ity of citrate synthase (CS) measured in
skeletal muscle biopsy samples before and
after 2 wk of training. B: protein content of
one subunit from each complex of the elec-
tron transport chain measured in skeletal
muscle biopsy samples obtained before and
after training. Representative Western blots
for each protein for two individual subjects
are also shown. Values are means ⫾ SD
(N ⫽ 7). *P ⬍ 0.05.

3B). The protein content of Mfn2 was elevated following post: 124 ⫾ 37 W, P ⫽ 0.03). Training reduced heart rate
training (⬃71%, P ⫽ 0.02; Fig. 4), as was total GLUT4 protein (pre: 73 ⫾ 7 vs. post: 66 ⫾ 6%HRmax, P ⬍ 0.001) and RPE
(⬃369%, P ⫽ 0.003; Fig. 5). (pre: 2.4 ⫾ 0.7 vs. post: 1.3 ⫾ 1.2, P ⫽ 0.01) during the
walk test.
Functional Exercise Performance
Maximal workload achieved on the ramp cycling test was
increased by ⬃10% following training (pre: 111 ⫾ 36 vs.

Fig. 5. Two weeks of high-intensity interval training increases GLUT4

Fig. 4. Two weeks of high-intensity interval training increases mitofusin 2 protein content. Glucose transporter 4 (GLUT4) protein content measured
(Mfn2) protein content. Mfn2 protein content measured in skeletal muscle in skeletal muscle biopsy samples obtained before and after training.
biopsy samples obtained before and after training. Representative Western Representative Western blots from 2 subjects. Values are means ⫾ SD
blots from 2 subjects are shown. Values are means ⫾ SD (N ⫽ 7). *P ⫽ 0.02. (N ⫽ 7). *P ⫽ 0.003.

J Appl Physiol • VOL 111 • DECEMBER 2011 • www.jap.org

Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (125.166.116.202) on November 2, 2018.
Copyright © 2011 American Physiological Society. All rights reserved.
1558 INTERVAL TRAINING IN TYPE 2 DIABETES

