Article Text

Download PDFPDF

Short intensive exercise increases the migratory activity of mesenchymal stem cells
  1. A Schmidt1,
  2. S Bierwirth1,
  3. S Weber1,
  4. P Platen2,
  5. T Schinköthe3,
  6. W Bloch1
  1. 1
    Department of Molecular and Cellular Sport Medicine, German Sport University Cologne, Cologne, Germany
  2. 2
    Ruhr University Bochum, Bochum, Germany
  3. 3
    Laboratory of Molecular Oncology, Universidad National de Quilmes, Buenos Aires, Argentina
  1. Dr A Schmidt, Department of Molecular and Cellular Sport Medicine, German Sport University Cologne, Am Sportpark Müngersdorf, 50933 Cologne, Germany;{at}


Background: It is known that exercise increases the number of stem cells within the circulation; however, it has not been clear which cellular processes are responsible for this increase. To answer this question, we analysed the influence of athletes’ blood sera on human mesenchymal stem cells (MSC).

Methods: Sera were taken before and after short intensive exercise. As cellular parameters of MSC proliferation, apoptosis and migratory activity were analysed.

Results: A change in stimulation of proliferation or apoptosis was not seen after exercise. In contrast, the migratory activity of MSC was significantly increased after exercise. To identify potential factors that could be responsible for this effect, we also analysed the semiquantitative serum concentration of 120 cytokines. Of these factors brain-derived neurotrophic factor, cutaneous T-cell-attracting chemokine, epidermal growth factor receptor, glucocorticoid-induced tumour necrosis factor receptor ligand, growth-regulated oncogene-α, interleukin (IL)1a, IL6, IL8, IL15, pulmonary and activation-regulated chemokine and soluble tumour necrosis factor receptor II showed a significant increase whereas migration inhibitory factor howed a decrease in concentration after exercise.

Conclusions: IL6 is known to stimulate migration in MSC. It is recognised that contracting skeletal muscles synthesise and release IL6 into the systemic circulation in response to exercise. We therefore hypothesise that there is a direct relationship between exercise, IL6 release and stem cell recruitment.

Statistics from

What is already known on this topic

Exercise increases the number of cells in the circulation.

What this study adds

Short intensive exercise leads to the secretion of chemokines and cytokines that are responsible, among others, for increased migratory activity of mesenchymal stem cells.

Sport-induced remodelling of tissues can be controlled by local or systemic cellular processes. In cases of local controlled adaptation as a result of increased requirements, this can give rise to new tissue such as muscle fibres or blood vessels from existing proliferating stem cells. Alternatively, stem cells will be attracted to the region of need. In this case, cells have to pass the circulation to reach their target area. In case of blood vessels supply, there is increased evidence that circulating stem cells are responsible.15 Using animal models, Laufs et al showed that exercise increases the number of circulation stem cells.6 There can be various reasons for an increase in stem cells within the circulation. The number of stem cells can be increased by enhanced proliferation. Cell accumulation is also possible by decreasing cell senescence. The third method of increasing the number of stem cells within the circulation is via increased migratory activity, causing more stem cells to leave their home environment and enter the circulation.

This study aimed to investigate which cellular processes are responsible for the described relationship. Therefore, athletes were exhausted within a short period of time, and the effect of their blood sera on stem cells was analysed. To define the influence of exercise, samples were taken before, directly after training and after 1 hour.

As stem cell source, we used human MSC, which are the progenitors of all connective tissue cells. In adults of multiple vertebrate species, MSC have been isolated from bone marrow and other tissues, expanded in culture, and differentiated into several tissue-forming cells such as bone, cartilage, fat, muscle, tendon, liver, kidney, heart and even brain cells.713 It is known that MSC are mobilised into the peripheral blood by hypoxia.14

Comparative studies have shown that MSC from different sources show comparable characteristics even though the expression profile of a large number of genes differs.15 The most commonly used source for MSC is the bone marrow.


