Epidemiological studies suggest that physical activity and exercise have preventive effects against colon cancer (11,15,16,29,41), but the underlying biological mechanisms of these effects are not fully known. Moderate-intensity exercise (MIE) training was demonstrated to reduce the number of chemically induced aberrant crypt foci (ACF) (2,13). ACF are defined as lesions composed of enlarged crypts that are slightly elevated above the surrounding mucosa and more densely stained with methylene blue compared to normal crypts (4). ACF are considered putative preneoplastic colon lesions that may be early indicators of colon carcinogenesis (4,9,20). The proposed multiple steps of colon carcinogenesis may start when ACF appears in the colon (39), and the results of the above-cited animal studies (2,13) thus suggest that exercise training may have a preventive effect on the early phase of colon cancer development.
Epidemiological research has suggested that not only MIE but also higher-intensity exercise may reduce the number of colon adenomatous polyps (16), which comprise the next step of colon cancer development after ACF (9). We are interested in the effects of high-intensity exercise training on the number of chemically induced ACF, and in the present investigation, we found that high-intensity intermittent exercise (HIIE) training reduced the number of rat colon ACF induced by 1,2-dimethylhydrazine (DMH).
Because high-intensity exercise may reduce immunological functions (24) and increase the production of free radicals (38), it may be speculated that in terms of the prevention of any type of cancer, high-intensity exercise training may have no effects or adverse effects. For our evaluation of the results obtained in the present study, we investigated mechanisms that may explain the effectiveness of high-intensity exercise training on ACF reduction, with special reference to the protein known as “secreted protein acidic and rich in cysteine” (SPARC), which is secreted from skeletal muscle and upregulated by regular exercise and downregulated by aging (1). This myokine is a matricellular glycoprotein involved in the development, remodeling, and tissue repair by modulating cell–cell and cell–matrix interactions (5,6) as well as other functions, such as antitumorigenesis (32).
In terms of colon cancer prevention, SPARC was reported to be increased by acute MIE and after training using the exercise, and to inhibit the initiation of colon tumorigenesis via physical exercise by increasing the apoptosis of ACF in rat colon (1). We hypothesized that, like other proteins that are increased by exercise and exercise training, the expression of SPARC may also be regulated by exercise intensity signals of 5′ AMP-activated protein kinase (AMPK) (27), probably through its transcriptional stimulation effects on peroxisome proliferator-activated receptor γ coactivator-1α (PGC1α). If our hypothesis is right, the higher the exercise intensity, the more SPARC there would be, and consequently the less ACF would be observed.
There are a great number and variety of high-intensity exercise programs, and thus the choice of a specific exercise program with which the possible effects of high-intensity exercise on colon ACF development can be elucidated is not straightforward. In the present investigation, we used a typical HIIE protocol that has been popular as “Tabata protocol” and enjoyed by many people and for which an animal model has been fully developed (34,36,37,42). Because this HIIE protocol was shown to elevate the expressions of PGC1α (34), glucose transporter 4 (GLUT4) (37) and mitochondrial oxidative proteins in rat skeletal muscle (36,42), we used the protocol based on our hypothesis that with this exercise protocol, colon cancer preventive effects could be expected simultaneously with preventive effects against noncommunicable diseases (e.g., diabetes, obesity, and cardiovascular diseases).
MATERIALS AND METHODS
The present investigation consisted of five animal experiments and two human experiments.
The protocols for the animal experiments were approved by the ethics committee of Ritsumeikan University. The animals were housed in rooms lighted from 8:00 AM to 8:00 PM and were maintained on an ad libitum diet of standard chow and water. The room temperature was maintained at 20°C to 25°C.
Rat high-intensity and low-intensity swimming training experiment
For both the high-intensity intermittent swimming training (HIIST) and low-intensity swimming training (LIST), 4-wk-old Fischer 344 male rats were purchased from CLEA (Kyoto, Japan). After 1 wk of acclimatization to the housing environment (thus, at 5 wk of age), the rats were randomly assigned to either the swimming training group (n = 8) or the control group (n = 8) for each training study. Before the training, all rats were given a subcutaneous injection of DMH, a chemical that is carcinogenic in the colon, at 20 mg·kg−1 bodyweight 1× per week for 2 wk. The DMH was dissolved in 0.1 mM ethylenediaminetetraacetic acid (pH, 6.5) immediately before the administration (13).
One week after the last injection of DMH (when the rats were 7 wk old), the swimming training (i.e., HIIST) was started. Before the swimming exercise experiment, all rats were acclimated to swimming exercise for 10 min·d−1 for 2 d as described by Ren et al. (28). In the HIIST, the rat performed twelve 20-s swimming bouts while bearing weight equivalent to 16% of its body weight. A 10-s pause was provided between the exercise bouts (33). Each rat performed the swimming exercise alone in a barrel filled to a depth of 50 cm. This high-intensity exercise was shown to increase the blood lactate concentration in rats to approximately 11 to 12 mM (37).
