An active lifestyle is associated with reduced breast cancer risk and improved survival in breast cancer patients (Bodai and Tuso, 2015). Besides a reduction of chronic inflammation, physical exercise has also been shown to regulate hormonal balance including catecholamine, prostaglandins, and cortisol (Keast et al., 1988). Additionally, exercise is known to influence cytokine levels and number, proportions and functions of various immune cells which are involved in the host tumor defense (Lancaster, 2006). One of the most frequently investigated immune cell populations in the context of exercise and cancer are natural killer cells (NK cells) (Pedersen et al., 2016). As a part of the innate immune system NK cells have the potential to identify and eliminate tumor cells without prior priming. Activity of NK cells is dependent on the interplay between its activating and inhibiting receptors. Indeed, higher levels of intra-tumoral NK cell numbers are associated with better prognosis in some types of cancer (Larsen et al., 2014).

Studies have revealed that an acute bout of exercise affect the NK cell distribution and function. In detail, NK cells are elevated immediately after exercise, strongly decrease in the subsequent hours and return to baseline within 24 h (Nieman, 1997), depending on exercise duration and intensity. In view of NK cell function, various studies reported elevated NK cell cytotoxicity levels after a single bout of exercise. On a more mechanistic level, Radom-Aizik et al. (2013) reported acute effects of a brief bout of exercise in 12 participants on the gene and miRNA expression pattern in NK cells. Their findings reveal seven pathways that are related to cancer and cell communication and changed the expression level of 23 miRNAs. The seven pathways are possibly regulated by miRNAs. Activation of NK cells is dependent on the interplay between its activating and inhibiting receptors. Additionally, their results also found upregulation of three KIR activation receptors, KIR2DL4, KIR2DS3, and KIR2DS4, in the circulating NK population in response to acute exercise. Another recent study has found that acute exercise leads to global histone modifications in NK cells and a subsequent increase in the expression of the activating NK cell receptor NKG2D (Zimmer et al., 2015).

In contrast to acute effects of exercise on NK cell mobilization and function, the knowledge about training effects on baseline NK cell function is still sparse. Although inconsistent, some studies reported changes of resting NK cell numbers and proportions as well as altered NK cell cytotoxicity after longer training periods (Gleeson et al., 1995).

Since acute effects of exercise on NK cell cytotoxicity are accompanied by changes of the transcriptome we hypothesized that chronic physical exercise may have a similar effect. Thus, in this pilot study we investigated the chronic effect of 12-week resistance exercise on NK-cell gene expression in breast cancer patients undergoing chemotherapy or radiation therapy.

Transcriptome analysis from one participant of Chemotherapy intervention group (BEATE) was not successful; hence, the total sample size is 19. Participant’s characteristics are shown in Table 1.

There were no significant group × time interactions after correcting for multiple testing (all adjusted p-values > 0.9). However, we could explore the top candidates, e.g., the attached heat map contains the genes with raw p < 0.001 (Figure 1).

One of the interesting candidates is the TET1 gene (10–11 translocation methyl cytosine dioxygenase 1). The expression levels after the treatment tend to be higher in the exercise group compared to the non-exercise group (raw p = 0.0008, adjusted p = 0.990912).

FXYD4 is an ion transport regulator gene expression and has a significant difference in expression between exercise and non-exercise groups with a higher expression in exercise group (raw p = 0.0003, adjusted p = 0.990912).

The expression of VAPA, Homo sapiens vesicle-associated membrane protein (VAMP)-associated protein A was higher in exercise group compared to the non-exercise group after treatment (raw p = 0.00007, adjusted p = 0.990912).

The expression levels of Homo sapiens WD repeat domain 68 (WDR68) after treatment tend to be higher in the exercise group compared to the non-exercise group. This effect was still visible when analysis was performed without the “outlier” sample in the non-exercise group with very low expression before treatment (raw p = 0.0007, adjusted p = 0.990912).

Homo sapiens chromosome 20 open-reading frame 199 (C20orf199) has a significantly higher expression in exercise group compared to the non-exercise group after treatment (raw p = 0.0001, adjusted p = 0.990912).

The gene expression of SLC22A10 is significantly higher in exercise groups compared to the non-exercise group after treatment (raw p = 0.0002, adjusted p = 0.990911).

After a 3-month exercise intervention several genes showed higher expression compared to the control group. However, after adjusting for multiple testing there were no significant differences in expression observed. This is the first study investigating the chronic effects of resistance training on NK cell gene expression profile in breast cancer patients undergoing adjuvant therapy.

