Methods

Subjects and study design

This was a case-control study with pre- and post-TCC exercise testing of immune function parameters. Outcome measures were collected before and after the 12-week TCC exercise. Subjects studied were patients with diabetes and age-matched normal adults who lived in Kaohsiung County, Taiwan and responded to a recruitment notice posted in diabetic clinics and community culture centres. The recruitment poster announced the search for normal and diabetic volunteers to participate in TCC exercise in order to improve immunity. Subjects with autoimmune disorders treated with immunosuppressive drugs, malignancies treated with chemotherapy or other diseases treated with corticosteroids were excluded from the study because these diseases usually have altered immunity that might interfere with interpretation of the results.

Thirty-two pairs of diabetic patients and age-matched controls initially participated in the programme; 30 pairs of participants completed the 12-week TCC exercise programme and underwent blood collections before and after TCC. Demographic data including age, gender, educational level and body mass index (BMI) were also collected for analysis.

Training protocol

TCC exercise is referred to as “meditation through movement,” incorporating elements of balance, postural alignment and concentration.7 Subjects learned to perform 37 standardised movements (Cheng’s Tai Chi 37 forms)7 20 21 under the guidance of an expert TCC master with 31 years of experience who conducted all treatment sessions throughout the 12-week intervention period and gave the participants a take-home video tape for demonstration of the Tai Chi 37 forms. Each TCC session was 60 min which included a 10 min warm-up (stretching and balancing exercises), a 40 min practice and a 10 min cool-down. Sessions were given three times a week and were administered from 08.00 to 09.00 h at a community cultural centre.

Measurement of fasting blood glucose, haemoglobin A1c (HbA1c) and complete blood counts

Peripheral blood (3 ml) was collected into anticoagulant-containing conical tubes for analyses of fasting blood glucose, HbA1c levels and complete blood counts (CBC) before and after the exercise programme (3 days after the last exercise). The first 1.0 ml of blood was used for measurement of blood glucose by an enzymatic UV test for the quantitative determination of glucose (Olympus Diagnostica GmbH, Ireland).21 A senior technician recalibrated the assay monthly, based on the Olympus System Calibrator that was traceable to the National Institute of Standard and Technology (NIST) Standard Reference Material (SRM) 965. The second 1.0 ml of blood was used to measure HbA1c levels by the glycated protein analyser CLC385 (Primus Co, Kansas City, USA) before and 3 days after completion of the 12-week TCC exercise regime. The third 1.0 ml of blood was used to measure CBC by an automatic haematology analyser (Fluidautomatic Cell Count, Sysmex Co, Long Grove, Illinois, USA) that was calibrated every day to assure precision.

Measurement of plasma Th1/Th2/Treg (IL-12/IL-4/TGFβ) cytokine profiles

Plasma was collected from peripheral blood (2 ml) in anticoagulant-containing tubes by centrifugation at 1500 g for 15 min. Paired plasma samples were collected from 29 patients and age-matched controls for measurement of the cytokines IL-12, IL-4 and TGFβ, representing Th1, Th2 and Treg reaction as previously described.21 22 23 These cytokines were tested by an enzyme linked immunosorbent assay (ELISA) purchased from BioVendor Laboratory Inc (Modrice, Czech Republic). The detection limits of IL-12, IL-4 and TGFβ were 4.0 pg/ml, 0.22 pg/ml and 31.2 pg/ml, respectively, according to the manufacturer’s recommendation.

Measurement of Th1/Th2/Treg transcription factors (T-bet/GATA-3/Foxp3) expression of leucocytes

