*3.4. Effects of Metal-Doped Cryogels upon Subcutaneous Implantation*

#### 3.4.1. In Vivo Model Overview

Based on previous surgical procedures [30–32], subcutaneous implantation model in Wistar rats was optimized to assess well-defined effects of the cryogel-formulated TMs. Two subcutaneous square pockets (1.5 × 1.5 cm) were formed at the upper dorsal surface by means of incision at the lower side of the outlined square zone followed by the skin detachment from underlying tissues (Figure 7). Square cryogel sheets (1 × 1 cm) were aseptically inserted into the formed pockets followed by skin suturing. In comparison with skin excision, the model allows for informative analysis of localized effects of the materials on intact host tissues upon biodegradation and release of TMs. Furthermore, earlystage host tissue responses to the TM-doped cryogels at days 5 and 10 after implantation were selected to better compare regeneration and inflammation-related processes. The main skin layers and appendages from the subcutaneous muscle to the epidermis were histologically examined (Figure 7). The analysis of non-doped and TM-doped materials in the same animal enabled improved assessment of specific activities of TMs, as immune and regenerative responses in rats are intrinsically variable [37], allowing one to reduce the number of animals in each group.

#### 3.4.2. Biodegradation of Cryogels

Lateral sections of the skin in contact with the cryogel were stained with Giemsa or Hematoxylin-eosin; the former stain was selected for histological differentiation between the main skin structures designated as numbers from **1** to **6** (Figure 8). Biodegradation of the implanted cryogels was analyzed by quantifying their residual area (Figure 8, **1**). The detected amount of the non-doped cryogel was reduced by ca. 38% from day 5 to day 10 post-implantation, suggesting initial resorption kinetics of the material. The components Zn and, to a lesser extent, Cu increased the rate of degradation of the cryogel, whereas Co had a relatively weak effect on the process. When the Zn dopant was used, ca. 80% of the cryogel implant was degraded at day 5, whereas comparable degradation was achieved for Cu at day 10 (Figure 8B, **1**). The acceleration of the biodegradation of Zn- and Cu-doped materials could be attributed to the TM-mediated increase in the catalytic activity of matrix metalloproteinases (MMP) [38].

**B**

(Ctrl) and TM-doped cryogels.

3.3.2. Angiogenic Differentiation

**Figure 5.** (**A**) Effect of metal dopants (0.2 mM) on proliferation of HUVECs cultured in TM-doped cryogels (MTS assay, 72 h). The data are presented as mean ± SD (*n* = 3, \* *p* < 0.05, \*\*\* *p* < 0.001). (**B**) Corresponding bright-field microscopy images of HUVECs stained with cresyl violet in metal-free

The markers of HUVEC differentiation in the TM-doped cryogels were detected by LSCM. In the presence of TMs, phalloidin CruzFluor™ 647-stained cells showed profound

morphological changes with reorganization of actin cytoskeleton (Figure 6A).

**Figure 6.** LSCM images of HUVECs grown in TM-doped cryogels (0.2 mM) at day 3 post-seeding. (**A**) Cells stained with phalloidin CruzFluor™ 647 conjugate for F-actin (red). (**B**) Immunofluorescence detection of VEGF. (**C**) Immunofluorescence detection of ICAM-2. Cell nuclei were stained with DAPI (blue). (**D**) Relative density of VEGF and ICAM-2 expression per field of view (mean ± SD, \* *p* < 0.05, \*\*\* *p* < 0.001). **Figure 6.** LSCM images of HUVECs grown in TM-doped cryogels (0.2 mM) at day 3 post-seeding. (**A**) Cells stained with phalloidin CruzFluor™ 647 conjugate for F-actin (red). (**B**) Immunofluorescence detection of VEGF. (**C**) Immunofluorescence detection of ICAM-2. Cell nuclei were stained with DAPI (blue). (**D**) Relative density of VEGF and ICAM-2 expression per field of view (mean ± SD, \* *p* < 0.05, \*\*\* *p* < 0.001).

In the Cu-doped cryogel, the cells were characterized by increased spreading with ca. 2.2-fold bigger area than control cells (mean cell area detected was 5807 ± 506 and 2595

approximately 1.6–2.9-fold for VEGF and 2.6–5-fold for ICAM-2 (Figure 6B,C). This further suggests profound angiogenic activity of the TMs, especially Cu, under

Based on previous surgical procedures [30–32], subcutaneous implantation model in Wistar rats was optimized to assess well-defined effects of the cryogel-formulated TMs. Two subcutaneous square pockets (1.5 × 1.5 cm) were formed at the upper dorsal surface by means of incision at the lower side of the outlined square zone followed by the skin detachment from underlying tissues (Figure 7). Square cryogel sheets (1 × 1 cm) were aseptically inserted into the formed pockets followed by skin suturing. In comparison with skin excision, the model allows for informative analysis of localized effects of the materials on intact host tissues upon biodegradation and release of TMs. Furthermore, early-stage host tissue responses to the TM-doped cryogels at days 5 and 10 after implantation were selected to better compare regeneration and inflammation-related

, respectively). According to immunofluorescence analysis, the expression of

*3.4. Effects of Metal-Doped Cryogels upon Subcutaneous Implantation*

± 468 μm<sup>2</sup>

experimental conditions.

