Exercise-induced endothelial progenitor cell mobilization is attenuated in impaired glucose tolerance and type 2 diabetes

Circulating endothelial progenitor cells contribute to vascular homeostasis and angiogenesis through engraftment and paracrine mechanisms. This study is the first to identify an impairment in acute exercise-induced CD34+/VEGFR2+ endothelial progenitor cell mobilization in older adults with either impaired glucose tolerance or type 2 diabetes mellitus. This impairment could play an underlying role in the link between diabetes, capillary rarefaction, and other vascular complications.

Abstract

Circulating endothelial progenitor cells (EPCs) contribute to vascular homeostasis and are fewer in those with type 2 diabetes mellitus (T2DM) compared with normal glucose tolerance (NGT), suggesting a link between EPCs and T2DM-associated vasculopathies. The purpose of this study was to assess EPC number and mobilization by acute submaximal exercise in older adults with NGT, impaired glucose tolerance (IGT) or T2DM. We tested the hypothesis that EPC mobilization is lower in IGT compared with NGT and further reduced in older adults with T2DM. Forty-five older (50-75 yr of age) men and women with NGT (n = 18), IGT (n = 10), or T2DM (n = 17) were characterized and underwent submaximal aerobic exercise tests with blood sampling for enumeration of vascular endothelial growth factor receptor 2+ (VEGFR2+) cells, CD34+ hematopoetic progenitor cells, and CD34+/VEGFR2+ EPCs by flow cytometry before and after exercise. Basal EPC number was 65 and 61% lower in the IGT and T2DM groups, respectively, compared with the NGT group (P < 0.05). EPC number increased 23% after acute exercise in the NGT group (P < 0.01), but did not change in the IGT or T2DM groups. Before and after exercise, VEGFR2+ cell number was lower in a stepwise manner across the NGT, IGT, and T2DM groups (P < 0.05). Basal CD34+ cell number was lower in the IGT group compared with NGT (P < 0.05), but did not change after exercise in any group. These findings suggest a CD34+/VEGFR2+ EPC mobilization defect in IGT and T2DM that could play a role in the cardiovascular diseases and capillary rarefaction associated with insulin resistance.

NEW & NOTEWORTHY

Circulating endothelial progenitor cells contribute to vascular homeostasis and angiogenesis through engraftment and paracrine mechanisms. This study is the first to identify an impairment in acute exercise-induced CD34+/VEGFR2+ endothelial progenitor cell mobilization in older adults with either impaired glucose tolerance or type 2 diabetes mellitus. This impairment could play an underlying role in the link between diabetes, capillary rarefaction, and other vascular complications.

endothelial progenitor cells (EPCs) are circulating cells that contribute to vascular homeostasis in vivo through engraftment into the existing endothelium and paracrine mechanisms (5, 33). Low-circulating EPC number and reduced EPC function are associated with higher risk for cardiovascular disease, independent of traditional cardiovascular disease risk factors (12, 18, 36, 37). Circulating EPC number is lower than normal in people with type 2 diabetes mellitus (T2DM) (11, 13) and may contribute to fourfold greater risk of cardiovascular disease and other vascular complications (8, 17). These findings suggest a link between reduced EPC number and the vasculopathies associated with T2DM (4, 7).

In healthy individuals, circulating CD34+ hematopoetic progenitor cells, cells expressing the endothelial cell marker vascular endothelial growth factor receptor-2 (VEGFR2), and EPCs (often defined as cells expressing both CD34 and VEGFR2) are mobilized from the bone marrow milieu in response to pathological (31) and physiological stimuli, such as acute exercise (20, 22, 26, 28, 35). People with T2DM have an attenuated increase in CD34+/VEGFR2+ EPC number in response to an acute cardiac event (4), suggesting a defect in their ability to mobilize CD34+/VEGFR2+ EPCs in response to ischemia. To date, less is known about EPC mobilization in response to physiological stimuli, such as aerobic exercise in individuals with T2DM, but reduced mobilization of EPCs could contribute to the chronically low circulating number of these cells in T2DM.