Total RNA of blood leucocytes was extracted by Tri-Zol solution (Invitrogen, California, USA) after plasma was harvested for measurement of cytokines. In brief, blood cell pellets were mixed with 0.5 ml Tri-Zol solution followed by phase separation by 0.5 ml chloroform (Merck KGaA, Darmstadt, Germany). The upper aqueous phase was transferred to a fresh DEPC-treated Eppendorf for RNA precipitation by isopropanol (Merck KGaA, Darmstadt, Germany) at −20°C for 1 h. The RNA was further precipitated by 70% ethanol (Merck KGaA), then subjected to real-time RT-PCR detection with hydrolysis probes (TaqMan) using the ABI PRISM 7700 instrument (Applied Biosystems, Foster City, California, USA) as previously described.24 25 TaqMan probes and primers for the quantitative detection of target mRNAs were designed using Primer Express computer software (Applied Biosystems Inc). For measurement of T-bet mRNA expression, the forward primer was 5′-AAC ACA GGA GCG CAC TGG AT-3′, the reverse primer was 5′-TCT GGC TCT CCG TCG TTC A-3′ and the probe sequence for detecting T-bet mRNA was 5′-TGA CCC AGA TGA TTG TGC TCC AGT CC-3′. For detecting GATA-3 mRNA, the forward primer was 5′-ACC GGC TTC GGA TGC AA-3′, the reverse primer was 5′-TGC TCT CCT GGC TGC AGA C-3′ and the probe sequence for GATA-3 was 5′-TCC AGC ACA GGC AGG GAG TGT GTG-3′. For detecting FoxP3 mRNA expression, the forward primer was 5′-TCA CCT ACG CCA CGG TCA-3′, the reverse primer was 5′-CAC AAA GCA CTT GTG CAG-3′ and the probe sequence was 5′-ATC CGC TGG GCC ATC CTG GAG GCT C-3′. Primer sequences for the internal control (β-actin) were 5′-GGC CAA CCG CGA GAA GAT-3′ for the forward primer, 5′-CGT CAC CGG AGT CCA TCA C-3′ for the reverse primer and 5′-TGC TAT CCC TGT ACG CCT CTG GCC-3′ for the probe sequence. These probes were purchased from the Assays-on-Demand gene expression of Applied Biosystems Inc. Each sample was analysed with duplicate experiments. The RT-PCR cycling parameters were set as follows: RT reaction at 50°C, 2 min; 60°C, 30 min; and 95°C, 5 min; followed by 40 cycles of PCR reactions at 94°C, 20 s and 60°C, 1 min. The results were detected in real-time and recorded on a plot showing fluorescence versus time. RT-PCR products were also visualised on ethidium bromide-stained 1.5% agarose gel (Pierce Co, Rockford, Illinois, USA) with a 100 bp ladder (Pharmacia Biotech, Piscataway, New Yersey, USA) as a reference. The increases in T-bet, GATA-3 and FoxP3 mRNA expression were therefore calculated assuming 100% efficient PCR where each C_t was normalised to β-actin mRNA expression as shown by the equation at 2^{{[Ct1(target) - Ct1(actin)] - [Ct2(target) - Ct2(actin)]}}. The C_t1 (target) and C_t2 (target) represent the C_t values for the T-bet, GATA-3 or FoxP3 mRNA expression in pre- and post-TCC samples, respectively. C_t1 (actin) and C_t2 (actin) represent the C_t values for the β-actin gene expression in pre- and post-TCC samples, respectively.

Data analysis and statistics

Data from this study were coded, double checked and analysed using descriptive analysis, χ² and Student t tests. Changes in fasting glucose and HbA1c levels as well as immune parameters before and after the TCC exercise were compared using paired t test. A p value of ⩽0.05 was considered statistically significant.

Results

Demographic data of patients with DM and age-matched controls

Thirty-two pairs of patients with DM and age-matched controls were initially recruited for the study and 30 pairs of participants completed it. There were no significant differences in demographic data between patients and controls (table 1). The male to female ratios in patients and controls were 16:14 and 10:20, respectively. The mean (SD) ages of the patients and controls were 56.7 (10.8) and 52.5 (11.9) years, respectively. No significant difference was seen in the number of subjects receiving post-high school education between patients with DM and controls. The BMI in patients with type 2 DM was higher than in normal controls before TCC exercise (24.28 (3.85) vs 22.10 (2.99) kg/m²), but the difference was not significant (p = 0.06). After TCC exercise the changes in BMI were not significantly different between patients with DM and controls (0.09 (0.72) vs −0.15 (0.58) kg/m², p = 0.16). The mean (SD) duration of the diabetes in the participating patients was 5.48 (5.91) years (table 1). The patients with DM committed themselves to take regular diabetic medication during the TCC exercise period. The mean (SD) number of diabetic medications was 1.8 (0.88) (range 0–4). Most of the patients (15/30, 50%) took metformin and glimepiride; four patients (13.3%) received single drug therapy (either metformin or glimepiride), eight (26.7%) took metformin plus glipizide, glibenclamide or pioglitazon, two received no medication and only one patient received four drugs.