The LIST consisted of a 2-h swimming session with no extra weight load. Eight rats swam simultaneously in a barrel filled to a depth of 50 cm and with an average surface area of 190 cm2 per rat. The intensity of this training was considered to be low because as reported elsewhere (35), rats demonstrated their ability to continue swimming in this manner for >6 h without exhaustion.
The LIST and HIIST groups underwent their training exercise for five consecutive days per week for 4 wk. For both levels of training, the water temperature was maintained at 35°C during the swimming exercise.
One day after the last day of exercise, the rats were anesthetized with an intraperitoneal injection of pentobarbital sodium (50 mg·kg−1 body weight). Then, the rats' colons were dissected and gently flushed with 10% neutralized formalin to remove residual bowel contents, cut open longitudinally, fixed flat between filter papers, and submerged in 10% neutralized formalin overnight at 4°C.
The fixed colons were stained with 0.2% methylene blue, as described by Bird (4). The number of ACF and the total number of aberrant crypts (ACs) comprising ACF were counted for each colon. ACF were identified as lesions composed of enlarged crypts with an increased pericryptal area, a slightly elevated appearance above the surrounding mucosa with an oval or slitlike orifice, and a higher 0.2% methylene blue staining intensity compared to normal crypts (4).
Rat HIIST and acute high-intensity intermittent swimming experiment
Four-week-old Wistar rats (60–65 g, n = 32) purchased from SLC Japan were randomly assigned to a training group (n = 16) and a nontraining control group (n = 16). Training using the HIIS described above was executed 1 time per day for 5 d. Two days after the last training day, eight rats of the training group executed the HIIS (trained HIIS, n = 8). Thirty minutes after HIIS, the rats were sacrificed, and muscle and blood were sampled. Samples from the other eight rats of the training group that did not undergo the HIIS (trained rest, n = 8) were obtained at approximately at the same time as when the trained HIIS group rats were sacrificed. At the same time of the same day as when the trained HIIS and trained rest rats were sacrificed, control group rats were sacrificed either 30 min after HIIS (control HIIS, n = 8) or at rest without HIIS (control rest, n = 8).
Rat in vivo AMPK stimulation experiment
5-aminoimidazole-4-carboxamide ribonucleoside (AICAR: an AMPK activator) (1.0 mg·g−1 body weight, n = 16) (40) or saline vehicle (n = 16) was intraperitoneally injected to 4-wk-old Wistar rats (60–65 g) purchased from SLC Japan. Six hours (n = 8) and 12 h (n = 8) after the AICAR injection, the rats were anesthetized by an intraperitoneal injection of pentobarbital sodium, and the epitrochlearis muscles were disected quickly, rinsed in ice-cold saline, weighed, and frozen in liquid nitrogen until analyzed. Blood was collected via the abdominal aorta and centrifuged (1500g, 15 min, 4°C), and then serum samples were stored at −80°C until use. Samples of the saline injected rats were obtained at the same time as when the AICAR injected rats were sacrificed 6 h (n = 8) and 12 h (n = 8) after the injection.
Rat in vitro experiment
Six 4- to 5-wk-old sedentary Wistar rats were used for this experiment. After sacrifice, two epitrochlearis muscles were dissected out from each rat. The muscles were incubated for 6 h in glass vials in a shaking incubator maintained at 35°C, in either the presence (n = 6) or absence (n = 6) of 0.5 mM AICAR. The suitability of the epitrochlearis muscle for long-term in vitro incubation experiments was demonstrated (33). The glass vials, containing 3 mL of the medium described below, were gassed continuously with 95% O2–5% CO2 throughout the incubation. The medium was replaced with fresh medium every 3 h. After incubation, the muscles were washed in 3 mL of phosphate-buffered saline (PBS) for 10 min, blotted, clamp frozen, and stored at −80°C until their further use.
The culture medium consisted of MEMα (12000-022; Life Technologies, Carlsbad, CA), 10% fetal bovine serum (Life Technologies), 50 U·mL−1 purified human insulin (Eli Lilly, Indianapolis, IN), 100 U·mL−1 penicillin (Life Technologies), and 100 g·mL−1 streptomycin (Life Technologies) (33).
The protocols for the two human experiments, including acute high-intensity intermittent exercise (HIIE) and high-intensity intermittent exercise training (HIIT) experiment, were approved by the ethics committee of Ritsumeikan University, and the experiments were conducted in accord with the Declaration of Helsinki. All subjects were given an oral and written briefing of the study, and each of the subjects provided written informed consent to participate.
Acute human HIIE experiment
Eight young men (age, 22 ± 1 yr) volunteered to participate in the study after they provided written informed consent. Each subject passed a comprehensive physical examination. The subjects’ weight and V˙O2max measured for bicycle ergometer exercise were 67.8 ± 18.3 kg and 49.4 ± 8.0 mL·kg−1·min−1, respectively.