3.4.1. In Vivo Model Overview

one to reduce the number of animals in each group.

**Figure 7.** Schematic representation of implantation of cryogel sheets (1 × 1 cm) into subcutaneous pockets (1.5 × 1.5 cm, dotted line) in rats and main skin structures subjected to histological analysis. See Section 2.8.2 for details. **Figure 7.** Schematic representation of implantation of cryogel sheets (1 × 1 cm) into subcutaneous pockets (1.5 × 1.5 cm, dotted line) in rats and main skin structures subjected to histological analysis. See Section 2.8.2 for details.

processes. The main skin layers and appendages from the subcutaneous muscle to the epidermis were histologically examined (Figure 7). The analysis of non-doped and TMdoped materials in the same animal enabled improved assessment of specific activities of TMs, as immune and regenerative responses in rats are intrinsically variable [37], allowing

#### 3.4.2. Biodegradation of Cryogels 3.4.3. Subcutaneous Muscle and Adipose Tissue

Lateral sections of the skin in contact with the cryogel were stained with Giemsa or Hematoxylin-eosin; the former stain was selected for histological differentiation between the main skin structures designated as numbers from **1** to **6** (Figure 8). Biodegradation of the implanted cryogels was analyzed by quantifying their residual area (Figure 8, **1**). The detected amount of the non-doped cryogel was reduced by ca. 38% from day 5 to day 10 post-implantation, suggesting initial resorption kinetics of the material. The components Zn and, to a lesser extent, Cu increased the rate of degradation of the cryogel, whereas Co had a relatively weak effect on the process. When the Zn dopant was used, ca. 80% of the cryogel implant was degraded at day 5, whereas comparable degradation was achieved for Cu at day 10 (Figure 8B, **1**). The acceleration of the biodegradation of Zn- and Cudoped materials could be attributed to the TM-mediated increase in the catalytic activity of matrix metalloproteinases (MMP) [38]. The subcutaneous muscle (SM) and adipose are well-vascularized tissues that host different specialized and multipotent cells and play important roles in cell recruiting, nutrition, biosynthesis of extracellular matrix (ECM) and angiogenesis [39,40]. At day 5, the Zn dopant induced a profound 1.7-fold increase in the thickness of SM (Figure 8, **2**), which is in immediate contact with the cryogel, whereas this parameter was weakly affected by Cu and showed a tendency to decrease in the presence of Co. At day 10, the stimulating effect of Zn on SM decreased, probably in relation to rapid resorption of the Zn-doped cryogel, whereas Cu exhibited comparable stimulation to Zn at day 5. Furthermore, in the presence of Zn and Cu dopants, SM was reorganized so that round rolled muscle structures became unrolled; the latter SM configuration favors cell migration and proliferation upon skin regeneration [40]. Furthermore, SM was greatly disrupted and mostly replaced by an amorphous connective tissue when the Co-doped cryogel was applied (day 10), complicating corresponding analysis (Figure 8B, **2**).

Adipose tissue (Figure 8, **3**) detected as hollow reticular areas (corresponding to washed-out lipid contents of adipocytes) [41] was also enlarged and deeply penetrated into the dermis in the presence of Zn and Cu components, respectively, by ca. 1.5 times (day 5) and over 1.7 times (day 10), compared to the control (non-doped cryogel). This effect of TMs was observed along with morphological change of some adipocytes from round- to ellipsoid-shaped cells. In great contrast to Cu and Zn, Co generally disrupted the adipose structures by day 10 (Figure 8B, **3**).

#### 3.4.4. Vascular System

In comparison with the control material, the Zn- and Cu-doped cryogels did not significantly change the distribution pattern of vascular structures (Figure 8A), which were predominantly localized in adipose and SM tissues, though causing noticeable expansion of these structures (Figure 8B, **4**). At day 5, the skin vascularization assessed by the relative area of the vasculature was increased by the Zn and Cu dopants (Cu > Zn), whereas at day 10 this effect was shown to be dissipated for Zn and maintained for Cu, presumably reflecting resorption properties of the corresponding materials (Zn > Cu). Co, though disintegrating intact SM and adipose layers, at day 10 induced a profound formation of many relatively small vessels in different skin layers, including upper dermis tissues (Figure 8A), which is not typical for intact rat skin. The results suggest all the metal dopants as angiogenic factors as follows: Co ≥ Cu > Zn. This assumes excessive activity of the Co component (also noting decreased degradation of the Co-doped cryogel).

*Gels* **2022**, *8*, x FOR PEER REVIEW 14 of 29

**Figure 8.** (**A**) Bright-field microscopy images of Giemsa-stained cross-sections of skin explants contacted with subcutaneously implanted TM-doped cryogels. (**B**) Mean morphometric parameters of the treated skin (designated as numbers from **1** to **6**) per cross-section (mean ± SD, \* *p* < 0.05). **Figure 8.** (**A**) Bright-field microscopy images of Giemsa-stained cross-sections of skin explants contacted with subcutaneously implanted TM-doped cryogels. (**B**) Mean morphometric parameters of the treated skin (designated as numbers from **1** to **6**) per cross-section (mean ± SD, \* *p* < 0.05).

