**3. Discussion**

Arteriogenesis is the natural compensation mechanism through which collateral circulation develops. This formation is stimulated by an increase of shear stress on the endothelium [20]. An increase of blood flow can be achieved by a high demand and walking exercise is the best possibility to maximize the flow physiologically [6]. Therefore, the aim of the study was to establish a training protocol in a murine model of PAD that increases arteriogenesis through exercise. Our results sugges<sup>t</sup> that a FAL surgery in ApoE−/− mice having free access to a voluntary running wheel serves best for future studies of exercise-induced arteriogenesis.

To increase physiologic shear stress on the arterioles it is preferable to develop an exercise program that maximizes intensity.

Initially we compared activity of healthy young C57BL/6 mice trained either with forced exercise by treadmill twice daily or with a 24 h accessible running wheel. Forced exercise is controlled and reproducible, but usually depends on a negative impulse. In models with forced exercise regimens, increased distress values, depressive behavior, inflammation reactions, and elevated corticosterone levels have been shown [26,28–30]. Distress, for example, may limit physiological remodeling normally associated with exercise training in humans [31,32]. The transferability of forced exercise models into humans may thus be limited. In contrast, similar structural and functional cardiac changes occurred in forced and voluntary exercise regimens [33]. The pros and cons of a forced versus a voluntary exercise model are thus not finally delineated.

We further showed that voluntary training leads to a much greater distance covered than forced exercise. Furthermore, voluntary running resulted in a higher running speed. Resulting from this higher exercise dose, a higher training effect response of voluntary running than of forced treadmill walking would be expected. Free-to-access running wheels are an easy way to record and store activity data without disturbing the habitual behavior. Likewise, voluntary access to running wheels permits reasonable adaptation to exercise after surgery and provides an excellent tool to monitor the behavior of mice. We could show that mice do tolerate this voluntary training much better with an increase of the distance travelled compared to forced exercise. For the reasons given above we continued our studies with voluntary training knowing well that this cannot be directly translated to the human situation. There is a discrepancy in intrinsic exercise capacity and response to exercise training between mice and humans. Mice do have a natural drive for running. Humans with sedentary behavior do not push their maximum limit. The focus of our study was to establish a protocol in mice which was adapted to their natural behavior and allowed for future investigations on collateral growth.

Since in PAD a stenosis is progressing over time and involves the whole arterial system, there is an uncertainty if healthy animals can be used to simulate the human patient's illness [3,23,34,35]. An acute occlusion in healthy participants by e.g., arterial emboli or trauma demands an instant intervention. The sudden tissue hypoxia can lead to an acute inflammatory–angiogenic–myogenic response which could result in massive loss of tissue. Patients suffering from PAD usually better tolerate an acute occlusion [36].

In order to increase the similarity to patients presenting PAD we used ApoE−/− mice fed with a HFD, that show numerous plaques throughout the whole arterial system [21,23,24,37] including fat deposition in collaterals.

It was expected that different mouse strains don't have similar responses to voluntary training [27,38]. Our findings showed that C57BL/6 and ApoE−/− animals accepted the voluntary training without a notable difference between the two strains. In this study we showed that there was just a short postoperative readjustment period of 10 days needed to ge<sup>t</sup> mice back to the initial distance. Whether this delay was due to the surgical intervention alone (opening and closing of the skin) cannot be fully excluded, because a sham treatment group without FAL was not investigated.

Next, the voluntary training protocol (running wheel) was tested in both strains to evaluate the reperfusion recovery after FAL.

The LDPI data acquired showed a significant higher perfusion immediately after the occlusion in ApoE−/− mice. This could be explained due to an increased collateral growth as a result of arteriosclerotic plaques in the major arteries [21]. C57BL/6 as well as ApoE−/− mice presented a maximum re-perfusion up to 69% and 77% with no significant difference in between the two strains.

It could be shown that having training possibility allowed ApoE−/− mice to reach the maximum reperfusion alignment one week post-FAL.

In order to correlate the increased perfusion to arteriogenesis we performed histological analyses of collateral tissue of ApoE−/− mice with and without exercise training. It is well accepted that mechanical, cellular, and molecular factors influence collateral growth [39]. Macrophages accumulate around the growing collaterals and cytokine secretion improves that process [40,41]. After training, ApoE−/− mice show a higher accumulation of CD68+ macrophages in the vascular nerve sheath of the collateral vessels than without training.

The presented experimental setup involves atherosclerotic ApoE−/− mice subjected to an acute FAL. Functional as well as histological findings implicate an improvement of arteriogenesis after exercise training in the proposed model.

### **4. Materials and Methods**

### *4.1. Ethics Statement*

Animal handling and all experimental procedures carried out were in full compliance with the Directive 2010/63/EU of the European Parliament on protection of animals used for scientific purposes. Approval was given by the responsible local authority, the Darmstadt governmental council for animal protection and handling (permit reference numbers V54-19c20/15-B2/360, permit date: 30 October 2013). Throughout this study all mice had access to water and food ad libitum.

### *4.2. Femoral Artery Ligation (FAL)*

Twenty-eight male C57BL/6 mice (Charles River, Sulzfeld, Germany) and 40 male ApoE−/− mice were subjected to FAL as described [22]. The contralateral leg served as the reference. During the surgical procedure mice were kept on a heating plate with a temperature of 38 ◦C. Anesthesia was applied using ketamine (120 mg/kg BW) and xylazine (16 mg/kg BW) i.p.. For postoperative analgesia carprofen (5 mg/kg BW) was injected s.c.. After termination of experiments the mice were euthanized by an anesthetic overdose.

### *4.3. Forced Exercise on Treadmill*

Mice were first accustomed by using a treadmill (Exer 3/6, Columbus Instruments, Columbus, OH, USA) with a motivation grid for 15 min/day. This applied small amounts of electric shock for conditioning when the mice stopped running.

After being conditioned, the mice started training at a light intensity with a preset speed of 0.2 m/s leading up to a maximum of 0.3 m/s with no further resistance. Training frequency was 2 times per day. Training was terminated at exhaustions. Exhaustion was defined as pause of more than 5 s at a time, or three times for two or more seconds on the shock pad without trying to ge<sup>t</sup> back on the treadmill.

### *4.4. Voluntary Running Wheels*

To evaluate voluntary training each animal was individually housed in a cage equipped with a free-to-access running wheel. The running wheels were connected to a computer equipped with TSE PhenoMaster V5.1.6 (2014-4115) (TSE Systems GmbH, Bad Homburg, Germany). This setup gave accurate data on each animal, recording time and traveled distances and it allowed evaluation of activity patterns during an adaptation period as well as identifing post-surgery e ffects.