All bicycle experiments and pretests were conducted on a mechanically braked cycle ergometer (Monark, Stockholm, Sweden) at 90 rpm (23,30). For each subject, oxygen uptake was measured during the last 2 min of six to nine different 10-min bouts of bicycling at constant power. After a linear relationship between the intensity of bicycling and the steady-state oxygen uptake was determined in this pretests, the subject's oxygen uptake was measured during the last two or three 30-s intervals during several bouts of supramaximal-intensity bicycle exercise that lasted for 2 to 4 min. The highest V˙O2max observed was taken as the subject's V˙O2max for each exercise (22).The criterion for exhaustion was that the subject was unable to maintain the pedaling frequency at or above 85 rpm near the end of the bout. These pretests were carried out on three to five separate days for each exercise.
After an overnight fast, the subjects arrived at the laboratory at approximately 9:00 AM. At 9:30 AM, the subjects started 10 min of warm-up bicycling exercise at 50% V˙O2max. Ten minutes after the conclusion of the warm-up exercise, the subjects conducted two different exercises. The order of the two exercises was randomized. At least 1 wk was allowed between the two exercises. For the MIE, the subjects exercised for 30 min on a bicycle ergometer at 70% V˙O2max. The exhaustive HIIE consisted of six to seven sets of high-intensity exercise with a 10-s rest between the exercise bouts. The exercise intensity for HIIE was 170% of the subject's V˙O2max.
Blood was sampled from each subject's antecubital vein immediately before the warm-up exercise and at 30, 60, 90, 120, and 180 min after the completion of the two different exercises. Blood from the subject's fingertip was obtained immediately after and at 1, 3, 6, and 9 min after the completion of the HIIE. The highest value of the lactate concentration was used as the subject's peak lactate concentration.
Human HIIT experiment
The subjects of this experiment were 11 sedentary or moderately active healthy men (age, 23 ± 1 yr; height, 173.7 ± 7.2 cm). All subjects were instructed to not alter their regular physical activity levels or dietary habits for the duration of the investigation.
All of the subjects underwent the 6-wk HIIT program. Each subject's V˙O2max and body weight were measured before and after the 6-wk HIIT program. For our examination of the skeletal muscle transcriptome, vastus lateralis (VL) muscle biopsies (described in the next section) were collected before and after the 6-wk HIIT program. All exercises, including the HIIT, were conducted on a mechanically braked cycle ergometer (828E; Monark, Stockholm, Sweden) at 90 rpm. Exercise tests before and after the HIIT were carried out on two to three separate days.
All subjects completed 6 wk of HIIT (100% compliance), including 24 training sessions (4 d·wk−1). Each session consisted of six to seven sets of 20-s exercise on a bicycle ergometer at an intensity of 170% of V˙O2max with a 10-s rest between each bout (30). The subjects were encouraged to complete six to seven sets of this exercise until exhaustion. The exercise was terminated when the pedaling frequency dropped below 85 rpm. When a subject was able to complete 8 or more sets of the high intensity bicycling during the training program, the exercise intensity was increased by 11 W.
VL muscle biopsy
Muscle biopsies of the VL were performed before and 48 to 72 h after the completion of the 6-wk HIIT program. The subjects arrived at the laboratory in the morning after an overnight fast. After resting for 15 min in a supine position, local anesthesia (xylocaine) was administered to the subject's skin and fascia. A small incision was made in the skin and muscle fascia, and then a muscle biopsy was taken from the VL muscle with the use of a disposable core biopsy instrument (MONOPTY Max-Core, 14G × 100 mm; Bard Biopsy Systems, Tempe, AZ). The muscle sample was quickly rinsed with ice-cold saline, freed from any visible nonmuscle material, and frozen immediately in liquid nitrogen. The samples were stored at −80°C until use.
Immunoblot analysis for phosphorylated-AMPK, total-AMPK, PGC-1α and SPARC proteins
Each muscle specimen was homogenized with 20 mM Tris–HCl (pH 7.8), 300 mM NaCl, 2 mM ethylenediaminetetraacetic acid, 2 mM dithiothreitol, 2% Nonidet P-40, 0.2% sodium lauryl sulfate, 0.2% sodium deoxycholate, 0.5 mM phenylmethylsulfonyl fluoride, 60 μg·g−1 aprotinin, and 1 μg·mL−1 leupeptin. The homogenate was rotated gently for 30 min at 4°C and then centrifuged at 12,000g for 15 min at 4°C. The protein concentration of the resulting supernatant was determined by a Bradford protein assay (25).
Samples (30 μg protein) were denatured at 96°C for 7 min in Laemmli buffer. A Western blot analysis was performed to detect the phosphorylation of AMPK (Threonine 172), and the expressions of the PGC-1α and SPARC protein. Briefly, each sample was separated on a 10% sodium lauryl sulfate-polyacrylamide gel and transferred to a polyvinylidene difluoride (Millipore, Billerica, MA) membrane. The membrane was treated with blocking buffer, that is, 5% skim milk in PBS with 0.1% Tween 20 (PBS-T) for 24 h at 4°C. The membrane was probed with phosphorylated-AMPK, total-AMPK (Cell Signaling, Beverly, MA), anti-PGC-1α (Merck Millipore, Darmstadt, Germany) and anti-SPARC antibody (Abcam, Cambridge, UK); all antibodies were diluted 1:1000 in blocking buffer.