Changes in blood glucose, HbA1c and CBC

All the participating patients with type 2 DM and age-matched controls had normal haemoglobin levels before the regular TCC exercise (table 2). However, patients with type 2 DM had higher blood glucose and HbA1c levels than normal controls. After the 12-week TCC exercise programme the HbA1c levels decreased significantly (p = 0.047) but the fasting glucose levels showed only an insignificant decrease (165.3 (53.2) vs 152.6 (48.8); p = 0.120, table 2). Patients with type 2 DM had significantly higher total white blood cells (WBC) than age-matched normal controls before the TCC exercise (6646 (1869) vs 5905 (1478); p = 0.01). Total WBC and differential counts did not change significantly before and after the TCC exercise programme. Platelets and red blood cells also showed no significant change after the TCC exercise (table 2).

Changes in Th1/Th2/Treg cytokines after TCC exercise

There were no significant differences in IL-12, IL-4 and TGFβ levels between patients with DM and age-matched controls before TCC exercise (fig 1). The mean IL-12 levels in patients with DM were not significantly lower than those in controls (5.96 (1.10) pg/ml vs 6.13 (2.01) pg/ml). IL-4 levels in patients with DM and controls were 0.84 (0.01) pg/ml and 0.85 (0.02) pg/ml, respectively, and the TGFβ levels in patients with DM and controls were 1125 (61) pg/ml and 1048 (45) pg/ml, respectively. After the 12-week TCC exercise the IL-12 levels in patients with DM increased significantly (p = 0.035; fig 1A) but in control subjects IL-12 levels did not increase significantly (fig 1A). In contrast, IL-4 levels decreased significantly in patients with DM (p = 0.003; fig 1B) but not in normal adult controls (fig 1B). The TGFβ levels in patients with DM and controls did not change significantly after the TCC exercise (fig 1C).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Changes in interleukin (IL)-12, IL-4 and transforming growth factor (TGF)β levels before and after TCC exercise between controls and patients with diabetes mellitus (DM). (A) Blood IL-12 levels increased significantly in patients with DM after exercise (p = 0.035). (B) Blood IL-4 levels decreased significantly in patients with DM after exercise (p = 0.003). (C) No significant changes were seen in TGFβ levels in patients with DM and controls after TCC exercise. Data presented are calculated from 29 paired samples.

Changes in Th1/Th2/Treg transcription factors

To study the molecular mechanism of Th1/Th2/Treg changes after TCC exercise, the expression of Th1/Th2/Treg specific transcription factors was measured by RT-PCR analysis of T-bet/GATA-3/FoxP3 mRNA expression. After normalisation to β-actin mRNA expression, T-bet mRNA expression was significantly lower in patients with DM than in controls (7.5 (1.5) vs 16.7 (4.7)-fold increase, p = 0.042; fig 2A). There were no significant differences in GATA-3 (fig 2B) and FoxP3 (fig 2C) mRNA levels between patients with DM and age-matched controls before TCC exercise. After the 12-week TCC exercise the T-bet mRNA levels in patients with DM increased significantly (7.5 (1.5) vs 19.1 (4.6), p = 0.009; fig 2A) and the T-bet mRNA levels in controls showed only an insignificant increase (p = 0.509, fig 2A). GATA-3 mRNA expression was not significantly different before and after TCC exercise in both controls and patients with DM (fig 2B). FoxP3 mRNA expression revealed a significant increase after TCC exercise in patients with DM but not in controls (p = 0.026, fig 2C).

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

Changes in T-bet, GATA-3 and FoxP3 mRNA expression after TCC exercise between controls and patients with diabetes mellitus (DM). (A) T-bet mRNA expression levels were significantly increased in patients with DM (p = 0.009) but not in controls (p = 0.109) after exercise. (B) GATA-3 expression showed an insignificant increase between controls and DM patients before and after exercise. (C) No significant change in FoxP3 expression was seen between controls and patients with DM before exercise but a significant increase in FoxP3 expression after exercise was found in patients with DM (p = 0.026).