While many of the vascular differences observed between individuals with T2DM and those with normal glucose tolerance (NGT) were previously attributed to chronically high plasma glucose levels, impaired glucose tolerance (IGT) represents a state of impaired glucose metabolism with more moderate elevations in plasma glucose. We previously identified low skeletal muscle capillarization in older adults with IGT compared with those with NGT (24), and the reduced capillarization is associated with lower clonogenic potential of circulating angiogenic cells, of which EPCs are one subset (25). Low EPC number could be another contributor to this capillary rarefaction in insulin-resistant individuals; however, there remains a paucity of data on EPC number and especially EPC mobilization in early stages of insulin resistance such as IGT. Thus the purpose of this study was to determine CD34+/VEGFR2+ EPC mobilization in response to acute submaximal exercise in older adults with T2DM, IGT, and NGT and also to assess the independent responses of the CD34+ and VEGFR2+ cell populations to acute exercise. We tested the hypothesis that EPC mobilization is lower in IGT compared with NGT, and further reduced in older adults with T2DM.

METHODS

Subjects and Screening

Subjects were sedentary (<20 min of aerobic exercise 2 times/wk), overweight-to-obese [body mass index (BMI) >25–35 kg/m²], middle-aged to older (50–75 yr) adults, recruited from the Baltimore, MD metro area and screened by physical examination, medical history, fasting blood chemistry, and graded maximal exercise test. Subjects all had normal graded treadmill exercise tests and no history of coronary artery disease, heart failure, peripheral arterial disease, stroke, liver, kidney disease, or lung disease. All subjects were current nonsmokers; those with a history of smoking had not smoked for >2 yr. The women in the study were postmenopausal for at least 1 yr. Subjects with a previous diagnosis of T2DM were included in the T2DM group; those without a known history of T2DM were given a 75-g oral glucose tolerance test and grouped according to American Diabetes Association criteria (3). The University of Maryland Baltimore Institutional Review Board approved all study procedures, and subjects provided written, informed consent.

Maximal Oxygen Consumption

Maximal oxygen consumption (V̇o_2max) was measured by indirect calorimetry (Quark, Cosmed USA, Chicago, IL) during a graded treadmill exercise test. Subjects walked at a constant velocity with the grade initially set to 0% and increasing every 2 min thereafter until volitional exhaustion. V̇o_2max was defined as the highest oxygen consumption (V̇o₂) value obtained for a full 30-s increment and was verified by standard physiological criteria (i.e., respiratory exchange ratio ≥ 1.10 or a plateau in V̇o₂, despite an increase in workload).

Acute Submaximal Exercise

After an overnight (∼12 h) fast, subjects completed a 30-min bout of treadmill exercise at 60% of their V̇o_2max, as measured by indirect calorimetry. A catheter was placed in an antecubital vein, and blood samples were drawn before aerobic exercise to determine basal EPC number and 30 min after completion of aerobic exercise to determine exercise-induced EPC mobilization, as this is the time point of maximum increase in CD34+/VEGFR2+ EPC number observed in healthy adults (20).

Circulating EPC, CD34+, and VEGFR2+ Cell Numbers

Peripheral blood mononuclear cells (PBMCs) were analyzed for the presence of surface antigens using flow cytometry. PBMCs were separated from whole blood by density gradient centrifugation using Lymphocyte Separation Medium (Mediatech, Corning Life Sciences, Manassas, VA). For each sample, 1 × 10⁶ cells were FcR blocked (Miltenyi Biotech, San Diego, CA) and immunostained with FITC-conjugated anti-human monoclonal CD34 (BD Biosciences, San Jose, CA) and phycoerythrin-conjugated anti-human monoclonal VEGFR2 (R&D Systems, Minneapolis, MN), then fixed in 2% paraformaldehyde. Samples were then processed in the University of Maryland Baltimore Flow Cytometry Core Facility with a MoFlo Legacy flow cytometer (Beckman Coulter, Brea, CA) by experienced operators blinded to subject clinical status and exercise time. Data were analyzed with Summit software (Beckman Coulter) using the appropriate isotype and single color controls (BD Biosciences). Cells were initially gated on the lymphocyte population based on forward and side scatter. The CD34/VEGFR2 analysis gate was established daily using isotype control samples and peripheral blood lymphocytes labeled with FITC-conjugated anti-human monoclonal CD3 and phycoerythrin-conjugated anti-human monoclonal CD3 antibodies (BD Biosciences). EPCs were defined as CD34+/VEGFR2+ cells. Individual enumeration of CD34+ circulating progenitor cells and VEGFR2+ cells was also completed. For EPCs, CD34+ cells were initially gated from total PBMCs and then examined for dual expression of VEGFR2+. Data are expressed as the number of positive cells per 10⁶ events. EPC mobilization was defined as the difference in circulating EPC number before (0 min) and 30 min after acute exercise (60 min).