DISCUSSION The mechanisms mediating the improvement in glycemic

control following HIT remain to be determined. Training had
The present study demonstrates that low-volume HIT can
no effect on body mass, and, while not assessed directly, it is
rapidly reduce hyperglycemia and increase skeletal muscle
unlikely that such a short exercise intervention would lead to
oxidative capacity in patients with T2D. These improvements
any substantial changes in body composition. Therefore, it is
were realized despite a small total volume of exercise that
tempting to speculate that adaptations in skeletal muscle were
consisted of six training sessions over 2 wk. The training
involved. Since reduced mitochondrial capacity in skeletal
protocol involved a total of only 30 min of high-intensity
muscle has been reported in insulin resistance and T2D (22)
exercise and a total time commitment of only 75 min/wk. This
and muscle oxidative capacity has been shown to be a signif-
is much lower than current physical activity guidelines for T2D
icant predictor of insulin sensitivity (3), it is possible that the
that recommend a total of 150 min of moderate to vigorous
rapid increase in skeletal muscle mitochondrial content follow-
intensity exercise each week (6). Given that the majority of
ing low-volume HIT may be a contributing factor related to
individuals with and without T2D does not accumulate suffi-
reduced insulin resistance and improved glycemic control.
cient exercise to achieve health benefits (6) and the most
However, the notion that mitochondrial deficiency mediates
common cited barrier to regular exercise is lack of time (26),
insulin resistance has been questioned recently (12), indicating
our results suggest that low-volume HIT may be a viable,
time-efficient strategy to improve health in patients with T2D. that other adaptations in skeletal muscle may be more impor-
tant. The training-induced increase in GLUT4 protein content
Low-Volume HIT and Glycemic Control likely plays a role in improving glucose regulation. Studies in
rodents indicate that the exercise-induced increase in GLUT4
Glycemic control is an important aspect of T2D treatment protein is directly related to the increase in muscle glucose
and is an independent risk factor for the development of uptake at any given insulin concentration (20). Thus, even in
diabetic complications (24, 27). We used CGM to assess the the face of insulin resistance, an increase in skeletal muscle
effects of short-term low-volume HIT on overall glycemic GLUT4 could facilitate greater muscle glucose uptake and
exposure and postprandial glucose responses. CGM provides contribute to improved glycemic regulation. In addition to
information about direction, magnitude, and frequency of skeletal muscle adaptations, training-induced alterations in
blood glucose excursions and may provide a sensitive means to hepatic glucose output cannot be ruled out. The effect of
detect acute changes in blood glucose throughout the day (15). exercise training on hepatic insulin resistance in T2D has not
Although exercise is regarded as an effective strategy to improve been directly assessed in humans, although there is evidence to
glycemic control (2, 6, 25), there are limited data regarding the suggest that endurance exercise training improves hepatic in-
effect of exercise training on glucose control using CGM. Studies sulin signaling and glycemic control in rodents (10).
using CGM technology in patients with T2D have reported that We did not directly assess the effects of training on insulin
acute resistance exercise reduces the prevalence of hyperglycemia sensitivity using hyperinsulinemic-euglycemic clamps and there-
(19) and acute endurance exercise reduces 24-h average blood fore cannot conclude whether low-volume HIT improves muscle
glucose concentration (17). In the only training study conducted to insulin sensitivity. CGM assesses exposure to hyperglycemia as
date, Cauza et al. (4) reported a greater reduction in 24-h average well as glycemic excursions throughout the day. Exposure to
blood glucose concentration following 4 mo of resistance training hyperglycemia over time may be a better indicator of diabetic
compared with endurance-type training in individuals with T2D, complications than insulin sensitivity per se (5) and therefore
but interpretations on the effects of exercise are potentially limited CGM may provide greater insight into the clinical benefits of
by significant changes in body composition and an apparent lack exercise training. Changes in HbA1C are commonly used to assess
of dietary control. the effectiveness of glucose-lowering interventions in T2D but
To our knowledge, this is the first study to examine the effects due to the short duration of the current study was not measured.
of HIT on glycemic regulation using CGM. Average blood
glucose concentration and area under the glucose curve measured Low-Volume HIT and Skeletal Muscle Mitochondrial
under standardized dietary conditions from ⬃48 to 72 h after the Adaptations
final training session were significantly lower than pretraining,
indicating that short-term, low-volume HIT improved glycemic Individuals with insulin resistance and T2D have been
control, particularly glycemic excursions after meals. Although shown to have reduced mitochondrial content (22), impaired in
reducing fasting hyperglycemia is a significant aspect of T2D vivo mitochondrial function (23), and/or reduced markers of
treatment, increasing evidence suggests lowering postprandial mitochondrial biogenesis (18) in skeletal muscle. These find-
hyperglycemia is as important, if not more important, for achiev- ings have led to the hypothesis that reduced mitochondrial
ing targeted HbA1C levels (27). Additionally, elevated postmeal capacity or impaired regulation of mitochondrial biogenesis in
blood glucose excursions have been implicated in the develop- skeletal muscle may play a role in the pathogenesis of T2D (18,
ment and progression of T2D related comorbidities such as 22, 23). Although it is currently unclear whether skeletal
cardiovascular disease (5, 27). Following HIT, the sum of the muscle mitochondrial impairment causes insulin resistance
postprandial areas under the glucose curve for breakfast, lunch, (12), interventions that increase muscle mitochondrial content
and dinner was significantly lower than pretraining, highlighting may be effective for the treatment and prevention of T2D (9,
the potency of HIT to lower postmeal glucose excursions. These 13). Given the potency of low-volume HIT to induce mito-
findings demonstrate that low-volume HIT may be an effective chondrial biogenesis in young, healthy subjects (8, 16), we
strategy for improving glycemic regulation in individuals with hypothesized that HIT might also increase mitochondrial ca-
T2D and suggest that CGM may be a sensitive technique to pacity in skeletal muscle of individuals with T2D. Low-volume
measure the effects of exercise on glucose control. HIT was a potent stimulus to increase mitochondrial capacity