The membrane was then washed three times with PBS-T and incubated for 1 h at room temperature with a horseradish peroxidase–conjugated secondary antibody, anti-rabbit immunoglobulin, diluted 1:3000 in blocking buffer (Cell Signaling, Beverly, MA). The membrane was then washed three times with TBS-T. Finally, phosphorylated-AMPK, total-AMPK, anti-PGC-1α, and SPARC proteins were detected using an enhanced chemiluminescence system (ECL Plus; GE Healthcare Biosciences) and visualized using an Image Quant LAS 4000 system (GE Healthcare Biosciences). The value of AMPK phosphorylation (arbitrary unit [AU]) was normalized by that of total AMPK expression.
The human serum levels of SPARC (SPARC Quantikine ELISA Kit, R&D Systems, Minneapolis, MN) were determined by a sandwich enzyme immunoassay. All techniques and materials used in the analysis were in accord with the manufacturer's protocol. The levels of SPARC in rat serum were determined using a sandwich enzyme immunoassay kit (Aviscera Bioscience, Santa Clara, CA). Immobilized polyclonal antibodies were raised against SPARC, whereas the secondary horseradish-peroxidase-coupled antibodies were monoclonal. Optical density at 450 nm was quantified using a microplate reader (xMark microplate spectrophotometer; Bio-Rad Laboratories, Hercules, CA). All samples were assayed in duplicate.
RNA extraction and real-time polymerase chain reaction for rat in vitro experiment
The epitrochlearis muscles were homogenized in TRIzol (Invitrogen, Carlsbad, CA) and total RNA was precipitated using chloroform and isopropanol. The DNase-treated total RNA (1 μg) was reverse-transcribed into cDNA by using random primer with a High Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Foster City, CA). Real-time RCR was performed using the FAST 7500 real-time polymerase chain reaction (PCR) system (Applied Biosystems). Gene expression was determined using commercially available Taqman primers and probes for SPARC and β-actin and Taqman Universal PCR Master Mix (Applied Biosystems). Gene expression levels were determined using the ΔΔCT method using β-actin as the housekeeping gene.
Quantitative PCR for human experiment
Total RNA was isolated from muscle biopsy samples using the miRNeasy® Mini Kit (Qiagen, Hilden, Germany), and a quality check of the total RNA was conducted using an Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA). Total RNA was reverse-transcribed using an Omniscript RT kit (Qiagen) in a 20-μL reaction mixture with oligo-d(T) 12 to 18 primer (ThermoFisher Scientific, San Jose, CA). The reaction was performed at 37°C for 120 min and at 94°C for 5 min.
The mRNA expression level of SPARC (Assay ID: Hs00234160_m1) was analyzed by a TaqMan® gene expression assay and the StepOneTM Real-Time PCR System (ThermoFisher Scientific). Beta-2 microglobulin (B2M) was used as an internal expression control (Assay ID: Hs00187842-m1). A PCR was performed in a 12.5-μL reaction mixture containing TaqMan Universal Master Mix II with UNG (20×), TaqMan Gene Expression Assay mix (20×), and cDNA (containing 10 ng of total RNA). All reactions were performed in duplicate, using the following profile: 1 cycle at 50°C for 2 min and at 95°C for 10 min, and 40 cycles at 95°C for 15 s, and at 60°C for 1 min. Standard curves were generated from five concentrations of total RNA (3.125, 6.25, 12.5, 25, and 50 ng) in duplicate for each experimental gene. The quantitative value of SPARC mRNA expression was normalized by that of B2M mRNA expression.
For the quantification of oxygen uptake during human exercise, the fractions of oxygen and carbon dioxide in the expired air were measured by a mass spectrometer (Arco 2000; Arcosystems, Chiba, Japan). The gas volume was measured by a gasometer (Shinagawa Seisakusho, Tokyo, Japan).
Values are expressed as means ± SD. All analyses were performed, using Sigma Plot 12 software (Systat, San Jose, CA). For the rat HIIST and LIST experiments, the differences between the control and training groups were compared using the nonpaired Student t test. A two-way analysis of variance (ANOVA) was performed for the rat HIIST and acute HIIS experiment, and rat in vivo AMPK stimulation experiment. For the rat in vitro experiment, a one-way ANOVA was performed. For the acute human HIIE experiments, the differences in serum SPARC levels between groups and times of blood sampling were analyzed by two-way ANOVA. For the human HIIT experiment, we used paired Student t tests to compare each parameter's before- and after-training values. Differences were considered significant when P < 0.05.
The rat HIIST experiment
The HIIST resulted in significantly lower body weight (control, 264 ± 15 g; training, 247 ± 111 g) (P < 0.05). The numbers of ACF were significantly lower in the HIIST group compared with the control group (P < 0.05) (Fig. 1). The number of total ACs was also significantly less in the training group (81 ± 47, n = 8) than in the control group (279 ± 134, n = 8) (P < 0.05).
