**4. Discussion**

The process of arteriogenesis mainly involves the proliferation of ECs and SMCs. Whereas the mechanisms relevant for EC proliferation are relatively well defined, little is known about the mechanism involved in SMC proliferation. Using a murine hindlimb model of arteriogenesis, here, we report that the voltage-gated potassium channel KV1.3, but not the Ca2<sup>+</sup>-gated potassium channel KCa3.1, is of importance for SMC proliferation in collateral arteries. Selectively blocking KV1.3 with PAP-1 resulted in a reduced perfusion recovery (Figure 4), which was associated with reduced numbers of proliferating SMCs (Figure 5). More in-depth in vivo and in vitro studies demonstrated a role for KV1.3 in the expression of the tyrosine kinase receptors *Fgfr1* and *Pdgfrb,* as well as of the transcriptional regulator *Egr1* (Figure 8), all relevant for proper SMC proliferation in arteriogenesis.

Elevated shear stress is the driving force for arteriogenesis [2,3]. This mechanical stress can be sensed by ECs but not by SMCs. The mechanisms how this mechanical force is translated into biochemical signals resulting in endothelial proliferation have been described in [4]. However, little is known about the mechanisms triggering SMC proliferation in arteriogenesis. To address this point, we decided to study the relevance of the potassium channels Kv1.3 and KCa3.1, which have been shown to play a role in smooth muscle proliferation in other experimental settings and processes [20,21]. Our immunohistological investigations demonstrated that both Kv1.3 and KCa3.1 are mainly localized in SMCs of collateral arteries of murine hindlimbs (Figures 2 and 3).

To investigate the relevance of KV1.3 for arteriogenesis, we performed blocking studies employing PAP-1, described as a selective KV1.3 blocker [31]. The laser Doppler perfusion measurements evidenced a significant reduction in perfusion recovery when mice were treated with PAP-1 (Figure 4). Moreover, our histological results showed a significant reduction of proliferating SMCs but not ECs in growing collateral arteries (Figure 5).

Proliferating SMCs are characterized by a reduced expression of the contractile marker αSM-actin *a* [34], which has been demonstrated by our group for the process of arteriogenesis [9] and confirmed in the present study (Figure 5). Blocking KV1.3 during collateral artery growth, however, interfered with the downregulation of αSM-actin (Figure 5). To investigate whether the reduced proliferation rate of SMCs in arteriogenesis was directly related to KV1.3 blockade in SMCs, but not in other cells such as leukocytes, which also play an important role in arteriogenesis [18,35,36], we analyzed the proliferative behavior of primary murine SMCs under KV1.3 blockade in vitro. Our results revealed a correlation between the concentration of PAP-1 in culture medium and the inhibition of mouse primary artery SMC proliferation (Figure 7), attributing KV1.3 with a role in SMC proliferation. Together, our data sugges<sup>t</sup> a direct correlation between the blockade of the potassium channel KV1.3 and the inhibition of SMC proliferation during the process of arteriogenesis. Our data are in line with results from Cidad et al., who demonstrated an inhibition of femoral artery SMC proliferation when Kv1.3 was blocked pharmacologically with PAP-1 or with Margatoxin or when Kv1.3 was knocked down by siRNA treatment [22].

Previous results have shown that SMC proliferation in arteriogenesis is dependent on the activation of FGFR-1. Already in 2003, we had demonstrated that FGFR-1, which was expressed in SMCs but not in ECs, was upregulated in the early phase of arteriogenesis, i.e., within the first 24 h after induction of collateral artery growth by femoral artery ligation, and that blocking this tyrosine kinase receptor with polyanetholsulfonic acid (PAS) interfered with the process of arteriogenesis [10]. A parallel study showed that a combined treatment of rodents with the cognate ligand of FGFR-1, namely FGF-2, and the cognate ligand of PDGFR-β, namely PDGF-BB, significantly promoted the process of arteriogenesis [6]. In that study, an upregulation of PDGF receptors by FGF-2 was suggested and was later confirmed by Zhang et al. [37]. PDGF-BB has been described as a potent inducer of the synthetic phenotype of a SMC and has been shown to act synergistically with FGF-2 to induce the downregulation of contractile genes such as αSM-actin during vascular SMC proliferation [38]. In particular, it has been demonstrated that PDGF-BB activates FGFR-1 via engaging PDGFR-β, thereby mediating the downregulation of αSM-actin and smooth muscle 22 α (SM22- α) expression. The PDGFR-β/PDGF-BB