Body Composition

Height and weight were measured to calculate BMI. Fat mass and fat-free mass were measured by dual-energy X-ray absorptiometry to determine percent body fat (Prodigy, LUNAR Radiation, Madison, WI).

Oral Glucose Tolerance Test

To determine glucose tolerance status, subjects reported to the laboratory between 7 and 9 AM after a 12-h overnight fast, and plasma glucose concentrations were measured (2300 STAT Plus, YSI, Yellow Springs, OH) before and 120 min after consumption of a beverage containing 75 g of glucose.

Plasma Lipoprotein-Lipid Profiles and Blood Pressure

Plasma triglyceride (TG) and cholesterol levels were analyzed using enzymatic methods (UniCel DxC880i; Beckman Coulter) and high-density lipoprotein (HDL) cholesterol (HDL-C) was measured in the supernatant after precipitation with dextran sulfate. Low-density lipoprotein (LDL) cholesterol (LDL-C) was calculated using the Friedewald equation: LDL-C = total cholesterol − (TG/5 + HDL-C) (16). Seated blood pressure was measured during three separate study visits, and the average was used in analyses.

Statistical Analyses

Statistical analyses were completed using IBM SPSS Statistics v22 (IBM, Armonk, NY). Analyses were completed using a two-factor (group × time) repeated-measures ANOVA with pairwise comparisons and linear trend analyses where appropriate. The χ² analysis was used to test for differences in medication use among groups. Pearson correlation analyses were used to assess relationships between subject characteristics (plasma lipids, blood pressure, body composition, V̇o_2max, or HbA1c) and circulating cell numbers. Statistical significance was accepted at P ≤ 0.05.

RESULTS

Subject Characteristics

Subject characteristics are reported in Table 1. Hemoglobin A1c and fasting plasma glucose concentrations were higher in the T2DM group compared with both the IGT and NGT groups (P < 0.01), but there were no significant differences in age, body weight, BMI, or percent body fat among groups. Blood pressure was similar among groups, with the exception of diastolic blood pressure, which was slightly higher in the T2DM group compared with the NGT group (P < 0.05). In the T2DM group, V̇o_2max (ml·kg⁻¹·min⁻¹) was 12 and 17% lower compared with the IGT and NGT groups, respectively (P < 0.05). Lipoprotein-lipid profiles were similar in the NGT and IGT groups; however, the T2DM group had lower total cholesterol and LDL-C compared with the NGT and IGT groups (P < 0.05), as well as lower HDL-C and higher TGs compared with the NGT group. Compared with the NGT group, a higher proportion of the participants in the IGT and T2DM groups were taking medications for dyslipidemia or hypertension (P < 0.05 for both). For participants with T2DM, the length of time since T2DM diagnosis was 6 ± 1 yr. There were no differences among the groups in the relative exercise intensity during the submaximal exercise bout (actual %V̇o₂ reserve).

Table 1.