The rat LIST experiment
The LIST resulted in significantly lower body weight (control, 214 ± 16 g; training, 174 ± 15 g) (P < 0.001). The numbers of ACF observed in the LIS group (14 ± 10) were significantly low compared with the control group (23 ± 14) (P < 0.05). The number of total ACs was also significantly less in the training group than in control group (control, 66 ± 44; training, 39 ± 28) (P < 0.05).
The rat HIIST and acute HIIS experiments
As shown in Figure 2A, the serum SPARC concentrations of the HIIS rats among the control rats (control HIIS) were not significantly different from those of the resting-control rats (control rest). The serum SPARC concentrations of the resting trained rats (trained rest) did not differ significantly from those of the resting control rats (control rest). The serum SPARC concentration observed after the HIIS of the 5-d trained rats (trained HIIS) was significantly higher than that observed in the other three groups of rats (P < 0.05) (Fig. 2A).
The SPARC protein levels measured after the HIIS in the epitrochlearis muscle of the control rats (control HIIS) were not significantly different from those of the resting controls (control rest) (Fig. 2B), whereas the levels in the muscle of the trained rats after HIIS (trained HIIS) are significantly higher than those of the nonexercised trained rats (trained rest). The SPARC protein levels of the trained rats after HIIS (trained HIIIS) were significantly higher than those observed in two control groups (Fig. 2B).
The PGC1α content in the epitrochlearis muscle after the HIIS of the nontrained resting rats (control HIIS) was not significantly different from that of the nontrained resting rats (control rest) (Fig. 3A). The PGC1α content in the epitrochlearis muscle after the HIIS of the trained rats (trained HIIS) was not significantly different from that of the resting trained rats (trained rest). The PGC1α content in the muscle measured after the HIIS of the trained rats (trained HIIS) was significantly higher than that observed in the nontrained resting rats (control rest) (P < 0.05).
The AMPK phosphorylation in the epitrochlearis muscle was increased by the HIIS in both the control and trained rats (P < 0.05) (Fig. 3B). In addition, the AMPK phosphorylation measured after the HIIS of the trained rats (trained HIIS) was significantly higher than that of the resting trained rats (trained rest) (P < 0.05).
Rat in vivo AMPK stimulation experiment
Six hours after the AICAR injection, the phosphorylation of AMPK in the rat epitrochlearis muscle was significantly higher than that in the non–AICAR-treated muscles, whereas no difference was observed between the control and AICAR-treated muscles at 12 h after the AICAR injection (Fig. 4A).
The expression of PGC-1α protein was elevated significantly at both 6 and 12 h after the AICAR injection compared with the noninjected control muscles at 6 and 12 h (P < 0.05), respectively (Fig. 4B).
As shown in Figure 5A, the SPARC content observed in the muscle at 6 h after the AICAR injection was significantly higher compared to the control rats’ muscle (P < 0.05). However, no significant difference in SPARC content was observed between the AICAR-injected and noninjected muscles at 12 h after the injection.
The serum SPARC concentrations of the rats measured 12 h after the AICAR injection were significantly higher than both those of the noninjected rats at the same time point and the concentrations measured at 6 h postinjection (P < 0.05) (Fig. 5B).
Rat in vitro experiment
The mRNA of SPARC in the rat epitrochlearis muscle at 6 h after the AICAR incubation (1.61 ± 0.14 AU, n = 6) was significantly higher than that of the control muscles (1.08 ± 0.12 AU, n = 6) (P < 0.05).
Acute human HIIE experiment
Immediately after both the MIE and HIIE, the serum SPARC concentrations rose significantly from the preexercise values (P < 0.05) (Fig. 6). No difference in serum SPARC concentration was observed between MIE and HIIE at this time point. At 0.5 h after the completion of the MIE, the subjects’ serum SPARC concentrations returned to the preexercise value, whereas their serum SPARC concentration after completion of the HIIE remained significantly higher the preexercise values (P < 0.05). In addition, the serum SPARC concentrations after HIIE at this time point were significantly higher than those after the MIE (P < 0.05). Serum SPARC concentrations returned to the preexercise value 1 to 3 h after the HIIE. The serum SPARC concentrations after the HIIE were significantly higher than those observed after MIE at 1 and 3 h after the exercises (P < 0.05).
The peak blood lactate concentration after the HIIE was 17.2 ± 2.7 mM, whereas that at the end of the MIE was 6.8 ± 1.1 mM, which was significantly less than that observed after the HIIE.
Human HIIT experiment
After the 6-wk HIIT program, the V˙O2max of the subjects was significantly increased by 9.2% (pretraining, 48.2 ± 1.8 mL·kg−1·min−1; posttraining, 52.4 ± 1.4 mL·kg−1·min−1; P < 0.01), but their body weights did not change (pretraining, 67.1 ± 2.1 kg; posttraining, 67.7 ± 2.0 kg).