#### 3.4.5. Dermis

The Cu dopant caused a noticeable thickening of the dermal layer (Figure 8B, **5**) by 1.3 and 1.7 times for days 5 and 10, respectively, attributed to extracellular matrix (ECM) overgrowth as a result of Cu-mediated tissue-vascularization. The Zn and Co dopants

insignificantly impacted this parameter (Figure 8). Furthermore, the relative area of mature collagen in the dermis was quantified using polarization microscopy with Picrosirius red (Figure S4), which stains mature (type I) and immature (type III) collagens in yellow-red and green colors, respectively [42]. The parameter increased as follows: Co ≤ Ctrl < Zn < Cu, as if the Zn and Cu dopants enhanced collagen maturation by ca. 1.3 and 1.6 times, respectively. These data show the increased growth-promoting effect of Cu over Zn toward the dermis. Co, weakly affecting the thickness and collagenization of the dermis, prompted certain disorganization of its intact fibrillar structure, though to a lesser extent than that observed for SM and adipose tissues (Figure 8A), suggesting a specific distance-dependent disturbing effects of the Co dopant on surrounding tissues.

#### 3.4.6. Hair Follicles and Epidermis

At day 5, the Cu-doped cryogel induced a noticeable increase in the number of hair follicles (**6**) in the dermis, which was 5-fold higher than in the control group, whereas the Zn-doped cryogel approached comparable effect at day 10 (Figure 8A). Most follicles in the Cu and Zn groups contained dermal papillary cells and had defined intensively stained outer covering, which is characteristic of the anagen phase [43,44]. The results show that the Cu and Zn dopants have folliculogenic activity (Cu > Zn), and the effect of Zn seems to be delayed, taking into account the rapid resorption of the corresponding matrix. The Co dopant demonstrated a lack of significant stimulation of folliculogenesis.

The skin treated with both the Cu- and Zn-doped cryogels was characterized by a well-structured epidermis, similar to that of the control group but with a somewhat more developed stratum spinosum layer and increased keratinization (Figure S3). In addition, the Cu dopant noticeably promoted epithelial invagination (Figure S3), which is involved in re-epithelization and folliculogenesis [45]. Under the same conditions, the Co-doped cryogel here and there disturbed the epidermis structure (Figure S3) presumably in relation to the disorganization of underlying skin layers.

#### 3.4.7. Immune Cells Appearance

The TM-doped cryogels differently affected the appearance of immune cells with distinct morphological features [46] throughout the whole skin. In particular, large intensively stained cells were attributed to mononuclear macrophages having smoothly shaped elongated or roundish morphology with their interior predominantly occupied by a large nucleus (Figure 9, MM) and dendritic cells, which are irregularly shaped due to noticeable cytoplasmic extensions (Figure 9, DC). Furthermore, multinucleated giant cells usually formed as a result of macrophages fusion appeared as huge elongated cells with multiple nuclei [47] (Figure 9, GC).

Upon treatment with the non-doped and Zn-doped cryogels, mononuclear macrophages prevailed over dendritic cells, though the latter material significantly increased overall immune cell number at day 5 (Figure 9). Unlike the above materials, the Cu-doped cryogel (days 5 and 10) and the Co-doped cryogel (day 5) induced a predominant appearance of dendritic cells over macrophages. Furthermore, the Cu group was characterized by a somewhat decreased number of immune cells compared to the Zn and Co groups. The Co dopant in a specific manner resulted in the appearance of numerous giant cells in the dermis at day 10. The provided histological data (Figures 8 and 9) were additionally supported by the corresponding analysis of H&E-stained skin sections (Figure S2). In particular, the latter analysis of the Co group revealed that giant cells are closely located to small capillary structures filled with erythrocytes attributed to newly forming vessels (Figure 9, H&E).

#### 3.4.8. Immunohistochemical Analysis

Additional immunohistochemical analysis of the treated skin (Figure 10 and Figure S5) showed that all TM-doped cryogels increased the number of CD31-positive cells attributed to ECs by a maximal factor of 3.8 (Cu) and 1.9 (Zn) observed at day 5 and 2.9 (Co, day 10) (Figure 10B). These data are consistent with the angiogenic activity of the metal dopants

in vivo according to the histological analysis (Figure 8 and Figures S2–S4). Furthermore, the TMs differently affected the production of HIF-1α as follows: Ctrl ≤ Zn < Cu < Co. These data show the ability of prooxidant Co and Cu dopants to induce a hypoxia-like state in the surrounding tissues, which apparently underlies their enhanced angiogenic effects, whereas the Zn component somewhat increased the appearance of ECs and vascular structures without a significant overproduction of HIF-1α. *Gels* **2022**, *8*, x FOR PEER REVIEW 15 of 29

**Figure 9.** Representative bright-field microscopy images of Giemsa-stained cross-sections of skin explants contacted with subcutaneously implanted TM-doped cryogels. Mononuclear macrophages (MM), dendritic cells (DC), and multinucleated giant cells (GC) were identified. H&E-stained section was additionally shown in right image of lower panel. **Figure 9.** Representative bright-field microscopy images of Giemsa-stained cross-sections of skin explants contacted with subcutaneously implanted TM-doped cryogels. Mononuclear macrophages (**MM**), dendritic cells (**DC**), and multinucleated giant cells (**GC**) were identified. H&E-stained section was additionally shown in right image of lower panel.