Subject characteristics

	NGT	IGT	T2DM
n	18	10	17
Sex (female/male)	11/7	5/5	8/9
Race (black/white)	11/7	3/7	11/6
Age, yr	60 ± 2	64 ± 2	60 ± 2
Weight, kg	87 ± 4	88 ± 6	96 ± 4
BMI, kg/m2	29.9 ± 1.3	32.2 ± 1.8	32.8 ± 1.4
Body fat, %	36.6 ± 1.7	38.4 ± 2.0	39.1 ± 1.6
V̇o2max, ml·kg−1·min−1	26.6 ± 1.3	25.3 ± 1.8	22.1 ± 1.3‡
V̇o2max, l/min	2.23 ± 0.08	2.21 ± 0.11	2.08 ± 0.08
V̇o2 during acute exercise, %V̇o2 reserve	60 ± 1	60 ± 1	62 ± 1
Hemoglobin A1c, %	5.5 ± 0.17	5.6 ± 0.23	6.9 ± 0.17*
Fasting plasma glucose, mmol/l	5.07 ± 0.26	5.51 ± 0.35	7.22 ± 0.27*
2-hr Postprandial glucose, mmol/l	5.47 ± 0.31	10.08 ± 0.42†
Diabetes medications, n	n/a	n/a	13 Biguanide, 3 Sulfonylurea, 1 GLP-1 analog, 1 DPP-4 inhibitor, 2 Insulin
Systolic blood pressure, mmHg	122 ± 3	114 ± 4	124 ± 3
Diastolic blood pressure, mmHg	73 ± 3	75 ± 3	79 ± 2‡
Taking medication for hypertension, %	17	60‡	59‡
Total cholesterol, mmol/l	4.94 ± 0.16	4.88 ± 0.22	3.92 ± 0.17*
LDL-cholesterol, mmol/l	3.00 ± 0.15	2.94 ± 0.20	2.11 ± 0.16*
HDL-cholesterol, mmol/l	1.49 ± 0.08	1.32 ± 0.10	1.13 ± 0.08†
Triglycerides, mmol/l	0.94 ± 0.14	1.35 ± 0.18	1.53 ± 0.14†
Taking medication for dyslipidemia, %	0	50‡	88†
Statins, %	0	30‡	65*

Values are means ± SE; n, no. of subjects.

NGT, normal glucose tolerance; IGT, impaired glucose tolerance; T2DM, type 2 diabetes mellitus; n/a, not applicable.

^*Significant difference compared with the NGT and IGT groups, P < 0.05.

^‡Significant difference compared with the NGT group, P < 0.05.

^†Significant difference compared with the NGT group, P < 0.01.

Circulating Cell Number Before and After Acute Exercise

CD34+/VEGFR2+ EPCs.

See Fig. 1A. Repeated-measures ANOVA revealed a main effect of group on CD34+/VEGFR2+ EPC number (P = 0.019). In the basal state, CD34+/VEGFR2+ EPC number was 65 and 61% lower in the IGT and T2DM groups, respectively, compared with the NGT group (P < 0.05 for both). These differences persisted after exercise. In response to an acute bout of submaximal exercise, CD34+/VEGFR2+ EPC number increased by 23% in the NGT group (P < 0.01), but there was no change in CD34+/VEGFR2+ EPC number in the IGT or T2DM groups (P > 0.69).

Fig. 1.

Circulating CD34+/VEGFR2+ (A), VEGFR2+ (B), and CD34+ (C) cell number in older adults with normal glucose tolerance (NGT), impaired glucose tolerance (IGT), and type 2 diabetes mellitus (T2DM), before and after a 30-min bout of submaximal treadmill exercise. Values are means ± SE. *Significant difference compared with NGT subjects within the same condition (basal or exercise), P < 0.05. †Significant within-group difference after acute exercise, P ≤ 0.01. ‡Significant linear trend for lower basal VEGFR2+ cell number (P = 0.04) and VEGFR2+ cell mobilization (P = 0.01) across NGT, IGT, and T2DM groups.

VEGFR2+ cells.