The quantitative value of SPARC mRNA expression normalized by that of B2M mRNA expression in the subjects’ VL muscle was significantly increased by approximately 40% after the HIIT program (pretraining, 1.37 ± 0.30 AU; posttraining, 1.90 ± 0.44 AU; P < 0.01).
The results of our study demonstrated that the HIIST reduced the number of DMH-induced ACF in the rat colon, suggesting that high-intensity exercise training may have a preventive effect on colon cancer. Our findings also suggest that SPARC, a myokine whose expression is regulated by intensity-related signals of AMPK, may intervene in the effect of high-intensity exercise training on colon cancer prevention.
The HIIT program used herein has been shown to improve both aerobic and anaerobic fitness in humans (30). Our modified version of the HIIT program for rats was demonstrated to induce health-related changes. For example, after this training, PGC1α, GLUT4, and mitochondrial oxidative proteins in the rat skeletal muscle were increased (34,36,37,42). Using a different type of HIIT, Gibala and McGee (14) studied the effect of high-intensity exercise training on the expression of health-related proteins in human skeletal muscle. Their results and our present data suggest that HIIT may have preventive effects on diabetes, obesity, and cardiovascular diseases.
Our earlier research showed that a moderate-intensity, prolonged running training protocol reduced the number of DMH-induced ACF in rat colon (13), and our present rat experiments also demonstrated that low-intensity prolonged swimming training reduced the number of ACF. This result is in accord with a study by Lunz et al. (19), who showed that long-term prolonged swimming reduced the number of DMH-induced ACFs. They reported that swimming training with a load of 2% body weight was the most effective compared with exercise training groups with 0% and 4% body weight loads, suggesting that in terms of the prevention of colon cancer, an appropriate exercise intensity exists.
Similar to other types of cancer, these previous findings that exercise can reduce the number of ACF supports the hypothesis that the preventive effects of low-intensity exercise to MIE against colon cancer can be attributed to elevated immunological function brought about by the exercise (24). Because high-intensity exercise may reduce immunological functions (24) and increase free radical production (38), one may speculate that, in terms of the prevention of any type of cancer, high-intensity exercise training may have no effects or adverse effects. However, we observed herein that the HIIT reduced the DMH-induced ACF number (Fig. 1) and the total ACs in rat colon. This result cannot be explained by previously suggested mechanisms.
A long-term intervention study showed that compared to women, men who tended to exercise more vigorously had a greater risk reduction of colon cancer after a 12-month exercise intervention (21), suggesting that higher-intensity exercise may have more preventive effects on colon cancer. Kono et al. (16) also reported that only hard-to-severe exercise has preventive effects on colon cancer. These studies suggest that relatively high-intensity exercise may have preventive effects against colon cancer.
In terms of exercise-related colon cancer reduction, Aoi et al. (1) proposed a new mechanism. They reported that in rats, SPARC secreted from skeletal muscle by exercise is delivered by the blood stream to the colon and induces the apoptosis of colon ACF, resulting in fewer ACF compared with controls. In the present study's human experiments, the serum SPARC concentration after the high-intensity exercise (HIIE) was as high as that observed after the moderate-intensity prolonged exercise (MIE) (Fig. 6). Moreover, differences in the subjects’ serum SPARC concentration between HIIE and MIE were noted until 3 h after the exercises. These results might indicate that HIIE results in the secretion of the same or a greater amount of SPARC into the blood. This level of plasma SPARC might also enable the same or a greater amount of SPARC to reach the colon, resulting in similar or greater effects of SPARC on ACF reduction as those obtained in our present rat training experiment. These consequences may explain our finding that the HIIST reduced the number of DMH-induced ACF to a level comparable to that after the MIE training.
Although low-intensity exercise to MIE/training has been recommended to general populations, our present findings may indicate that, provided that an individual's immune function is well maintained, high-intensity exercise/training should not necessarily be excluded as an exercise recommendation for healthy young- to middle-age people, at least in terms of colon cancer prevention.
Using a Western blot analysis, Aoi et al. (1) showed that 30 min of MIE at 70% V˙O2max elevated human subjects’ serum SPARC concentration by approximately 20%. In the present investigation using a sandwich enzyme assay for the SPARC analysis, the serum SPARC concentrations observed after 30 min of the same-intensity exercise are comparable, confirming the results reported by Aoi et al. (1). In addition, our present investigation is the first to demonstrate the effects of HIIE on the serum SPARC concentration of humans. We observed no significant difference in the serum SPARC concentrations observed immediately after the two exercise regimens with different intensities and duration.
Because the exercise time (<4 min) of the high-intensity intermittent bicycling exercise used in the present human experiments was too short to increase the expression of SPARC protein, we suspect that the secretion of SPARC is not related to the SPARC protein expression in the recruited muscles during the high-intensity intermittent swimming (HIIS) exercise by the rats. For example, the PGC1α content in the rat skeletal muscle recruited during the HIIS (4 min and 40 s) did not increase immediately after the exercise, but a significantly higher content of the transcription co-activator was observed 2 h after the exercise (34). Therefore, the subjects’ serum SPARC concentration immediately after the high-intensity intermittent bicycle exercise is not explained by the increased expression of SPARC during the short exercise.