As mentioned in the section “Introduction,” the chronic effect of resistance exercise training on differential gene expression of NK cells has not yet been investigated. Dias et al. (2015) found that after 18 weeks of aerobic endurance training 211 gene transcripts involved in cell cycle regulation, proliferation, and development of immune cells were changed in PBMCs. A study conducted by Liu et al. (2017) identified 72 transcripts in PBMCs involved in encoding ribosomal proteins and oxidative phosphorylation. The study was conducted on young athletes compared to non-athletes with a false discovery rate <0.05 (Liu et al., 2017). However, one should keep in mind that PBMCs are a heterogeneous mix of immune cells. Detected changes in gene expression over time may also be driven by alterations of cell proportions.

Natural killer cells subsets make up with 90% as surface markers of CD56-dim, classified as cytotoxic and 10% as CD56-bright, classified as immune-regulatory and least mature (Timmons and Cieslak, 2008). Both these subsets are localized in different compartments and also differ in their expression of β₁ and β₂-adrenergic receptors with higher expression on CD56^dim NK cells and are therefore more responsive to exercise induced epinephrine (Bigley et al., 2012; Pedersen et al., 2016). In fact, Shephard (1995) has shown that exercise training leads to changes in immune cell subsets. Moreover, differences in results may also be explained by the different exercise regimes (aerobic vs. strength) as well as intensity and duration of exercises performed. The results of these studies do not provide evidence that chronic exercise, especially chronic resistance exercise impacts the transcriptomic profile of NK cells in breast cancer patients.

In contrast to chronic effects of exercise, acute exercise has been reported to induce chromatin remodeling in NK cells and changes the expression of NK cell receptors (Zimmer et al., 2015; Schenk and Zimmer, 2018). Zimmer et al. (2014) also showed that a single bout of exercise could regulate the cytokine levels which consequently affect the epigenome of NK cells. Crisafulli et al. (2013) conducted studies on chronic exercise effects did not show any change in NK cell cytotoxicity in vitro in post-menopausal overweight women compared to the stretching control group. Recent animal studies by Pedersen et al. (2016) have shown that NK cells were mobilized by epinephrine and were redistributed to tumors by IL6-dependent manner.

Few studies that have reported NK cell gene expression pattern, migration potential, cytotoxicity, and change in NK cell subset population are in context of acute physical exercise, data regarding chronic exercise are few and contradictory.

Our study showed that the exercise intervention had negligible effects on the gene expression. The transcript TET1 is responsible for DNA methylation and its expression seems to be slightly higher post exercise. The TET1 gene (10–11 translocation methyl cytosine dioxygenase 1) is an epigenetic marker responsible for the conversion of the modified DNA base 5-methylcytosine (5-mC) to 5-hydroxymethylcytosine (5-hmC) and is the initial step for active DNA demethylation in mammals. Reduced levels of TET1 have been associated with higher proliferation levels (Takai et al., 2014). The other interesting candidates were as follows: FXYD4, an ion transport regulator. VAPA, Homo sapiens VAMP-associated protein A has a role in RRAS signaling which in turn attenuates integrin beta-1 (ITGB1) activation at the cell surface (Weber-Boyvat et al., 2015). Homo sapiens WDR68 interacts with other enzymes and may play a significant role in a signaling pathway regulating cell proliferation (Gangula and Maddika, 2013). C20orf199, SLC22A10 is a solute carrier family protein. Apart from TET1, the above mentioned genes are well documented but may not have a relevant function in respect to NK cells function or cytotoxicity. Their function in NK cell still needs to be elucidated. The other transcripts with a significant p-value prior to adjustments for multiple testing are hypothetical predicted protein coding genes, and our literature search did not show any relevant known function. They neither interact with each other nor do they form a pathway pattern.

Physical exercise indeed disturbs the immune system homeostasis. However, our study together suggests that these changes are probably driven by acute rather than chronic exercise effects. Additionally, chronic exercise may not induce any changes in NK-cell gene expression in breast cancer patients undergoing medical treatment.

One of the possible explanations for our findings can be, since medical treatments (chemotherapy or radiation therapy) strongly influence gene expression in various tissues (Klebanoff et al., 2005); this can in turn affect the NK cell gene expression pattern and hence affect our findings. It is indeed a limitation in our study that we do not have healthy control group to compare our findings.

It is of importance to note that similar to other investigations in this field our study was limited by small number of participants and should thus be considered a pilot study. Another limitation is that we did not measure NK cell functions or the change in NK cell number or proportions. However, ours is the first study that explored effects of chronic resistance training program as well as cancer therapy effects on NK cell gene expression of breast cancer patients. Further research in this context should include acute and chronic assessments of NK cell function, transcriptomics, and epigenomics in larger sample sizes with different exercise modalities.

All authors listed have made a substantial, direct and intellectual contribution to the work, and approved it for publication.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.