**2. Results**

C57Bl6J mice were made diabetic using streptozotocin. The diabetic mice showed a significant rise in blood glucose levels (328 ± 38 mg/dL) compared to non-diabetic mice (144 ± 10 mg/dl) 24 h post-ligation. In accordance with this, diabetic mice treated with insulin showed a significant decrease in glucose levels (216 ± 35 mg/dL) compared to the diabetic group (Table 1) at the same timepoint.


**Table 1.** Plasma glucose levels measured 24 h post-Femoral Artery Ligation prior to sacrifice (*n* = 7 in each group).

> \* *p* < 0.05 compared to Control group, \*\* *p* < 0.05 compared to Streptozotocin group.

#### *2.1. Reduced Upregulation of Egr-1 in Growing Collaterals of Diabetic Mice*

The Egr-1 mRNA expression levels were significantly increased in growing collaterals of control mice, diabetic mice, and diabetic mice treated with insulin compared to resting collaterals isolated from the sham-operated side (Figure 1a). However, the increase in the expression levels of *Egr-1* was significantly less pronounced in diabetic mice compared to non-diabetic mice and control mice 24 h post-FAL (2**ˆ** −**ddCT**, 3.19 ± 0.172 versus 22 ± 0.25, *p* < 0.05) (Figure 1a). Treatment of diabetic mice with insulin significantly increased *Egr-1* expression compared to the diabetic mice (3.45 ± 0.25) (Figure 1a), resulting in similar levels as in the control group. Western blot analyses revealed a significantly lower expression of Egr-1 protein in the diabetic group compared to both the non-diabetic control group and the insulin treated group (Figure 1b) 24 h after the surgical procedure.

**Figure 1.** Gene expression studies of (**a**) mRNA levels and (**b**) protein levels of collateral arteries obtained from occluded and sham-operated mice 24 h after the surgical procedure. (**a**) Dot plots representing the results of qRT-PCR analyses (*n* = 5 per group, \* *p* < 0.05 (each group compared to each other group), # *p* < 0.05 compared to corresponding occ group from one-way ANOVA with Bonferroni's multiple comparison test). Results were normalized to the expression level of the 18S rRNA. (**b**) Quantitative analyses (upper panel) and corresponding representative pictures of a Western blot (lower panel) showing the protein expression of Egr-1 as well as of α-tubulin, which was used for normalization, 24 h post-FAL (*n* = 4 per group, \* *p* < 0.05 (each group compared to each other group) from one-way ANOVA with Bonferroni's multiple comparison test).

#### *2.2. Expression of Egr-1 Downstream Genes in Collaterals of Diabetic Mice Was Restored by Insulin Treatment*

The mRNA expression levels of Egr-1 downstream target genes, namely, Intercellular Adhesion Molecule-1 (ICAM-1), urokinase Plasminogen Activator (uPA), and Monocyte Chemoattractant Protein-1 (MCP-1) [18,23,24], were measured via qRT-PCR. Our results revealed decreased expression of ICAM-1 (Figure 2a) and uPA (Figure 2b) in the diabetic group compared to the non-diabetic control group (1.49 ± 0.49 versus 2.84 ± 0.49 and 1.53 ± 0.64 versus 3.2 ± 0.60, respectively). Mice treated with insulin showed a significant rise in ICAM-1 (2.35 ± 0.34) and uPA (3.56 ± 0.10) expression compared to the diabetic group. The expression of *MCP-1* decreased in diabetic mice (2.34 ± 0.15); however, its expression levels did not significantly rise upon treatment with insulin (2.65 ± 0.29) (Figure 2c). As previous results have shown that the transcriptional repressor Splicing Factor-1 (SF-1), which controls smooth muscle cell proliferation, is upregulated in growing collaterals during the process of arteriogenesis [22], we also investigated the expression level of the corresponding transcript. Our data showed a significant upregulation of SF-1 in the diabetic group (2.25 ± 0.36) compared to the non-diabetic group (1.15 ± 0.26), but its expression level was normalized again by insulin treatment (0.72 ± 0.04) (Figure 2d).