Additional immunohistochemical analysis of the treated skin (Figures 10 and S5) showed that all TM-doped cryogels increased the number of CD31-positive cells attributed to ECs by a maximal factor of 3.8 (Cu) and 1.9 (Zn) observed at day 5 and 2.9 (Co, day 10) (Figure 10B). These data are consistent with the angiogenic activity of the metal dopants in vivo according to the histological analysis (Figures 8 and S2–S4). Furthermore, the TMs differently affected the production of HIF-1α as follows: Ctrl ≤ Zn < Cu < Co. These data show the ability of prooxidant Co and Cu dopants to induce a

3.4.8. Immunohistochemical Analysis

angiogenic effects, whereas the Zn component somewhat increased the appearance of ECs

and vascular structures without a significant overproduction of HIF-1α.

**Figure 10.** (**A**) Immunofluorescent analysis of cross-sections of skin explants contacted with subcutaneously implanted TM-doped cryogels. Simultaneous CD31 Alexa Fluor 488 (green), HIF-1α Alexa Fluor 647 (red) and DAPI staining was performed. (**B**) Mean number of CD31-positive cells (**left**) and relative density of HIF-1α expression (**right**) per field of view (mean ± SD, \* *p* < 0.05). **Figure 10.** (**A**) Immunofluorescent analysis of cross-sections of skin explants contacted with subcutaneously implanted TM-doped cryogels. Simultaneous CD31 Alexa Fluor 488 (green), HIF-1α Alexa Fluor 647 (red) and DAPI staining was performed. (**B**) Mean number of CD31-positive cells (**left**) and relative density of HIF-1α expression (**right**) per field of view (mean ± SD, \* *p* < 0.05).

Furthermore, the Zn- and Cu-doped cryogels were found to significantly increase the dermal level of MMP-2 and MPP-3 involved in the degradation of ECM proteins, the release of ECM-bound growth factors, angiogenesis and tissue remodeling. Similarly, MMP-9 level was elevated in the presence of these materials (data not shown). The effect of metal dopants decreased at day 10 over day 5, as more clearly observed for the rapidly

at day 10, there were distinct structures with highly expressed MMP-2 in the Co group (Figure S6C) attributed to the areas containing giant cells and newly formed capillaries

The cross-linked gelatin cryogel was used as a biodegradable hydrogel scaffold both to examine regenerative activities of incorporated TMs and to develop improved bioinductive materials. As previously shown, different TMs can be stably incorporated into the cryogel [19], presumably involving complex formation with collagen polypeptide groups in accordance with the earlier observation [48]. The advanced macroporous structure of the cryogels favors their bulk interactions with gases, nutrients and living cells, allowing one to assess regenerative factors in three-dimensional tissue-mimicking

Cu, Co, and Zn were studied here as recognized components of solid osteogenic materials [11–13,49]; however, therapeutic effects of these TMs in cryogel scaffolds toward soft tissues have not been compared to date. The composition of TM-containing cryogels was optimized so that the metal dopants did not show a cytotoxic effect, while they were able to stimulate the proliferative and functional activity of mammalian cells (Figure 2). The corresponding effective concentrations of the incorporated TMs lay within the range of up to 1 mM, which moderately affected the structure of cryogels (Figure 1). In particular, the metal dopants comparably increased the elastic over viscous behavior of the cryogels up to ca. 2 times (Figure 1B), attributed to additional TM-mediated crosslinking. These data suggest that the incorporated TMs participate in forming a polymer network of the materials and that the stiffness of the TM-doped cryogels should fit with

Earlier, the macroporous structure of cryogels allowed us to monitor bulk affinity interactions of the materials with fluorescently labeled peptide ligands [34]. Similarly, the

different soft tissues and support cell–matrix interactions [50].

(Figure 9).

**4. Discussion**

conditions [2,34].

Furthermore, the Zn- and Cu-doped cryogels were found to significantly increase the dermal level of MMP-2 and MPP-3 involved in the degradation of ECM proteins, the release of ECM-bound growth factors, angiogenesis and tissue remodeling. Similarly, MMP-9 level was elevated in the presence of these materials (data not shown). The effect of metal dopants decreased at day 10 over day 5, as more clearly observed for the rapidly degrading Zn-doped material (Figure S6). The Co-doped cryogel generally did not increase the content of MMPs compared to the control material (Figure S6A,B); however, at day 10, there were distinct structures with highly expressed MMP-2 in the Co group (Figure S6C) attributed to the areas containing giant cells and newly formed capillaries (Figure 9).