The study was approved by the local ethics committee and conformed to the Declaration of Helsinki.

Exercise and collection of cells

Five athletes underwent exercise using a cycle ergometer. After a warm-up for 12 minutes, the athletes exercised to exhaustion within a maximum time of 8 minutes. Blood samples were taken before warming up, directly after exhaustion was reached and again 1 hour later. The freshly collected sera were used for ex vivo cultivation of human MSC. In independent samples, MSC from the same source were cultivated over 12 hours.

Cell culture and manipulation

MSC of human origin were obtained from bone marrow of the femur shaft and head. Bone marrow was filtered through a 70 μm mesh before density gradient centrifugation (Ficoll-Paque Plus; Amersham Pharmacia Biotech, Uppsala, Sweden). The medium (α-minimal essential medium, 20% v/v fetal calf serum, 200 μmol/l L-glutamine, 100 U/ml penicillin, 100 U/ml streptomycin) was changed after 2 days of culture in 95% humidity and 5% CO2. Cells thus maintained were used for three passages. For every passage or experiment, cells were plated at 2000 cells/cm2 and the medium was changed twice a week.


Cells were treated with 0.25% Triton-X 100 and 0.5 mol/l NH4Cl in 0.05 mol/l Tris-buffered saline (TBS), for cell-membrane permeabilisation. Blocking was performed with 5% bovine serum albumin (BSA) in TBS for 1 hour at room temperature. Primary antibodies used were anti-active caspase-3 (1:500; Pharmingen, San Diego, CA, USA) and anti-Ki67 (1:150; Dianova, Hamburg, Germany), which were diluted in 0.8% BSA and incubated overnight at 4°C. The secondary antibody was goat-anti-rabbit conjugated Cy2 (1:200;), which was incubated for 1 h at room temperature.


To analyse MSC migration, a modified Boyden chamber with a 24-well insert system (HTS Fluoro Blok; Falcon Becton Dickinson GmbH, Heidelberg, Germany) with 8 μm pores was used; 104 cells were used per insert. After incubation for 8 hours, the cells were fixed using 4% w/v paraformaldehyde. After membrane was transferred onto a coverslip, stained using mounting medium containing DAPI (Vectashield, Vector Laboratories, Burlingame, California, USA) and the total number of migrated cells counted.

Serum analysis of cytokines and chemokines

A cytokine/chemokine array kit (Ray Biotech Inc., Norcross, Georgia, USA) was used to investigate a panel of 120 secreted cytokines and chemokines in the serum from healthy patients and those with breast cancer, using the manufacturer’s recommended protocol.


For densitometry analysis the grey-scale values were measured using a computer-based image analysis system (ImageJ V.1.33; National Institutes of Health; All values were converted into relative intensities with respect to the range between positive and negative controls.

Statistical analysis

All data are presented as mean (SD). Data analysis was performed using analysis of variance with Bonferroni posthoc test and/or Student t test for unpaired data. Significance was set at p<0.05.



After cultivation, the number of proliferative cells was analysed using fluorescence-immunohistochemical staining with Ki-67. Ki-67, which is not expressed in resting cells, was found to be positive in 58 (7)% within the nucleolus of the MSC cultivated using the sera taken before warm-up. In 61 (5)% of samples, Ki-67 indicated unchanged expression with the sera taken directly after exhaustion (p = 0.247). Sera taken 1 hour after exercise also did not show any significant influence on the percentage (56 (9)%) of Ki-67-positive cells (p = 0.299; fig 1A).

Figure 1

Influence of blood sera taken from athletes before, directly after exercise and 1 hour later. (A) For cell expression, the proliferation-associated protein Ki-67 was analysed using immunohistochemistry. (B) The number of apoptotic cells was quantified using immunohistochemical staining against activated caspase 3. (C) The migratory activity of the cells was analysed using a Boyden chamber. *Significant change.