Another possibility is that vesicles containing SPARC in skeletal muscle were released from skeletal muscle by exercise stimuli (18). Further, because SPARC is known to be expressed in other organs, SPARC might be released from other organs during exercise. As the changes in the serum concentration of SPARC reflect a balance of SPARC secretion from muscle and the degradation of SPARC, there is also a possibility that exercise reduces the degradation rate of SPARC and elevates the serum SPARC concentration. Further investigations are necessary to address this issue.
The SPARC concentrations observed 30 min after the high-intensity bicycle exercise by the human subjects were slightly but significantly higher than those observed after the moderate-intensity prolonged bicycle exercise at the same time point. This might indicate a slight advantage of the high-intensity intermittent bicycle exercise in terms of colon cancer prevention.
Aoi et al. (1) reported that after MIE (70% V˙O2max) training, the increase in the plasma concentration of SPARC after the exercise was higher than that observed after the same pretraining exercise, suggesting that more SPARC is secreted after exercise training. One possible reason for this is increased protein expression by exercise training. We used an animal model to investigate the mechanisms explaining exercise and exercise training effects on the SPARC expression and secretion from skeletal muscle, respectively. After the rats underwent 5 d of high-intensity swimming training, their serum SPARC concentrations tended to be higher, whereas at 30 min after the HIIT program, the serum SPARC concentrations were significantly higher than those of the resting nonexercised control rats. This result is in agreement with a similar human study (1) showing that not only exercise but also training has an impact on colon cancer prevention via an elevation of the serum SPARC concentration. We also observed changes in the SPARC content in the rat epitrochlearis muscle, which is recruited during high-intensity swimming exercise (34). The SPARC concentrations in the present study’s nontraining exercised rats tended to be higher than those of the nontraining resting rats, and those of the trained exercised rats tended to be higher than those of the trained resting rats and were significantly higher compared to the nontraining resting rats. The reason that the resting value of SPARC in the epitrochlearis muscles of the training rats tended to be higher but did not exceed the nontraining resting level might be that SPARC may be secreted before they rats started the exercise; that is, during the dark/nighttime period when rats are active. It is necessary to conduct further investigations into the mechanisms underlying the secretion of SPARC from the inside to the outside of exercised muscles.
We also examined the cellular mechanisms that may explain the increased expression of SPARC in activated (exercised) skeletal muscle. The AMPK is a potent exercise intensity-related signal, and the exercise intensity-dependent activity of this enzyme (27) enhanced the expression of a transcriptional coactivator of PGC-1α, which elevates the expression of many proteins by increasing the transcription of the proteins (8). Toward the elucidation of the mechanism explaining the training-induced increase in SPARC secretion into blood, we used the AMPK activator AICAR for in vitro and in vivo experiments. AMPK is known as a main exercise intensity-related signal in skeletal muscle (27).
The results of the in vitro experiment demonstrated that the prolonged incubation of rat epitrochlearis muscle with AICAR increased the mRNA of SPARC in the muscle, suggesting that the mRNA and possibly the protein expression of SPARC may be elevated by the exercise intensity-related signal of AMPK. Our in vivo experiment showed that the AICAR injection elevated the AMPK phosphorylation and increased the expressions of PGC1α and SPARC, suggesting that, like proteins including mitochondrial oxidative enzymes, PGC1α stimulated by AMPK might increase the expression of SPARC in skeletal muscle. Because the AMPK activity during high-intensity exercise is higher than that recorded during low-intensity exercise to MIE (12,27), it is reasonable to speculate that the higher the exercise intensity, the higher the AMPK activity and the more SPARC expression after exercise/training. A greater expression of SPARC may result in less ACF and subsequently in colon cancer prevention.
We also observed that the SPARC concentrations in the serum of the rats at 12 h after the AICAR injection were not higher than those of the control rats (Fig. 5). Because, for the values depicted as 12 h after the AICAR injection, blood was sampled at 8:00 AM from rats that are likely to have been active during the night (AICAR was injected at 8:00 PM, just before the light was turned off in the animal room), we speculate that the increased expression of SPARC, at least that expected at 2:00 AM (6 h after the injection), was secreted due to their activity during the period from 2:00 AM to 8:00 AM. Actually, the SPARC level in the rats’ epitrochlearis muscle was decreased at 12 h after the AICAR injection (8:00 AM).
After 5 d of HIIST, the serum SPARC levels after acute HIIS (Trained HIIS) were higher than those observed in the nontrained resting rats (control rest) (Fig. 2A), suggesting that HIIT (or high-intensity exercise) has not only an acute effect but also a training effect. This could be due to the increased content of SPARC in the trained muscle after the exercise (Fig. 2B). The changes in SPARC content in the muscle were similar to the response of PGC1α (Fig. 3A), suggesting that this line of response started from the PGC1α activated transcription of SPARC. Although HIIE increased the AMPK phosphorylation after the exercise for both the control and trained rats (Fig. 3B), the HIIS exercise did not increase the PGC1α content in epitrochlearis muscle both in the control muscle (control HIIS) and in the trained muscle (trained HIIS) (Fig. 3A). This result may be explained that 30 min is not long enough to induce significant increase in PGC1α protein expression after the HIIE (34). PGC1α content of the trained rats after HIIS (trained HIIS) is higher than that of the nonexercised control rat (control rest) (Fig. 3A). This result might be due to summation of nonsignificant effects of training (HIIST) and exercise (HIIS).