#### **4. Discussion**

The cross-linked gelatin cryogel was used as a biodegradable hydrogel scaffold both to examine regenerative activities of incorporated TMs and to develop improved bioinductive materials. As previously shown, different TMs can be stably incorporated into the cryogel [19], presumably involving complex formation with collagen polypeptide groups in accordance with the earlier observation [48]. The advanced macroporous structure of the cryogels favors their bulk interactions with gases, nutrients and living cells, allowing one to assess regenerative factors in three-dimensional tissue-mimicking conditions [2,34].

Cu, Co, and Zn were studied here as recognized components of solid osteogenic materials [11–13,49]; however, therapeutic effects of these TMs in cryogel scaffolds toward soft tissues have not been compared to date. The composition of TM-containing cryogels was optimized so that the metal dopants did not show a cytotoxic effect, while they were able to stimulate the proliferative and functional activity of mammalian cells (Figure 2). The corresponding effective concentrations of the incorporated TMs lay within the range of up to 1 mM, which moderately affected the structure of cryogels (Figure 1). In particular, the metal dopants comparably increased the elastic over viscous behavior of the cryogels up to ca. 2 times (Figure 1B), attributed to additional TM-mediated cross-linking. These data suggest that the incorporated TMs participate in forming a polymer network of the materials and that the stiffness of the TM-doped cryogels should fit with different soft tissues and support cell–matrix interactions [50].

Earlier, the macroporous structure of cryogels allowed us to monitor bulk affinity interactions of the materials with fluorescently labeled peptide ligands [34]. Similarly, the interaction of H2O<sup>2</sup> with TM-doped cryogels was assessed using an ROS-sensitive DCFDA probe, considering H2O<sup>2</sup> as both an extracellular and intracellular precursor of ROS generated in Fenton-like reactions [29,51]. The hydroxyl radical is particularly recognized as a secondary messenger involved in cell redox regulation via oxidizing thiols and activating transcriptional factors such as HIF-1α [52]. The ROS-generating ability of the Co and Cu dopants revealed in the presence of H2O<sup>2</sup> (Figure 3) should reflect hypoxiamimicking and concomitant angiogenic properties of these TMs [22,24,29]. H2O<sup>2</sup> is known to be released to the site of tissue injury (for example, in association with NADPH oxidase or dual oxidase activities), playing an important regulatory role in the healing process as well as a potential pathological role [53,54]. Therefore, the Co- and Cu-doped cryogels upon tissue implantation are expected to promote the extracellular activity of H2O<sup>2</sup> in contrast to the Zn-doped cryogel.

In comparison with the above reactions (Figure 3), the soluble TMs showed a different ROS-generating profile in 3T3 fibroblasts exposed to 1 or 10 µM compounds in antioxidant-free HBSS (Figure S7). Soluble Cu and, to a lesser extent, Zn, in contrast to Co, were found to induce some ROS overproduction accompanied by a weak decrease in glutathione level in cells. Together, these data support variable condition-dependent ROS-modulating activities of the TMs. The results, in particular, suggest that soluble Zn is also able to increase ROS formation in the fibroblasts, though weaker than soluble Cu (Figure S7), regardless of the ability of Zn co-dopant to inhibit the Fenton-like activity of the Co- and Cu-doped cryogels (Figure 3). The intracellular redox effect of Zn ions can be associated with their interaction with multiple thiol groups of Zn-binding cysteine-rich

proteins, particularly metallothionein [55], activation of ROS-producing mitochondrial lipoamide dehydrogenase (LADH) as well as the inhibition of LADH thiol oxidoreductase [56]. Such prooxidant activity is controlled by MTF-1 transcriptional factor activated by increased Zn concentrations, leading to the overexpression of metallothionein and Zn efflux transporters [55,57,58].

The lack of the ROS-generating effect of soluble Co on 3T3 cells (Figure S7) could be explained by the decreased physiological requirement and cellular uptake of this TM in fibroblasts, noting that in cancerous cells the TM readily caused oxidative stress under similar conditions [59]. Unlike soluble Co, nano- and microparticles of this TM were shown to induce oxidative stress in 3T3 cells probably due to the internalization of the particulate formulations by endocytosis and phagocytosis [60]. Interestingly, the proliferation-stimulating effect of the TM-doped cryogels on the fibroblast cells (Figure 2) is in accordance with prooxidant cellular effects of the soluble TMs (Figure S7), suggesting that the former effect involves intracellular uptake of the metal ions released from the materials. Together, our results demonstrate that the studied TMs possess multifaceted ROS-modulating activities depending on the conditions. Variable prooxidant effects of Cu and Co, both extracellular and intracellular, could be particularly expected, whereas Zn is expected to have dual prooxidant/antioxidant effects.

Among the metal dopants, Cu exhibited the highest stimulation of proliferation of all studied cells (3T3 fibroblasts, HSFs, HUVECs) with as high as 2-fold stimulation for HUVECs (Figures 2 and 5). The Co dopant was almost inactive toward 3T3 cells and HSFs, whereas it significantly increased the proliferation of HUVECs, suggesting increased sensitivity of these ECs to the TMs compared to other types of cells in accordance with reported effects of the GHK-Cu complex [20] and metal chelators [61].