Discussion

Most exercise immunology studies conducted with an exhaustive exercise design have concluded that immunosuppression associated with risk of upper respiratory tract infections occurs after an exhaustive exercise.1 2 Both innate and adaptive immune systems exhibit adverse changes after heavy exertion lasting longer than 90 min. These immune changes occur between 3 and 72 h, depending on the immune parameters measured.1 2 26 On the other hand, some clinical and experimental studies have shown certain beneficial effects of moderate exercise on immune functions. Shore et al27 reported that 6 months of supervised exercise training of previously sedentary elderly people resulted in nominal increases in some immune function measures. Sothern et al28 reported that a progressive moderate-to-intensive exercise programme did not adversely affect the immune system but benefited the IL-2/natural killer cell system. Moderate exercise in mice was also shown to enhance tumour immunity in the lungs.29 We have previously shown that regular TCC exercise for 12 weeks—which is classified as moderate exercise—increased the CD4/CD8 ratio and CD4CD25 regulatory T cells in one group of participants measured before and after TCC exercise.20 The present parallel case-control study has further shown that TCC exercise significantly increased Th1-specific T-bet mRNA expression together with an increase in blood IL-12 levels in patients with type 2 DM. To our knowledge, these are the first published results to demonstrate how moderate exercise such as TCC promotes a better Th1 response in patients with type 2 DM.

This study also found that levels of the Th2 mediator IL-4 decreased after TCC but levels of the Th2 transcription factor GATA-3 did not decrease, which suggests that the decrease in blood levels of IL-4 after TCC exercise may be caused by the augmented Th1 response which is known to suppress the Th2 reaction in a reciprocal way.22 23 24 25 26 It is not clear why enhanced expression of the Treg transcription factor FoxP3 (fig 2C) by TCC was not associated with an increase in TGFβ levels (fig 1C) in patients with type 2 DM. It may be because CD4CD25 regulatory T cells usually function through surface receptor contact regulation but not T cell regulatory mediator.30 It is also not completely understood why TCC exercise improves specific Th1 reaction and, to some extent, the Treg reaction in patients with type 2 DM but not in normal age-matched adults. We postulate that patients with type 2 DM have an altered balance in Th1/Th2/Treg reactions that could be significantly reversed by regular TCC exercise; in contrast, normal adults have a homeostatic balance in Th1/Th2/Treg reactions that is not significantly affected by regular TCC exercise. Recent studies have indicated that a perception of well being could enhance the Th1 response31 32 and that anxiety accelerates the Th2 response in patients with atopic dermatitis.33 Whether TCC exercise acts on the perception of well-being to enhance the Th1 response in patients with type 2 DM remains to be determined.

Many exercise programmes have tried to improve the metabolic dysfunction of patients with type 2 DM by regular physical exercise.34 35 36 Although these studies have been meta-analysed for their effects on physical fitness or HbA1c improvement,35 36 very few studies to date have investigated how regular exercise can improve both glucose metabolism and immune function. The present study has shown that a 12-week TCC exercise programme significantly decreased HbA1c levels and increased T-bet mRNA expression and blood IL-12 levels in patients with type 2 DM. It remains unclear whether the improvement in the Th1 reaction by TCC in patients with type 2 DM is directly related to the improvement in glucose metabolism reflected by the decrease in HbA1c or whether it is related to another mechanism independent of glucose metabolism. It is reasonable to postulate that TCC exercise may decrease blood glucose levels or improve glucose metabolism, resulting in a decrease in the inflammatory response elicited by hyperglycaemia. Hyperglycaemia has recently been shown to alter the inflammatory responses in subjects with metabolic syndrome, but these alterations are rapidly normalised in normoglycaemic subjects.37 38 The rationale for explaining how TCC exercise improves Th1 function independent of glucose metabolism includes (1) regular TCC exercise may counteract the proinflammatory reaction in diabetic patients who are usually obese and prone to proinflammatory response; (2) diabetic patients, who are prone to physical inactivity that can induce a proinflammatory reaction,17 may benefit from regular TCC exercise; and (3) TCC exercise can improve cardiopulmonary fitness and the perception of well-being39 which may result in a better Th1 reaction. Moreover, there are some other immune elements such as natural killer (NK) cell number, NK cell activity and its related cytokines IL-15 and IL-18 that could also affect Th1/Th2/Treg functions. Further studies are needed to investigate whether these immune elements can mediate TCC enhancement of Th1 function.

Conclusions

Patients with type 2 DM have higher HbA1C levels and lower Th1 transcription factor (T-bet) expression than age-matched normal controls. A 12-week course of TCC exercise significantly decreases HbA1C and increases T-bet expression, along with an increase in blood IL-12 concentration and a decrease in IL-4 concentration in patients with type 2 DM. A combination of TCC exercise with medication may result in an even better improvement in both metabolism and immunity of patients with type 2 DM.