J Appl Physiol • VOL 111 • DECEMBER 2011 • www.jap.org

Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (125.166.116.202) on November 2, 2018.
Copyright © 2011 American Physiological Society. All rights reserved.
 INTERVAL TRAINING IN TYPE 2 DIABETES 1559
in the current study, as evidenced by increased enzyme activity improves insulin action in young healthy males. BMC Endocr Disord 9: 3,
of CS as well as elevated protein content of several subunits 2009.
2. Boule NG, Haddad E, Kenny GP, Wells GA, Sigal RJ. Effects of
from complexes in the electron transport chain. exercise on glycemic control and body mass in type 2 diabetes mellitus: a
Another novel observation in the present study was that meta-analysis of controlled clinical trials. JAMA 286: 1218 –1227, 2001.
low-volume HIT increased the protein content of Mfn2. The 3. Bruce CR, Anderson MJ, Carey AL, Newman DG, Bonen A, Kriketos
primary role of Mfn2 is in mitochondrial fusion, although it AD, Cooney GJ, Hawley JA. Muscle oxidative capacity is a better
also appears to regulate the expression of electron transport predictor of insulin sensitivity than lipid status. J Clin Endocrinol Metab
88: 5444 –5451, 2003.
chain subunits and influence mitochondrial bioenergetic capac- 4. Cauza E, Hanusch-Enserer U, Strasser B, Kostner K, Dunky A,
ity (28). Our findings provide evidence that elevated Mfn2 may Haber P. Strength and endurance training lead to different post exercise
be involved in regulating the increase in mitochondrial capac- glucose profiles in diabetic participants using a continuous subcutaneous
ity following low-volume HIT. A role for Mfn2 in the patho- glucose monitoring system. Eur J Clin Invest 35: 745–751, 2005.
genesis of T2D is supported by studies reporting reduced Mfn2 5. Ceriello A. The possible role of postprandial hyperglycaemia in the
pathogenesis of diabetic complications. Diabetologia 46, Suppl 1: M9 –16,
expression in skeletal muscle of patients with T2D (11), 2003.
suggesting that alterations in mitochondrial fusion/fission may 6. Colberg SR, Sigal RJ, Fernhall B, Regensteiner JG, Blissmer BJ,
contribute to mitochondrial impairment. Whether a training- Rubin RR, Chasan-Taber L, Albright AL, Braun B. Exercise and type
induced increase in muscle Mfn2 is linked to improved meta- 2 diabetes: the American College of Sports Medicine and the American
bolic health is unknown. Further studies are needed to clarify Diabetes Association: joint position statement. Diabetes Care 33: e147–
167, 2010.
the role of mitochondrial dynamics and examine the effects of 7. Earnest CP. The role of exercise interval training in treating cardiovas-
exercise and other interventions on Mfn2 skeletal muscle. cular disease risk factors. Current Cardiovasc Risk Rep 3: 296 –301, 2009.
8. Gibala MJ, Little JP, van Essen M, Wilkin GP, Burgomaster KA,
Conclusions Safdar A, Raha S, Tarnopolsky MA. Short-term sprint interval versus
traditional endurance training: similar initial adaptations in human skeletal
Two weeks of low-volume HIT, involving only 30 min of muscle and exercise performance. J Physiol 575: 901–911, 2006.
vigorous exercise within a total time commitment of 75 min/ 9. Hawley JA, Lessard SJ. Mitochondrial function: use it or lose it.
wk, lowered 24-h average blood glucose concentration, re- Diabetologia 50: 699 –702, 2007.
duced postmeal blood glucose excursions, and increased mark- 10. Heled Y, Shapiro Y, Shani Y, Moran DS, Langzam L, Barash V,
Sampson SR, Meyerovitch J. Physical exercise enhances hepatic insulin
ers of skeletal muscle mitochondrial capacity in individuals signaling and inhibits phosphoenolpyruvate carboxykinase activity in
with T2D. The total weekly training time commitment in the diabetes-prone Psammomys obesus. Metabolism 53: 836 –841, 2004.
present study was 50% lower than recently revised guidelines 11. Hernandez-Alvarez MI, Thabit H, Burns N, Shah S, Brema I, Hatunic
that call for 150 min of moderate to vigorous exercise per M, Finucane F, Liesa M, Chiellini C, Naon D, Zorzano A, Nolan JJ.
week. While longer-term comparative studies are clearly war- Subjects with early-onset type 2 diabetes show defective activation of the