See Fig. 1B. There was a group × time interaction effect on circulating VEGFR2+ cell number (P = 0.046) with a significant linear trend for VEGFR2+ cell number (P = 0.04) and the exercise-induced change in VEGFR2+ cell number (P = 0.01) to be lower in a stepwise manner across the NGT, IGT, and T2DM groups, respectively. After acute submaximal exercise, VEGFR2+ cell number increased by 27% in the NGT group (P < 0.01). Whereas VEGFR2+ cell number tended to increase by 21% after exercise in the IGT group, this did not reach statistical significance (P = 0.08). There was no change in VEGFR2+ cell number after exercise in the T2DM group.

CD34+ hematopoetic progenitor cells.

See Fig. 1C. In the basal state, circulating CD34+ cell number was lower in the IGT group compared with the NGT group (P < 0.05); however, CD34+ cell number was not significantly different between the T2DM and NGT groups. There was a tendency for participants taking statins (IGT and T2DM combined) to have higher basal CD34+ cell number than those not taking statins (304 ± 47 vs. 189 ± 25 cells/10⁶ events, P = 0.058). None of the groups experienced statistically significant changes in CD34+ hematopoetic progenitor cell number after acute submaximal exercise, and this did not differ by statin use.

We subsequently conducted analyses to assess relationships between subject characteristics and basal and acute exercise-induced cell mobilization. In the basal state, CD34+/VEGFR2+ EPC number correlated with plasma TG level (r = −0.32, P = 0.03), and CD34+ hematopoetic progenitor cell number correlated with total cholesterol (r = −0.39, P = 0.01) and LDL-C (r = −0.36, P = 0.06). Basal VEGFR2+ cell number did not correlate with any subject characteristic. After acute exercise, the change in CD34+/VEGFR2+ EPC number correlated with V̇o_{2 max} in milliliters per kilogram per minute (r = 0.35, P = 0.02), the change in VEGFR2+ cell number correlated with LDL-C level (r = 0.30, P = 0.04), and the change in CD34+ hematopoetic progenitor cell number correlated with V̇o_{2 max} in milliliters per kilogram per minute (r = 0.34, P = 0.02) and percent body fat (r = −0.33, P = 0.04). When included in analyses accounting for group (NGT, IGT, or T2DM), none of these associations of cell number and subject characteristics remained significant. Neither hemoglobin A1c nor length of time since diagnosis of T2DM was associated with basal or acute exercise-induced changes in circulating cell number.

DISCUSSION

The number of circulating EPCs can serve as an independent predictor of cardiovascular disease outcomes (37) and may also play a role in the microvascular and macrovascular complications associated with T2DM (10). The major findings of this study are that both basal CD34+/VEGFR2+ EPC number and exercise-induced EPC mobilization are reduced in IGT and T2DM compared with NGT controls. We also found a similar trend for VEGFR2+ cells, with lower basal cell numbers and impaired mobilization of VEGFR2+ cells in IGT and T2DM individuals. To our knowledge, this is the first study to report lower CD34+/VEGFR2+ EPC and VEGFR2+ cell mobilization in response to acute submaximal exercise in individuals with IGT or T2DM. As there are known roles for EPCs in angiogenesis, vascular maintenance, and repair (5, 15), lower circulating number and mobilization of these cells in IGT and T2DM suggests a potential mechanism for the vascular impairments associated with impaired glucose metabolism.

Our findings are consistent with previous reports that basal EPC number is reduced in T2DM compared with NGT controls (11, 13, 32). We also find that basal circulating CD34+/VEGFR2+ EPC number is lower in participants with IGT compared with NGT controls, but there is no further reduction between the IGT and T2DM groups. This would seem to indicate that lower circulating CD34+/VEGFR2+ EPC number does not follow the progression of NGT to IGT to T2DM in a linear manner, but that the reduction occurs earlier in the development of impaired glucose metabolism. Given the role of EPCs in vascular homeostasis, this further suggests that low EPC number could contribute to the progression of insulin resistance through vascular mechanisms, and that low EPC number may not simply be a direct consequence of overt T2DM. It should be noted that this is in contrast to two previous reports that basal CD34+/VEGFR2+ EPC number is not lower than NGT controls in participants with impaired fasting glucose or IGT, whereas CD34+/VEGFR2+ EPC number was lower in those with T2DM (11, 13). At this time, the reason for discordance among these studies is unclear; however, the aforementioned studies (11, 13) included smokers as well as younger adults with lower BMI and generally worse lipid profiles compared with subjects in the present report.