**Figure 2.** Gene expression of downstream targets of *Egr-1* in collateral tissues obtained from mice after induction of arteriogenesis was performed by qRT-PCR (*n* = 5 in each group, \* *p* < 0.05 (each group compared to each other group) using a one-way ANOVA with Bonferroni's multiple comparison test). The relative expression is represented as 2 **ˆ** −**ddCT**. The relative expressions of (**a**) *ICAM-1*, (**b***) uPA,* (**c**) *MCP-1*, and (**d**) SF-1 were normalized to the expression level of the 18S rRNA.

### *2.3. Insulin Treatment Restored Vessel Growth in Diabetic Mice*

The luminal collateral vessel diameter, which was measured 7 days post-FAL, was found to be significantly decreased in the diabetic group (23.61 ± 2.1 μm) compared to the control group (30.76 ± 2.5 μm); however, it was restored to control levels by insulin treatment (29.58 ± 1.8 μm) (Figure 3a). Moreover, qRT-PCR results on the cell proliferation marker Ki-67 revealed significantly reduced Ki-67 mRNA expression in the diabetic group (1.36 ± 0.37) compared to the non-diabetic control group (2.35 ± 0.50); however, Ki-67 expression levels increased when STZ mice were treated with insulin (3.02 ± 0.15) (Figure 3b).

**Figure 3.** Upper panel: pictures of superficial collateral arteries of STZ-treated mice 7 days after femoral artery ligation (FAL, occ, left picture) or sham operation (right picture). Arrows indicate pre-existing (resting) collaterals of the sham-operated site or growth-induced collaterals of the experimental site. The ligation (\*) of the femoral artery was executed downstream of the profound artery. Scale bars: 5 mm. Lower panel: (**a**) dot plots represent the inner luminal vessel diameter measured 7 days post-FAL (*n* = 4 per group, at least two collateral arteries per mouse and two sections were evaluated, \* *p* < 0.05 (each group compared to each other group), # *p* < 0.05 compared to the corresponding occ group using a one-way ANOVA with Bonferroni's multiple comparison test). (**b**) Dot plots show the expression level of the proliferation marker Ki-67 in collaterals of control, STZ, and STZ + insulin-treated mice 24 h after FAL. Results were normalized to the expression level of the 18S rRNA (*n* = 5 per group, \* *p* < 0.05 (each group compared to each other group) using a one-way ANOVA with Bonferroni's multiple comparison test).

#### *2.4. Diminished Leukocyte Infiltration Was Improved in Diabetic Mice by Insulin Treatment*

Fluorescent-activated cell sorting (FACS) studies from blood collected from non-ligated mice showed significantly increased levels of CD11b+ cells in diabetic (55%) and insulin-treated groups (77%) compared to the non-diabetic control group (33%) with respect to CD45+ cells, whereas the levels of CD19+ or CD3+ cells decreased in both the diabetic (10% and 14%, respectively) and the insulin-treated group (7% and 30%, respectively) (Figure 4a). There were no significant changes in leukocyte count in the adductor muscle of sham-operated mice in STZ-treated and STZ and insulin-treated mice compared to control mice (data not shown). However, in diabetic mice, there was a significant decrease in CD11b+ (27% versus 13%), CD19+ (45% versus 18%), and CD3+ (55% versus 33%) cells three days post-ligation compared to control mice. Interestingly, insulin treatment increased the number of leukocytes in STZ-induced DM mice (Figure 4b).

**Figure 4.** Dot plots represent the results of FACS analyses on CD11b<sup>+</sup>, CD19<sup>+</sup>, and CD3+ cells performed on (**a**) whole blood of control, STZ, and STZ + insulin-treated mice without any surgical treatment and on (**b**) adductor muscles isolated from mice 3 days post-FAL (**<sup>a</sup>**,**b**: *n* = 3 per group, each group compared to each other group using a two-way ANOVA with Bonferroni's multiple comparison test).