and FGFR-1/FGF-2 signaling pathways have also been effectively described to promote the upregulation of the transcription factor Egr1 [8], the expression of which was regulated in opposition to that of contractile genes, and which we have found to mediate cell cycle progression in arteriogenesis [9]. In the present study, we found that treatment of mice with the Kv1.3 blocker PAP-1 during the process of arteriogenesis resulted in a downregulation of *Fgfr1*, *Pdgfrb*, and *Egr1* (Figure 8), whereas αSM-actin was upregulated (Figure 5). Accordingly, all genes are regulated in the opposite way as described for proper arteriogenesis [9,10]. Although one could speculate that the impaired expression of *Fgfr1* and/or *Pdgfrb* could be responsible for the impaired expression of their downstream genes, i.e., *Egr1* and αSM-actin, this is somehow unlikely as all genes show a hampered expression at the same point of time. Therefore, we wondered if another factor could be involved in Kv1.3 mediated gene expression. Interestingly, in silico analyses (data not shown) revealed several binding sites for the transcription factor specificity protein 1 (Sp1) in the promoter regions of *Fgfr1*, *Pdgfrb*, and *Egr1*. However, whether Sp1 is indeed involved in potassium channel Kv1.3 mediated gene expression remains to be determined by further studies. Our data indicate that Kv1.3 plays a major role in SMC proliferation, especially in the process of arteriogenesis, by influencing signal transduction cascades associated with the expression of the growth factor receptors *Fgfr1* and *Pdgfrb* and their downstream genes being involved in phenotype switch and cell cycle regulation.

In contrast to our findings regarding PAP-1 administration, treatment of mice with the KCa3.1 selective blocker TRAM-34 did not influence vascular SMC proliferation or differential gene expression in growing collaterals in vivo (Figures 5 and 8). The laser Doppler perfusion measurements, however, evidenced a reduced perfusion recovery after femoral artery ligation (Figure 4). Interestingly, KCa3.1 has been shown to be upregulated by fluid shear stress [39], the driving force for arteriogenesis [2,3]. Moreover, it has been demonstrated by blocking studies employing TRAM-34 in vivo, that KCa3.1 plays a role in EC proliferation during angiogenesis (Grigic, Eichler 2005) and in SMC proliferation, e.g., during atherogenesis [27]. Our study, however, revealed that KCa3.1 is not involved in EC or in SMC proliferation in collateral artery growth (Figure 5). Of course, one could argue that the dose of TRAM-34 used in the present study was not high enough to block KCa3.1 in vivo, but identical dosages were shown to be effective in hampering vascular cell proliferation in a model of intima hyperplasia [25] and atherosclerotic lesions in mice [27]. Together, these data indicate that the mechanisms of SMC proliferation in the different pathophysiological situations are diverging. Indeed, it has been shown by Bi et al. in vitro [20] that KCa3.1 mediated SMC proliferation blocked by TRAM-34 was not associated with any change in expression of *Pdgfrb*, supporting the data of the present investigations (Figure 8). As our laser Doppler perfusion measurements revealed a reduced perfusion recovery upon femoral artery ligation (Figure 4), which was not associated with a reduced collateral artery cell proliferation (Figure 5), we hypothesize that KCa3.1 could overtake a function in EDHF-mediated collateral vasodilation, a well described function of this potassium channel [29,40]. However, further studies are necessary to prove this hypothesis. A similar effect on reduced perfusion recovery upon femoral artery ligation has been described for nitric oxide synthase 3 (NOS3)-deficient mice, also attributing nitric oxide a role in vasodilation during arteriogenesis [41]. In terms of KCa3.1, it could be interesting to know that the Ca2+-channel transient receptor potential cation channel, subfamily V, member 4 (TRPV4) has previously been shown to play a role in arteriogenesis by promoting vascular cell proliferation [42]. TRPV4 is described as shear stress sensitive channel which plays an important role in the regulation of vascular tone by modulating intracellular Ca2+ levels [43]. However, TRPV4 has also been shown to promote collateral artery growth in several animal models [42,44,45]. It has been suggested that upon activation of this receptor, a first increase in intracellular Ca2+ levels could result in EDHF-mediated vasodilation, whilst a prolonged raise could activate transcription factors causing vascular cell proliferation [42,45]. It is tempting to speculate that KCa3.1 is involved in this EDHF mediated vasodilation, but further studies are necessary to investigate this assumption.