With respect to overall CD34+ hematopoetic progenitor cell levels, similar to previous reports, we found that basal CD34+ hematopoetic progenitor cell number is lower in IGT compared with NGT (11, 13). However, we found no difference between the NGT and T2DM groups, which is discordant with those same studies. One possibility for the discrepancy in these findings is differential use of medications by the participants with T2DM. Of the two previous studies, one included only subjects taking no medications (13); while the other study included participants taking medications (11), the proportion of subjects on statins and anti-hypertensive medications was lower than that in the present report. Within this study, a greater percentage of individuals in the T2DM group, compared with the IGT or NGT groups, were on medications such as statins and angiotensin-converting enzyme inhibitors, which are associated with higher circulating CD34+ hematopoetic progenitor cell number (9). As there was a tendency for those taking statins to have higher basal CD34+ hematopoetic progenitor cell number, this could explain the relatively higher CD34+ cell number in the T2DM subjects in this study. However, it is apparent that, if medication use did influence basal CD34+ hematopoetic progenitor cell number in T2DM, it did not appear to enhance mobilization of these cells in response to exercise. Indeed, while there was a numerical exercise-induced increase in CD34+ cell number across groups, there was no statistically significant mobilization of CD34+ cells in any group. This finding is consistent with three other reports in healthy young (35) and older adults (29, 34); however, in the latter study (29), exercise did mobilize CD34+ cells in healthy young adults, indicating a possible effect of age on CD34+ cell mobilization when young and older adults were directly compared.

The study participants with NGT significantly increased the number of CD34+/VEGFR2+ EPCs after an acute bout of aerobic exercise, consistent with previous studies in healthy adults (20, 22, 26, 35). Results for cells expressing the endothelial marker VEGFR2+ were similar. The NGT group, but not the IGT or T2DM groups, had a significant, exercise-induced increase in VEGFR2+ cell number, whereas basal and postexercise VEGFR2+ cell number were lower in a stepwise manner across groups. The aforementioned studies (20, 22, 26, 35) did not independently report the number of VEGFR2+ cells, so these results cannot be directly compared among studies. The congruent findings for EPCs and VEGFR2+ cells within the present study do, however, suggest that the impaired mobilization of CD34+/VEGFR2+ cells may be linked to impaired mobilization of the VEGFR2+ cell population, as there were no differences across groups in the mobilization of CD34+ cells. A previous study linked the degree of glycemic control with lower EPC number and mobilization after an acute cardiac event in individuals with T2DM (4), but our results do not support a link between impaired exercise-induced EPC mobilization and glycemic control per se (assessed as hemoglobin A1c). The IGT and T2DM groups had similar impairments in mobilization, yet the IGT group had hemoglobin A1c values nearly identical to those of NGT participants. Furthermore, hemoglobin A1c values were not associated with basal levels or mobilization of any cell type in our analyses. The precise relationships between mobilization of CD34+/VEGFR2+ EPCs and VEGFR2+ cells, glycemic control, and the progression of IGT and T2DM require further study.

Although we did not directly assess the mechanisms responsible for impaired EPC mobilization, there are several putative mechanisms that could be considered. In healthy people, acute exercise causes release of VEGF (22) and nitric oxide (39) that serve as mobilizing factors for EPCs (1, 6, 19). It is possible that lower levels of, or lesser increases in, VEGF and nitric oxide in response to exercise may have contributed to the impaired EPC mobilization in the subjects with IGT or T2DM. Additionally, diabetes causes bone marrow microangiopathy and dysfunction in animal models (23), which is thought to play a role in low EPC number and mobilization in diabetic humans (2). In mouse models of diabetes, the mobilization of hematopoietic progenitor cells is impaired due to alterations in the bone marrow stroma (14, 38), a finding that corresponds with impaired mobilization of hematopoietic progenitor cells in diabetic humans in response to granulocyte-colony stimulating factor administration (14). The exercise-induced release of EPC mobilizing factors and the “health” of bone marrow warrant further study to determine the causes of impaired EPC mobilization in IGT and T2DM.