As a marker for apoptotic cells, immunohistochemical staining for the cleaved and therefore active form of caspase-3 was used. As found for Ki-67 also, the number of cells positive for cleaved caspase-3 had not significantly changed directly after exhaustion (p = 0.117) or 1 hour later (p = 0.449; fig 1B).


For analysing the influence of exercise on the migratory activity of MSC, the cells were plated into the upper well of a Boyden chamber in the presence of different sera. After 8 hours, the cells were fixed and the number of cells that had migrated through the membrane was counted. Using the pre-exercise sera, 110 (31) cells migrated through the membrane. The sera taken directly after exhaustion increased the migratory activity of the analysed MSC by 120% (p = 0.021), and the sera taken 1 hour after exercise increased the value by 132% (p = 0.015; fig 1C).


The influence of sera on the behaviour of cells must be mediated by soluble factors. To answer the question of which factor within the sera could be responsible for the observed effect, we analysed the relative concentration of a broad range of secreted cytokines using a cytokine array, which can simultaneously analyse 120 soluble factors. These arrays provide semi-quantitative results for relative concentration changes. Examples of these membranes are given in fig 2. Of the 120 factors, 12 showed a significant change in concentration. The concentrations of 11 factors were significantly increased directly after the performed exercise, and only the serum concentration of migration inhibitory factor was decreased. The greatest increases were found for brain-derived neurotrophic factor (45%, p = 0.008), interleukin (IL)8 (40%, p = 0.015) and IL6 (39%, p = 0.47). The other factors showed an increase between 11% and 27%, and all were significant (fig 3).

Figure 2

Semi-quantitative analysis of cytokine blood serum concentrations. Rectangles, positive and negative controls; double spots, interleukin-8 and migration inhibitory factor. Serum profile (A) before exercise and (b) directly after exercise, taken from the same donor.

Figure 3

Overview of all factors with a significant change in relative concentration. Relative concentrations before (white bars) and after (black bars) exercise.


It is known from animal models that exercise increases the number of stem cells within the circulation5 and that hypoxia increases the number of circulating MSC in periphery.14 However it was not known which cellular reasons were responsible for such an increase. The present study has shown for the first time that exercise has a direct influence on the migratory activity of human MSC. MSC are domiciled within the bone marrow;13 by increasing the migratory activity of MSC, the cells leave the bone marrow and enter the circulation. This explains the relationship between exercise and the increase in stem cells within the circulation, and the study shows that neither increased proliferation nor decreased apoptosis is responsible for this effect.

Migration of stem cells is mediated by cytokines.16 We therefore analysed which factors could be responsible for the observed increase in migratory activity. Using the cytokine array, we found that the relative concentration of 11 factors was significantly increased in the sera after a short period of intensive exercise. Of these factors, IL6 is known to mediate migration in and increase the mobility of MSC.16 This observation leads to new opportunities in monitoring the effect of intensive exercise. Exercises that are focused on tissue remodelling or in formation of new tissue are dependent on the presence of stem cells. Based on these results, changes in serum concentrations of IL6 could be a useful marker for monitoring.

IL6 is mostly associated with anti-inflammatory effects of regular exercise.17 18 It is recognised that contracting skeletal muscle synthesise and release IL6 into the interstitium and the systemic circulation in response to exercise. Prolonged exercise involving a significant muscle mass in the contractile activity is necessary in order to produce a marked systemic IL6 response.19

A relationship between IL6 and performance has been under discussion for some time. Previous theories suggested that IL6 increased the insulin sensitivity of muscles, resulting in a direct increase in power. However, more recent studies have shown that this does not play a role in the relationship between IL6 and performance.20 Our study suggests a new relationship, in which increased levels of IL6 have a positive effect on performance. IL6 is produced and secreted from active contractile muscle cells. IL6 itself is a chemotactic cytokine that attracts MSC, a prerequisite of building up new tissues. This relationship might be a key player in improved performance.



  • Funding: WB received a grant for this study from the Novartis Foundation.

  • Competing interests: None.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles