**3. Results**

#### *3.1. KV1.3 and KCa3.1 are Localized in Collateral Arteries*

Employing a murine hindlimb model of arteriogenesis, we investigated whether the potassium channel KV1.3 or KCa3.1, respectively, were expressed in adductor collateral arteries. Immunofluorescence imaging revealed that KV1.3 and KCa3.1 labelling strongly colocalized with αSM-actin, a marker for SMCs, but weakly with CD 31, which is a marker for ECs (Figures 2 and 3).

**Figure 2.** Localization of KCa3.1 in ECs and SMCs of murine collateral arteries. (**a**) Representative confocal immunofluorescence images of transversal sections of collateral arteries isolated 3 h after induction of arteriogenesis. Tissue sections were stained with an antibody against KCa3.1 (green), together with the SMC marker αSM-actin (red), the EC marker CD31 (grey), and DAPI (blue); (**b**,**<sup>c</sup>**) Scatterplots showing the colocalization analysis, (left lower panel) represents pixels that have low intensity levels in both channels, green and red (**b**), or green and gray (**c**). Quadrant 4 (lower left bottom) represents pixels that are referred to as background and are not taken into consideration for colocalization analysis. Quadrant 1 represents pixels that have high green intensities and low red intensities and Quadrant 2 represents pixels that have high red intensities and low green intensities. Quadrant 3 represents pixels with high intensity levels in both green and red (b) or green and gray (**c**). These pixels are considered to be colocalized. Bright field image is also displayed. (**c**) 3D projection surface rendering is showing the localization of the KCa3.1 with the labelling CD 31 and αSM-actin display on the panel (**c**) right lower position. Scale bar 20 μm.

**Figure 3.** Localization of KV1.3 in ECs and SMCs of murine collateral arteries. (**a**) Representative confocal immunofluorescence images of transversal sections of collateral arteries isolated 3 h after induction of arteriogenesis. Tissue sections were stained with an antibody against KV1.3 (green), together with the SMC marker αSM-actin (red), the EC marker CD31 (grey), and DAPI (blue); (**b**,**<sup>c</sup>**) Scatterplots showing the colocalization analysis. Quadrant 4 (left lower left panel) represents pixels that have low intensity levels in both channels, green and red (**b**) or green and grey (**c**), and these pixels are referred to as background and are not taken into consideration for colocalization analysis. Quadrant 1 represent pixels that have high green intensities and low red intensities and Quadrant 2 represents pixels that have high red intensities and low green intensities. Quadrant 3 represents pixels with high intensity levels in both green and red in (**b**) and green and grey in (**c**). These pixels are considered to be colocalized. Scale bar 20 μm. Bright field image is also displayed. (**c**) 3D projection surface rendering is showing the localization of the KV1.3 with the labelling CD 31 and αSM-actin on the panel (**c**) right lower position.

#### *3.2. Blockade of KV1.3 But Not of KCa3.1 Impaired Arteriogenesis by Inhibiting Collateral SMC Proliferation*

To investigate the functional relevance of the potassium channels for arteriogenesis, the KV1.3 channel was blocked with PAP-1, and the KCa3.1 channel with TRAM-34. The laser Doppler perfusion measurements revealed that both treatments significantly interfered with reperfusion recovery after femoral artery ligation (Figure 4).

**Figure 4.** Laser Doppler perfusion measurements. Line plot (left panel) along with corresponding flux images (right panel) of laser Doppler perfusion measurements. Mice were treated with solvent (control), PAP-1, or TRAM-34, respectively, and the perfusion was calculated by right to left (occlusion (occ) to sham) ratio before, immediately after, and at day 3 and 7 after the surgical procedure (left panel). Data are means ± SEM, *n* = 6 per group. \* *p* < 0.05 (PAP-1 vs. control) and # *p* < 0.05 (TRAM-34 vs. control) from two-way ANOVA with Bonferroni's multiple comparison test. The right panel shows representative flux images of murine paws with the tail in the center. Cold colors (blue, green) indicate low perfusion, whereas warm colors (yellow, red) indicate high perfusion (see scale).

To quantify the effects of channel blockade on vascular cell proliferation, we performed immunohistochemical analyses of the proliferation marker BrdU in transversal sections of collateral arteries at day 7 after induction of arteriogenesis. The results showed that both treatment with the KV1.3 blocker PAP-1 and with KCa3.1 blocker TRAM-34 did not interfered with EC proliferation in growing collaterals. However, the PAP-1 treatment significantly reduced SMC proliferation, an effect that was not observed when mice were treated with TRAM-34 (Figure 5a–c).

During the transition from the synthetic to the proliferative phase, the mRNA expression level of αSM-actin has been shown to be downregulated 12h after induction of arteriogenesis [9], and confirmed in the present study by qRT-PCR analyses (Figure 5d). Interestingly, in TRAM-34 treated mice, the expression level of αSM-actin was comparable to that of the control mice at 12 h after induction of arteriogenesis, however, it was significantly increased in PAP-1 treated mice at the same point in time (Figure 5e).

