3.2.1. Biocompatibility Tests

Biocompatibility results on NHDF treated with medicinal ALI and GR waters for 24 h are plotted in Figure 3. No significant differences were found, even when multiple comparisons were performed. In fact, ALI and GR viability results were around 100% for all water dilutions. Therefore, mineral medicinal waters studied in this work did not hinder dermal fibroblasts viability, thus determining their biocompatibility. Internal comparisons between ALI and GR concentrations were also performed. Mann–Whitney marked differences in ALI viability were between 50% vs. 0.25% and 25% vs. 0.25% (Figure 3). Su et al. detected an optimal dilution of a concentrated underground mineral spring water that made skin cells to grow better in comparison with other dilutions [107]. By the same token, ALI concentrations higher than 25% (*v*/*v*) could allow slightly better NHDF growth performances with respect to more diluted samples.

PS9 clay mineral and corresponding peloids PS9ALI and PS9GR are plotted in Figure 4 (left) as well as G30 and its corresponding peloids G30ALI and G30GR Figure 4 (right). In a general view, the lower the sample concentration (either clay or thermal mud), the higher the cell viability. None of the samples nor their concentrations presented a cellular viability below 80%. This fact indicated the absence of drastic cellular cytotoxicity within 24 h. With respect to PS9 and G30 clays, it was possible to observe a slight reduction of the cellular viability (with respect to GM) as the clay concentration increased. In fact, when the clay mineral concentration corresponded to 1 mg/mL, cellular viability was close to 80% (for both pristine clays and thermal muds). Nonetheless, statistical analyses indicated no significant differences between PS9 and G30 1000 μg/mL concentrations vs. GM, which confirmed that PS9 and G30 were highly biocompatible against NHDF cells. G30 5 μg/mL concentration showed significant differences vs. GM. Notably, the lower the concentration of the clay, the better the cellular proliferation. It has been suggested by some studies that the precipitation of clay minerals could hinder cell viability

by some sort of physical effect [50,103]. The biocompatibility of Veegum HS © (Bentonite clay) was evaluated by Salcedo et al., finding that at 167 μg/mL of Veegum HS © concentration the viability of Caco-2 cells reduced approximately to 50% [50]. This effect was ascribed to the precipitation of clay particles, which blocked cell membrane channels. According to the authors, montmorillonite had a particle size ranging from 45–75 μm [50], which is a particle size range very similar to that of PS9 and G30. In these conditions, it was argued that if the precipitation of clay particles occurred, cell viability could be compromised at concentrations ≥160 μg/mL. On the contrary, neither PS9 nor G30 showed fibroblasts viabilities lower than 50% even at the highest concentration tested, thus demonstrating to possess a remarkable biocompatibility.

**Figure 3.** Biocompatibility tests (contact time 24 h) of ALI and GR medicinal waters. Viability (%) vs. medicinal water concentration (% *v*/*v*) toward NHDF. GM stands for "growth medium" and refers to control test. Mean values ± s.d.; n = 8. Significant differences are marked as (•) ALI 50% (*v*/*v*) vs. ALI 0.25% (*v*/*v*); (-) ALI 25% (*v*/*v*) vs. ALI 0.25% (*v*/*v*). Mann–Whitney (Wilcoxon) W tests, *p* values ≤ 0.05.

**Figure 4.** Biocompatibility tests (contact time 24 h) of PS9, PS9ALI and PS9GR (left) and G30, G30ALI and G30GR (right). Viability (%) vs. clay concentrations toward NHDF. GM stands for "growth medium" and refers to control test. Mean values ± s.d.; n = 8. Significant differences were reported within the figure (down) as "sig" which refers to "*p*-values".

Internal statistical studies for PS9 and G30 results detected significant differences in G30 1000 μg/mL vs. 50 μg/mL and 5 μg/mL. For PS9, concentrations of PS9 1000 and 500 μg/mL differed both between them as well as with 50 and 5 μg/mL (see Figure 4). These findings mean that, for both clays, smaller clay amounts in contact with fibroblasts were characterized by a higher degree of fibroblast biocompatibility. The biocompatibility results obtained for both clays were in agreement with the existing literature. Sepiolite clay mineral (Pangel S9) has been previously tested by Fukushima et al. towards fibroblasts and osteoblasts [108]. In this case, authors combined 10% of sepiolite with poly(butylene adipate-co-terephthalate) reporting no cytotoxicity. A Tolsa's Group sepiolite clay was also employed by Fernandes et al. [109]. In this case, they reported a reduction of HeLa cell viability of 50% when sepiolite was present in 1000 μg/mL concentration. No such a reduction was found in this work when PS9 was used at the same concentration. Nonetheless, these differences could be ascribed to differences in the type of cell culture [110]. Apart from anti-inflammatory, antibacterial and anti-oxidant properties of a natural Spanish palygorskite, its cytotoxicity against murine macrophages has been reported to start from 300 μg/mL onwards [58,59]. Particularly, authors reported a reduction of 20% viability at the aforementioned concentration. Once again, in the present study, G30 showed higher biocompatibility (at least against fibroblasts) because, even at 500 μg/mL concentration, fibroblasts did not show a viability reduction. In more recent studies, pharmaceutical grade palygorskite (Pharmasorb® colloidal) was able to protect human dermal fibroblasts against carvacrol cytotoxicity at clay concentrations ranging from 8 to 12 μg/mL [6].

Neither PS9ALI nor PS9GR showed significant differences between GM and tested concentrations, thus demonstrating the total biocompatibility of these hydrogels. Statistically, PS9ALI and PS9GR thermal mud showed similar significant differences between concentrations, following the same trend previously described for PS9. In fact, 50 and 5 μg/mL clay concentrations apparently favored cellular viability (Figure 4, left). The agreement among PS9ALI, PS9GR and PS9 biocompatibility results allowed the confirmation of their reproducibility. Both PS9ALI and PS9GR, as well as PS9, shared significant differences of concentration 50 μg/mL due to the higher viability results obtained at this concentration (Figure 4, left).

Regarding G30ALI and G30GR peloids, Mann–Whitney statistical analysis pointed out significant discrepancy in activity between G30ALI GM and 1000 μg/mL concentration vs. 5 μg/mL and between G30GR 1000 μg/mL vs. 50 μg/mL (as specified in Figure 4, right). Since the same trend has been reported for G30 samples, once again it was possible to confirm the reproducibility of the results.

Despite the complexity generated by the statistical analysis, all solid samples (Figure 4) showed a better result for 50 μg/mL concentration, always with cellular viabilities higher than 100%. According to these findings, 50 μg/mL clay concentration was the safest/most ideal one for fibroblast cultures, regardless of the type (PS9 or G30) or the origin of the clay (powder or thermal mud). It is also important not to forget the fact that all the tested concentrations were biocompatible for fibroblasts within 24 h of contact. Due to the conclusions reached with the MTT test, 50 μg/mL was the clay concentration selected for further studies including cell motility assay (wound healing) and proliferation CLSM tests. For spring waters, 50% *v*/*v* concentration was used.