Based on the above results, the mitogenic activity of the TM-doped cryogels could be explained by the activation of ROS-mediated signaling pathways in the cells [52]. Such an effect seems to require an appropriate cell-specific level of overproduced ROS, presumably achieved in both fibroblasts (Figure 2) and HUVECs (Figure 5) grown in the Cu- and Zndoped cryogels. This level should be essential for the stimulation of regenerative processes, particularly angiogenesis [62]. Excessive ROS levels can decrease cell viability and lead to sustained secretion of pro-inflammatory cytokines and endothelial permeability [63]. Therefore, double composition of prooxidant TMs, namely, Cu and Co, demonstrated lower or lack of stimulation toward HUVECs/HSFs compared to Cu alone (Figures 2B and 5A) attributed to excessive generation of cellular ROS. Moreover, the Zn co-dopant abolished stimulation of HUVECs in the Cu- and Co-doped cryogels (Figure 5A) and decreased it for HSFs in the Cu-doped cryogel (Figure 2B). This is attributed to the prevention of ROS production by the Zn co-dopant presumably via the inhibition of the Fenton-like reaction catalyzed by Cu and Co [52] and/or the possible reversal of HIF1-α upregulation induced by the latter TMs [64].

The mitogenic activity of the metal dopants was accompanied by the angiogenic differentiation of HUVECs in the presence of Cu and Co viewed through the morphological rearrangement and overexpression of VEGF and ICAM-2 (Figures 5B and 6). Furthermore, all the TM dopants induced the overexpression of a series of cytokines and growth factors by HSFs (Figure 4), including VEGF, FGF-2 and PDGF, the potent regulators of functional activity of ECs and blood vessels [65]. Moreover, the enhanced release of MCP-1 by the Zn and Cu dopant (Figure 4) may be essential for the recruitment of monocytes/macrophages involved in tissue repair and angiogenesis [66,67]. Of note, the aforementioned cytokines and growth factors can be produced and secreted by ECs, playing important roles in the autocrine regulation and recruitment of other cells to support neovascularization [20]. Thus, our results show that the incorporated TMs allow for the effective induction of pleiotropic growth factors and cytokines in the cryogel-grown cells. Such an activation of cryogels with TMs provides a promising alternative to the immobilization of low-stable recombinant polypeptides such as VEGF [68] and FGF-2 [69], solely or in combination [70], in order to improve regenerative and angiogenic properties of polymeric scaffolds.

The subcutaneous evaluation allows one to understand regenerative or toxicological mechanisms of (bio)materials at molecular, cellular and tissue levels [30–32,71,72], and these are relevant not only for skin but for other organs and tissues. The potential of subcutaneous implantation for the analysis of TM-containing bioactive hydrogels is poorly investigated, apart from a few reports [73,74] not dealing with TM comparison. Such a model was optimized here to establish specific localized effects of macroporous cryogelformulated TMs on various intact skin tissues and appendages (Figure 7).

The degradation rate of implanted TM-doped cryogels was in the order: Zn > Cu > Co ≥ Ctrl (Figure 8, **1**). The profound promoting effect of the Zn dopant is apparently associated with this TM activity as a key cofactor of different MMPs, including MMP-1, MMP-3, MMP-8, MMP-13, MMP-2, and MMP-9 [75]. The peptidase activity of MMP-1 and MMP-9 can be also promoted by Cu [38] in accordance with some lesser effect of the latter dopant. Some delayed effect of the Co dopant at day 10 (Figure 8, **1**) may be attributed to the induction of MMP-1 and MMP-2 expression under oxidative stress conditions [76,77]. Immunofluorescence analysis confirmed a significant increase in the expression of MMP-2 and MMP-3 (as well as MMP-9) in the dermis in the presence of Zn- and Cu-doped cryogels, which, however, had a comparable effect with each other (Figure S6). Therefore, the increased resorption of the former material should be due to the overexpression of MMPs and/or a higher contribution to specific peptidase activity by Zn ions as compared to Cu ions.

The Cu dopant induced a significant thickening of the dermal layer along with an increase in the number of hair follicles recognized as an important source of stem and progenitor cells for skin repair (Figure 8). These data confirm the positive role of this metal in tissue renewal and remodeling in concordance with the reported ability of peptidecomplexed Cu (AHK-Cu) [78] and Cu-doped bioactive glass [79] to promote the formation and growth of hair follicles. Such an effect could be mediated by the stimulation of angiogenesis facilitating the migration of follicle progenitor cells to the dermis [79]. A folliculogenic activity of the Zn-doped cryogel detected at day 10 is in agreement with the recently reported effect of a Zn-containing chitosan-PEG hydrogel in a full-thickness skin defect model in mice [80].

Stimulation of collagen deposition by the Cu and Zn dopants may be associated with effective vascularization or increased collagen stabilization by lysyl oxidase in the presence of Cu [81] as well as with early Zn-mediated attraction of fibroblasts that produce collagen, among other ECM components [80]. Considering some disorganization of skin layers by the Co dopant (Figure 8), its effect on skin functioning is assumed to be negative, presumably due to excessive oxidative stress. It has been previously shown that the use of the Co component at an increased concentration in hydrogels reduces cytocompatibility, collagen deposition and slows down wound closure and re-epithelization [18].