**Figure 5.** BrdU incorporation and αSM-actin expression in collaterals. (**<sup>a</sup>**,**b**) Bar graphs represent the results of quantitative analyses of BrdU+ ECs (left panels) and SMCs (right panels) in solvent (control), (**a**) PAP-1 or (**b**) TRAM-34-treated mice at day 7 after induction of arteriogenesis. Data are means ± SEM, *n* = 3 mice per group. \* *p* < 0.05 from unpaired student´s t-test. The numbers of BrdU+ cells in control collaterals were defined as 100%; (**c**) Representative picture of a BrdU stained collateral at day 7 after induction of arteriogenesis. Scale bar 20 μm; (**d**,**<sup>e</sup>**) The bar graphs represent the expression levels of αSM-actin (occlusion/sham (occ/sham)) in collateral arteries (**d**) at different time points after induction of arteriogenesis or (**e**) at 12 h after induction of arteriogenesis in control, PAP-1, or TRAM-34 treated mice. The qRT-PCR results were normalized to the expression level of the 18SrRNA. Data are means ± SEM, n > 3 per group. \* *p* < 0.05 from unpaired student's t-test and refers in (**d**) to occ vs. sham.

#### *3.3. KV1.3 and KCa3.1 Blockade Inhibits Mouse Primary Artery SMCs Proliferation In Vitro*

To gain further insights into the role of the potassium channels on SMC proliferation, we performed in vitro investigations on mouse primary artery SMCs. Immunocytological analyses showed that KV1.3, as well as KCa3.1, are localized perinuclear in mouse primary artery SMCs. Somehow, weaker signals were seen in the cytoplasm and at the cytoplasmic membrane (Figure 6).

**Figure 6.** Immunocytological analyses on KV1.3 and KCa3.1 localization in mouse primary artery SMCs. Cells were stained with antibodies against the KV1.3 (upper panels, green) or the KCa3.1 channel (middle panels, green) together with an antibody against the SMC marker αSM-actin (red) and counterstained with DAPI (blue) to show the nuclei. For negative control (lower panels) the primary antibody was omitted. Scale bar 40 μm.

To analyze the effects of KV1.3 and KCa3.1 blockade on SMC proliferation in in vitro mouse, primary artery SMCs were treated with different concentrations of PAP-1 (0.1, 1, and 5 μM) or TRAM-34 (10, 100, and 500 μM), respectively. Interestingly, in in vitro, both PAP-1 and TRAM-34 treatments interfered with SMC proliferation, as shown by the BrdU incorporation assay (Figure 7).

**Figure 7.** Proliferation assay of mouse primary artery SMCs. Mouse primary artery SMCs were cultured with 10% FCS with or without treatment of different concentrations of the KV1.3 blocker PAP-1 or the KCa3.1 blocker TRAM-34. Cell proliferation was investigated by means of BrdU incorporation. Values are expressed as percentages of the positive control (+), i.e., mouse primary artery SMCs stimulated with 10% FCS. For the negative control (–), mouse primary artery SMCs cultured with 2% FCS. Data are means ± SEM, *n* > 6 per group. \* *p* < 0.05 from one-way ANOVA with Bonferroni's multiple comparison test.

#### *3.4. KV1.3 Blockade Repressed the Expression of FGFR-1, PDGFR-ß, and Egr1 in Mouse Primary Artery SMCs In Vitro and During Arteriogenesis In Vivo*

Receptor tyrosine kinases such as FGFR-1 and PDGFR-ß are well described for their relevance in SMC proliferation. Our qRT-PCR results on the expression level of *Fgfr1* and *Pdgfrb* provided evidence that treatment of mouse primary artery SMCs with the KV1.3 channel blocker PAP-1 significantly interfered with the expression of both growth factor receptors, whereas the treatment with the KCa3.1

channel blocker TRAM-34 showed no significant influence (Figure 8a). Moreover, in collateral arteries 12 h after induction of arteriogenesis, a significant downregulation was evident for both *Fgfr1* and *Pdgfrb* when KV1.3 was blocked with PAP-1, while treatment with the KCa3.1 blocker TRAM-34 showed no significant effect (Figure 8b). To further investigate the relevance of the KV1.3 potassium channel for SMC proliferation in vitro and during arteriogenesis in vivo, qRT-PCR analyses were performed on the cell cycle regulator Egr1. Our results evidenced that blocking KV1.3 with PAP-1 in vitro, as well during arteriogenesis in vivo, significantly interfered with the mRNA expression of *Egr1* (Figure 8c,d).

**Figure 8.** The qRT-PCR results of the expression levels of *Fgfr1*, *Pdgfrb,* and *Egr1* in vitro and during arteriogenesis in vivo. (**<sup>a</sup>**,**<sup>c</sup>**) Bar graphs represent the mRNA expression levels of Fgfr1, Pdgfrb, or Egr1 in vitro and (**b**,**d**) in vivo. In vitro mouse primary artery SMCs were cultured without (control) or with 1 μM PAP-1 or 100 nM TRAM-34, respectively. In vivo the expression level of *Fgfr1*, *Pdgfrb,* and *Egr1* were investigated 12 h after induction of arteriogenesis in collateral arteries and are expressed as occlusion (occ) to sham ratio. All qRT-PCR results were normalized to the expression level of the corresponding 18S rRNA. Data are means ± SEM, *n* = 3 per group. \* *p* < 0.05 from one-way ANOVA with Bonferroni's multiple comparison test.