Our earlier study showed that the PGC1α content remained elevated until 18 h after HIIS, suggesting that elevated transcriptional activities by PGC1α are maintained for a long period (34). In the present investigation, the PGC1α content in rat epitrochlearis muscle measured 2 d after the last bout of HIIT (trained rest) was not significantly different from that of the nontrained control resting rats (control rest) (Fig. 3A). This result demonstrated that the elevation of PGC1α after HIIT did not persist for 2 d, and it may indicate that the effects of PGC1α on the transcription of SPARC and the majority of the subsequent SPARC content in the muscles wear off after 2 d.
Finally, the present investigation revealed that the mRNA content in the human muscle recruited by the HIIT program was elevated. The magnitude of the increase in mRNA (1.6-fold) is comparable to that reported in a study using 12-wk combined training consisting of moderate-intensity endurance and resistance exercise (7). Our findings suggests that the elevated SPARC mRNA in skeletal muscle after HIIT may increase the expression of SPARC in skeletal muscle and its secretion to blood, and that it may possibly decrease number of ACF, which is the first step of colon cancer development. Because the interval required for the evolution of colon cancer from ACF is known to be more than 20 yr (10), colon cancer is preventable if appropriate long-term primary care tools are available. Therefore, in addition to recommending low-intensity exercise to MIE for the prevention of noncommunicable diseases (26), such high-intensity exercises (including sports; e.g., recreational football) can be recommended as a primary prevention tool for colon cancer to young adults, young middle-age, and even older middle-age people, because such sports are enjoyable and may thus result in better adherence for a longer lifespan (3,17).
We also note that 12 h after the AICAR injection to rats, the AMPK phosphorylation returned to a level not higher than the control level, whereas the PGC1α content in the epitrochlearis muscle remained at the high level. Increased PGC1α persisted for 18 h after HIIE (34), suggesting that PGC1α that reaches a high level continues to be high for hours. The level of PGC1α that we observed at 12 h after the AICAR injection thus appears to reflect the amount of the protein expressed at 6 h after the injection.
Even the present investigation showed, in different experiments, that (1) HIIST reduced the number of DMH-induced ACF in the colon, and (2) the serum SPARC concentration of the trained HIIS rats was higher than that of the nontrained resting rats, suggesting that SPARC, which is known to induce the apoptosis of ACF (1), may have reduced the number of ACF of the DMH-injected rats. However, these results do not necessarily demonstrate that SPARC directly functioned to reduce the number of DMH-induced ACF in the colon of our study's HIIST rats. Further experiments are needed to test the direct relationship between the increased serum SPARC concentration induced by HIIE training and the number of ACF of individual DMS-injected rats.
Our findings demonstrated that HIIE training increases the serum concentration of SPARC, which was suggested to have a preventive effect against colon cancer by inducing the apoptosis of ACF, the appearance of which is the first step in colon cancer development (which has shown a “step by step” nature). It can thus be speculated that the reduced number of ACF after high-intensity exercise training may be linked to a reduced number of polyps in the future and finally to a reduced risk of colon cancer development, if the processes after the appearance of ACF are not different among individuals. However, our present findings did not prove a direct relationship between an increased blood SPARC concentration and future polyps or colon cancer development. Long-term animal experiments performed to determine the changes in not only the DMH-induced ACF number but also the appearance of polyps and the final outcome of colon cancer after high-intensity exercise training are necessary for the evaluation of preventive effects of high-intensity exercise training against colon cancer development. Moreover, in addition to cross-sectional human studies (31), longitudinal human studies that reveal the long-term effects of low-intensity physical activity and moderate- to high-intensity exercise training on the serum SPARC concentration, the number of ACF, and the development of polyps and colon cancer incidence should be conducted.
In summary, our present experiments demonstrated that high-intensity exercise training reduced the number of chemically induced ACF in rat colon, suggesting that this type of exercise can be recommended (or at least not excluded) for preventing colon cancer. Our findings contribute to the clarification of the mechanisms underlying the reduction in the number of ACF in rat colon by high-intensity exercise training, with special reference to the myokine SPARC, which was found to be expressed in proportion to the exercise intensity-related signal of AMPK.
The authors express their great appreciation to the study participants. This study was supported by a MEXT Kakenhi Grant-in-Aid for Scientific Research (B) 25282219.
The authors have no conflicts of interest to report. The results of the present investigation do not constitute endorsement by the ACSM. The results of the study are presented clearly, honestly, and without fabrication or inappropriate data manipulation.