The established ability of metal dopants to increase tissue vascularization (Figures 8 and 10B) is well explained by the generation of ROS (Figure 3 and Figure S7) in combination with the increased tissue level of ROS-sensitive HIF-1α (Figure 10B). It was reported that the modification of mesoporous bioactive glasses by Co [22] and Cu [24] promoted neovascularization by inducing hypoxic cascade and the expression of HIF-1α in human bone marrow stromal cells, yet the induction of HIF-1α was not investigated in vivo. Moreover, the incorporation of Co in a gauze calcium alginate hydrogel was shown to stabilize HIF-1α expression, stimulate angiogenesis and accelerate full-thickness skin wound healing in mice [18]. HIF-1α is a major transcription regulator of VEGF [18,82], which is consistent with a considerable boost in VEGF secretion by fibroblasts (Figure 4) and expression by HUVECs (Figure 6) cultured in Cu- and Co-doped cryogels. In addition, metal-induced VEGF and FGF-2 (Figure 4) are known to stimulate the production of MMPs that degrade basement membrane and ECM, allowing ECs to migrate and form sprouts [70].

The results suggest that the prooxidant activity of the TMs may occur via both extracellular and intracellular reactions. The antioxidant effect of the Zn co-dopant upon the induced ROS production (Figure 3) presumably was not manifested under experimental

conditions in vivo, since the TM was found to cause a weak increase in HIF-1α level and some angiogenic response, which were inferior to the effect of Cu and Co. Other Zn formulations, namely, zinc oxide (ZnO) and zinc peroxide (ZnO2) nanoparticles, embedded into a cross-linked polymeric hydrogel promoted angiogenesis via the generation of ROS, in particular, the onsite production of H2O<sup>2</sup> [83]. The Cu-doped cryogel provided a more physiological vascularization with a predominant hypodermic localization, whereas the Co-doped material, disturbing hypodermic tissues, caused the redistribution of newly formed vessel structures into the dermis. Such a distinct vascularization pattern, however, may result from the degree of prooxidant activity of the TMs (Co > Cu).

Consistent with these results, microvessel growth in the skin was earlier stimulated by Cu-containing hyaluronic acid hydrogel [73] and borate glass microfibers [84]. However, no mechanisms of Cu-mediated angiogenic activity and its comparison with other TMs were provided in these studies. According to the density of CD31-positive cells (Figure 10B), the Cu-doped cryogel should have better angiogenic potential than other proposed Cucontaining materials [73,79,84]. Furthermore, Co-containing collagen/alginate-based hydrogel was earlier demonstrated to exhibit an angiogenic effect on rat mesenchymal stem cells accompanied by the expression of CD31 and VEGF. However, this activity was not significantly affected by the Co component in vivo [85]. Likewise, implanted Co-doped sol-gel bioactive glasses promoted the appearance of blood vessel structures; however, angiogenic effects were not assessed quantitatively [86].

The angiogenic effect of the Zn-doped cryogel without a significant induction of HIF-1α (Figures 8 and 10B) could result from the activation of other pathway(s) implicating increased FGF-2 production (Figure 4), faster material degradation by MMPs promoting cell infiltration, and from the attraction of mononuclear cells (Figure 9). It is noteworthy that the in vivo effects of the Zn-doped cryogel generally decreased on day 10 compared to day 5. This suggests a transient activity of TM-doped cryogels presumably associated with the release of metal ions during degradation so that it disappears after resorption of the material. This also confirms the importance of assessing the effects of TM-doped cryogels in the early stages post-implantation.

In the presence of implanted Zn-doped cryogel, recruitment of mononuclear cells was observed (Figure 9). This process is known to be mediated by MCP-1 chemokine [87], which was stimulated by the material in vitro (Figure 4). It was reported the role of Zn in stimulating the infiltration of monocytes into the damaged tissues, where they can differentiate into macrophages [87], and regulating the transition of macrophages from pro-inflammatory to immune-regulatory tissue repair phenotypes [88]. Recruited macrophages can contribute to the angiogenic process via the degradation of ECM leading to EC migration, the release of angiogenic cytokines, and vessel wall formation by differentiating into ECs [66,67]. The incorporation of Zn into the hydroxyapatite-collagen scaffold promoted osteogenesis and angiogenesis by activating the p38 MAPK signaling pathway in the monocytes, further contributing to the release of TGF-β, VEGF, and PDGF, which stimulate the recruitment of BMSCs and ECs to the injury site [89]. The regulatory roles of Zn on immune system homeostasis have been reviewed [90]. These particularly involve a balance between normal immune response and potential tissue damage in relation with Zn distribution in the extracellular and intracellular compartments.

The detected overall number of immune cells in the case of Cu-doped cryogels was lower compared to the other implanted gels (Figure 9), which indicates that the host immune response to this material occurred smoothly. On day 10 post-implantation, the Co dopant caused the formation of multinucleated giant cells surrounding vascularization zones in dermis (Figure 9), in concordance with the reported enhanced angiogenesis by these VEGF-expressing cells within the implantation bed [47]. Subcutaneously implanted Co-containing bioactive glasses presented a significantly increased number of cell nuclei, morphologically resembling the detected giant cells [86]. This was attributed to the HIF-1αmediated recruitment of proinflammatory cells and coordination of regeneration processes by the material [86]. Given that lymphocytes are essential for giant cell formation [91], it

could be assumed that Co may trigger an adaptive immune response, the mechanism of which should be studied elsewhere.