skeletal muscle PGC-1alpha/Mitofusin-2 regulatory pathway in response
ranted, our findings indicate that low-volume HIT may repre- to physical activity. Diabetes Care 33: 645–651, 2010.
sent a time-efficient exercise strategy for the treatment of T2D. 12. Holloszy JO. Skeletal muscle “mitochondrial deficiency” does not medi-
Future research is needed to examine the long-term influence ate insulin resistance. Am J Clin Nutr 89: 463S–466S, 2009.
of HIT and to comprehensively examine how this type of 13. Holloway GP, Bonen A, Spriet LL. Regulation of skeletal muscle
training compares to traditional therapeutic exercise strategies. mitochondrial fatty acid metabolism in lean and obese individuals. Am J
Clin Nutr 89: 455S–462S, 2009.
GRANTS 14. Hood MS, Little JP, Tarnopolsky MA, Myslik F, Gibala MJ. Low-
volume interval training improves muscle oxidative capacity in sedentary
The Canadian Diabetes Association and Natural Sciences and Engineering adults. Med Sci Sports Exerc 43: 1849 –1856, 2011.
Council (NSERC) of Canada provided the funding for this project (Principal 15. Klonoff DC. Continuous glucose monitoring: roadmap for 21st century
Investigator: MJG). J. P. Little was supported by an NSERC doctoral schol- diabetes therapy. Diabetes Care 28: 1231–1239, 2005.
arship (PGS-D) and J. B. Gillen held a NSERC masters scholarship (PGS-M). 16. Little JP, Safdar A, Wilkin GP, Tarnopolsky MA, Gibala MJ. A
A. Safdar was supported by a Canadian Institutes of Health Research (CIHR) practical model of low-volume high-intensity interval training induces
doctoral scholarship and M. E. Percival held a CIHR undergraduate student- mitochondrial biogenesis in human skeletal muscle: potential mechanisms.
ship. We thank Medtronic of Canada for in-kind contribution of continuous J Physiol 588: 1011–1022, 2010.
glucose monitors and sensors and Dr. M. Riddell, Dr. R. Manders, and D. 17. Manders RJ, Van Dijk JW, van Loon LJ. Low-intensity exercise
Kelly for helpful advice and assistance with continuous glucose monitoring reduces the prevalence of hyperglycemia in type 2 diabetes. Med Sci
measurements. Sports Exerc 42: 219 –225, 2010.
18. Mootha VK, Lindgren CM, Eriksson KF, Subramanian A, Sihag S,
DISCLOSURES
Lehar J, Puigserver P, Carlsson E, Ridderstrale M, Laurila E, Houstis
No conflicts of interest, financial or otherwise, are declared by the authors. N, Daly MJ, Patterson N, Mesirov JP, Golub TR, Tamayo P, Spiegel-
man B, Lander ES, Hirschhorn JN, Altshuler D, Groop LC. PGC-
AUTHOR CONTRIBUTIONS 1alpha-responsive genes involved in oxidative phosphorylation are coor-
dinately downregulated in human diabetes. Nat Genet 34: 267–273, 2003.
Author contributions: J.P.L., J.B.G., M.A.T., Z.P., M.E.J., and M.J.G.
19. Praet SF, Manders RJ, Lieverse AG, Kuipers H, Stehouwer CD,
conception and design of research; J.P.L., J.B.G., M.P., and A.S. performed
Keizer HA, van Loon LJ. Influence of acute exercise on hyperglycemia
experiments; J.P.L., J.B.G., M.P., A.S., M.E.J., and M.J.G. analyzed data;
in insulin-treated type 2 diabetes. Med Sci Sports Exerc 38: 2037–2044,
J.P.L., J.B.G., A.S., M.A.T., Z.P., M.E.J., and M.J.G. interpreted results of
experiments; J.P.L. and J.B.G. prepared figures; J.P.L. drafted manuscript; 2006.
J.P.L., J.B.G., M.P., A.S., M.A.T., Z.P., M.E.J., and M.J.G. edited and revised 20. Ren JM, Semenkovich CF, Gulve EA, Gao J, Holloszy JO. Exercise
manuscript; J.P.L., J.B.G., M.P., A.S., M.A.T., Z.P., M.E.J., and M.J.G. induces rapid increases in GLUT4 expression, glucose transport capacity,
approved final version of manuscript. and insulin-stimulated glycogen storage in muscle. J Biol Chem 269:
14396 –14401, 1994.
REFERENCES 21. Richards JC, Johnson TK, Kuzma JN, Lonac MC, Schweder MM,
Voyles WF, Bell C. Short-term sprint interval training increases insulin
1. Babraj JA, Vollaard NB, Keast C, Guppy FM, Cottrell G, Timmons sensitivity in healthy adults but does not affect the thermogenic response
JA. Extremely short duration high intensity interval training substantially to beta-adrenergic stimulation. J Physiol 588: 2961–2972, 2010.