This study was designed to identify differences in CD34+/VEGFR2+ EPC mobilization among older adults with NGT, IGT, and T2DM in response to a well-controlled bout of aerobic exercise, but is not without limitations. First, we were unable to match across groups for V̇o_2max, blood pressure, and lipoprotein-lipid profiles. Participants with T2DM had lower V̇o_2max, higher diastolic blood pressure, and lower total cholesterol, LDL-C, and HDL-C levels compared with the NGT group. While these factors could potentially affect EPC mobilization, we feel that this is unlikely, given that the IGT group did not differ from the NGT group in any of these variables, yet the IGT and T2DM groups had similar mobilization impairments. One study reported that higher exercise-induced EPC mobilization was associated with less favorable LDL-C and HDL-C profiles in healthy younger subjects (35), but this was not this case in our older subjects with IGT and T2DM. We cannot, however, rule out a potential contribution of higher TG levels, as both the IGT and T2DM groups had numerically higher TG levels compared with the NGT group. We also did not control for medication use across our groups of participants, so we could not formally analyze the effect of specific medications on the number or mobilization of cells. There are several medications, including statins, that appear to increase circulating EPC number (9). If these did affect CD34+/VEGFR2+ EPC number in our subjects, we may actually have underestimated the lower EPC number and mobilization in the IGT and T2DM groups. While our results do not indicate that statin or antihypertensive medication use increased EPC number or mobilization in the IGT or T2DM groups, we cannot rule out the possibility that certain medications taken by our subjects affected the outcomes. In addition, cells were enumerated only at one time point after acute exercise. While this time point has previously been shown to represent peak mobilization in healthy participants (20), we are unable to definitively determine whether the appearance or mobilization of cells into circulation is delayed in T2DM. Lastly, we only studied sedentary people with no recent history of exercise training. Aerobic exercise training increases EPC number in people with cardiovascular diseases (21, 27, 30), but further research is needed to determine whether exercise training can restore EPC mobilization in older adults with IGT or T2DM.

In conclusion, our results indicate lower circulating CD34+/VEGFR2+ EPC and VEGFR2+ cell number in older adults with IGT and T2DM compared with NGT in the basal state, as well as impairment in the ability of individuals with IGT or T2DM to mobilize these cells into circulation in response to a bout of aerobic exercise. These findings suggest an impairment of CD34+/VEGFR2+ EPC mobilization that could play an underlying role in the link between diabetes, cardiovascular diseases, and capillary rarefaction. Strategies to improve or restore EPC mobilization could help mitigate the decrements in endothelial and cardiovascular health in older adults with IGT or T2DM.

GRANTS

S. J. Prior was supported by a Paul B. Beeson Career Development Award in Aging (K23-AG-040775 and the American Federation for Aging Research). A. H. Lutz was supported by a National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) Training Grant (TK35-DK-095737). This research was supported by a Veterans Affairs Merit Review Award (I01-CX000730), the Baltimore Veterans Affairs Medical Center Geriatric Research, Education and Clinical Center (GRECC), the University of Maryland Claude D. Pepper Center (P30-AG-028747), and the NIDDK Mid-Atlantic Nutrition Obesity Research Center (P30-DK-072488).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).

AUTHOR CONTRIBUTIONS

A.H.L., R.Q.L.-R., and S.J.P. performed experiments; A.H.L., R.Q.L.-R., and S.J.P. analyzed data; A.H.L., J.B.B., R.Q.L.-R., and S.J.P. interpreted results of experiments; A.H.L., R.Q.L.-R., and S.J.P. drafted manuscript; A.H.L., J.B.B., R.Q.L.-R., and S.J.P. approved final version of manuscript; A.H.L., R.Q.L.-R., J.B.B., and S.J.P. edited and revised manuscript; S.J.P. conception and design of research; S.J.P. prepared figures.