#### **5. Conclusions**

This study proves biodegradable cryogels as advanced hydrogel scaffolds both to clarify the regeneration-related effects of incorporated TMs in vitro and in vivo and to develop improved biomaterials activated with Cu, Co, Zn or their compositions. Since these effects are assumed to be associated with ROS generation, the ROS-modulating activity of the TMs was characterized. The results suggest Cu and Co as well as their binary compositions as probable extracellular Fenton-like prooxidants, which are antagonized by Zn. All the TMs are capable of intracellular ROS generation to different extents and depending on cellular uptake. The ROS-modulating activities of the TMs and compositions were consistent with the proliferation rate of mammalian cells, the production of regeneration-related cytokines and growth factors by HSFs and the angiogenic differentiation of HUVECs in the cryogels in vitro. The subcutaneous implantation model was optimized to elucidate and compare host tissue effects of TMs containing cryogels. Biodegradation of the materials was promoted mainly by the Zn and Cu dopants, which also induced the overexpression of matrix metalloproteinases, the stimulation of collagen deposition and hair follicle growth. All the TMs, especially Cu and Co, increased the level of ROS-sensitive markers and vascularization, where more physiological vasculature pattern was observed in the case of Cu compared to Co. The latter dopant, unlike Cu and Zn, caused an obvious disturbance in the organization of skin layers along with the appearance of multinucleated giant cells attributed to the excessive prooxidant effect. The results contribute to the understanding of regenerative and adverse activities of hydrogel-formulated TMs and provide the basis for tissue engineering and regeneration applications of TM-containing cryogels.

**Supplementary Materials:** The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/gels8020118/s1, Figure S1: (**A**) LSCM images of dual TM-doped cryogel sheets (top surface) visualized by autofluorescence upon argon laser excitation (488 nm). (**B**) Frequency sweep test (strain deformation δ = 1%) data for the cryogels. Figure S2: Representative bright-field microscopy images of H&E-stained cross-sections of skin explants contacted with subcutaneously implanted TM-doped cryogels. Figure S3: Visualization of epidermis of treated skin according to H&E staining. (**A**) Representative bright-field microscopy images for non-doped and TM-doped cryogels. (**B**) Localized area with dermatitis manifestations caused by Co-doped cryogel. Figure S4: (**A**) Representative polarization microscopy images of Picrosirius red-stained cross-sections of skin explants (dermal area) contacted with subcutaneously implanted TM-doped cryogels. (**B**) Relative area of mature collagen per field of view. Figure S5: Immunofluorescent analysis of cross-sections of skin explants contacted with subcutaneously implanted TM-doped cryogels at day 5 (Zn and Cu) and day 10 (Ctrl and Co). Simultaneous CD31 CruzFluor™ 488 (green), HIF-1α Alexa Fluor 647 (red) and DAPI staining was performed. Figure S6: (**A**) Immunofluorescent analysis of cross-sections of skin explants contacted with subcutaneously implanted TM-doped cryogels (MMP-2 Alexa Fluor 488 (green) and MMP-3 Alexa Fluor 647 (red)). (**B**,**C**) Relative MMP levels in the dermis per field of view. (**D**) The area with localized giant cells (MMP-2, Co-doped cryogel). Figure S7: Effect of dissolved metals on relative levels of (**A**) ROS and (**B**) glutathione in 3T3 fibroblasts according to DCFDA (λex/λem = 490/526) and monochlorobimane (λex/λem = 380/480) fluorescence.

**Author Contributions:** Conceptualization, T.I.A.; methodology, I.N.S. and T.I.A.; investigation, A.A.Y. and M.Z.; formal analysis: R.A.I.; writing—original draft preparation, A.A.Y., M.Z. and T.I.A.; writing—review and editing, M.Z., I.N.S. and T.I.A. All authors have read and agreed to the published version of the manuscript.

**Funding:** The reported study was funded by RFBR and VAST according to the research project 21-515-54003.

**Institutional Review Board Statement:** The animal study protocol was approved by the Ethics Committee of Kazan Federal University (protocol no. 435B from 28.02.2020).

**Data Availability Statement:** The data presented in this study are contained within the article.

**Acknowledgments:** This work was co-funded by the Russian Science Foundation (research project 20-73-10105 in studying ROS-generating activity of TM formulations) and is part of Kazan Federal University (KFU) Strategic Academic Leadership Program. Ilnur Salafutdinov (Institute of Fundamental Medicine and Biology, KFU) is greatly acknowledged for invaluable assistance. The authors thank S.V. Fedosimova and A.G. Daminova (Interdisciplinary Center for Analytical Microscopy, KFU) for LSCM analysis and I.T. Rakipov (Alexander Butlerov Institute of Chemistry, KFU) for rheological analysis.

**Conflicts of Interest:** The authors declare no conflict of interest.