J Appl Physiol • VOL 111 • DECEMBER 2011 • www.jap.org

Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (125.166.116.202) on November 2, 2018.
Copyright © 2011 American Physiological Society. All rights reserved.
1560 INTERVAL TRAINING IN TYPE 2 DIABETES

22. Ritov VB, Menshikova EV, Azuma K, Wood RJ, Toledo FG, 25. Snowling NJ, Hopkins WG. Effects of different modes of exercise
Goodpaster BH, Ruderman NB, Kelley DE. Deficiency of electron training on glucose control and risk factors for complications in type 2
transport chain in human skeletal muscle mitochondria in type 2 diabetic patients: a meta-analysis. Diabetes Care 29: 2518 –2527, 2006.
diabetes mellitus and obesity. Am J Physiol Endocrinol Metab 298: 26. Trost SG, Owen N, Bauman AE, Sallis JF, Brown W. Correlates of
E49 –E58, 2010. adults’ participation in physical activity: review and update. Med Sci
23. Schrauwen-Hinderling VB, Kooi ME, Hesselink MK, Jeneson JA, Sports Exerc 34: 1996 –2001, 2002.
Backes WH, van Echteld CJ, van Engelshoven JM, Mensink M, 27. Woerle HJ, Neumann C, Zschau S, Tenner S, Irsigler A, Schirra J,
Schrauwen P. Impaired in vivo mitochondrial function but similar Gerich JE, Goke B. Impact of fasting and postprandial glycemia on
intramyocellular lipid content in patients with type 2 diabetes mellitus overall glycemic control in type 2 diabetes. Importance of postprandial
and BMI-matched control subjects. Diabetologia 50: 113–120, 2007. glycemia to achieve target HbA1c levels. Diabetes Res Clin Pract 77:
24. Sheetz MJ, King GL. Molecular understanding of hyperglycemia’s 280 –285, 2007.
adverse effects for diabetic complications. JAMA 288: 2579 –2588, 28. Zorzano A. Regulation of mitofusin-2 expression in skeletal muscle. Appl
2002. Physiol Nutr Metab 34: 433–439, 2009.

J Appl Physiol • VOL 111 • DECEMBER 2011 • www.jap.org

Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (125.166.116.202) on November 2, 2018.
Copyright © 2011 American Physiological Society. All rights reserved.