*2.9. Release of G1Phy*

The corresponding membrane (12 mm diameter) was immersed into 10 mL of Tris-HCl 0.1 M buffer (pH 7.4) and incubated at 37 ◦C. Aliquots of 2 mL were taken at different periods of time (1,2,4,7 and 14 days) and replaced with fresh media. The different aliquots were diluted to 5 mL with Milli-Q water and measured by ICP-OES. For each period of time and sample, a minimum of four replicates were measured and averaged. Data were expressed as mean ± SD.

### *2.10. Cell Studies*

### 2.10.1. hMSCs Isolation and Culture from Adipose Tissue

hMSCs used in this study were isolated from human abdominal fat obtained from healthy donors undergoing liposuction plastic surgery. Ethical approval for the study was obtained from the Ethics Committee (number: 02/022010) of the Clinical University Hospital of Málaga, Spain. Informed patient consent was obtained for all samples used in this study. hMSCs were isolated from human adipose tissue and characterized as previously reported [30,31]. Cells were cultured in high-glucose Dulbecco's modified Eagle's medium (DMEM; Sigma-Aldrich, St Louis, MO, USA) supplemented with 10% fetal bovine serum (FBS; Sigma-Aldrich), 100 U/mL penicillin and 100 mg/mL streptomycin (Invitrogen Inc., Grand Island, NY, USA) at 37 ◦C in a humidified atmosphere containing 5% CO2. Medium was regularly changed every 3 days. At 80% of confluence, cells were subcultured. For all the experiments cells were used between passages 4 and 6.

### 2.10.2. hMSCs Culture in Hydrogel Membranes

hMSCs isolation and culture from adipose tissue were performed as a described in Section 2.10.1. Hydrogel membranes (12 mm diameter) were sterilized by immersion in 70% ethanol aqueous solution for 1 h, washed several times in PBS and then subjected to UV light (Philips, Pila, Poland) for 20 min on both sizes. Then, hydrogel membranes were incubated in 24 well plates with complete medium overnight before cells were seeded. hMSCs suspension containing 30,000 cells in 200 µL of medium was slowly dropped onto the surface of each membrane and incubated for 2 h at 37 ◦C. After that, 1 mL of fresh medium was added to each well plate. All samples were incubated under a 5% CO<sup>2</sup> atmosphere at 37 ◦C. The culture medium was replaced every 2 days and the hydrogel membranes were processed for subsequent analysis.

### 2.10.3. Cell Viability Assay

Cell viability was determined on days 1, 7 and 21 using Live/Dead™ Viability/Cytotoxicity Kit (Invitrogen Inc., Grand Island, NY, USA). The hydrogel membranes were incubated in PBS containing Calcein AM (2 µM) and ethidium homodimer (4 µM) at 37 ◦C for 30 min to stain live and dead cells, respectively. Membranes were imaged by confocal microscopy (Nikon Eclipse Ti-E A1, Amsterdam, The Netherlands) and analyzed using NIS-Elements software (Amsterdam, The Netherlands).

### 2.10.4. Cell Proliferation Assay

Cell proliferation was analyzed using AlamarBlue® assay (Bio-Rad Laboratories, Inc., manufactured by Trek Diagnostic System., Hercules, CA, USA) after 1, 5, 7, 14 and 21 days. The hydrogel membranes were incubated with AlamarBlue® solution at 37 ◦C for 3 h. Fluorescence of reduced AlamarBlue® was determined at 530/590 nm excitation/emission wavelengths (Synergy HT, BIO-TEK, Winooski, VT, USA).

### 2.10.5. Environmental Scanning Electron Microscopy (ESEM)

The hydrogel membranes were analyzed using a variable-pressure equipment FEI, mod. Quanta 400 (OR, USA). The analysis was performed to characterize the surface structure of the membranes and cell growth after 21 days in culture. Samples were fixed with 2% glutaraldehyde and, then, were rinsed in 0.1 M cacodylate buffer and incubated overnight at 4 ◦C. For critical point the samples were then maintained with osmium tetroxide 1% at room temperature during 1 h and dehydrated in a series of ethanol solutions (50%, 70%, 90% and 100%) by soaking the samples in each solution for 15 min. Subsequently, samples were critical point dried (Anderson, 1951) in a desiccator (Leica EMCPD300, Wetzlar, Germany) and covered by evaporating them in a carbon evaporator (Emitech K975X).

### 2.10.6. Statistical Analysis

All graphed data represent the mean ± SD from at least three experiments. Two-tailed Student T test analysis were performed for Ch/HA and Ch/HAMA samples with respect to Ch ones at each time point at significance level of \*\* *p* < 0.01, and for Ch/HA samples with respect to Ch/HAMA samples at each time point at significance level of (## *p* < 0.01).

### **3. Results and Discussion**

### *3.1. Physicochemical Characterization and Viscoelastic Properties of Membranes*

Elemental composition of the membranes was determined by elemental analysis. Table 1 shows the theoretical and experimental elemental compositions (C, H, and N) for the different membranes. Experimental compositions correlated very well with those calculated theoretically, which revealed the absence of impurities. Table 1 also shows a decrease of the experimental C/N value (5.25 ± 0.07) respect to the theoretical one (6.1) for Ch/HA membranes. In case of Ch/HAMA membranes, the experimental C/N ratio (5.83 ± 0.15) approached the theoretically expected (6.1). The decrease of C/N ratio in semi-IPN systems could indicate a possible release of HA from these membranes after washing steps since this phenomenon was not observed in IPN systems. This means that the covalent crosslinking of HAMA mediated by UV-light seemed to retain the HA polysaccharide in the IPN. In addition, this fact was confirmed by a decrease in the yield percentage of Ch/HA membranes. Nevertheless, both semi- and IPN systems can be considered promising candidates for tissue regeneration since both of them contain HA, an essential component of the native ECM [7]. Likewise, the presence of HA in the membranes will help to maintain the functionality and characteristic structure of regenerated tissues, which is essential for the final success of these scaffolds [6,11]. It is important to consider that the concentration range in which HA can provide these beneficial properties is quite wide [32].

**Table 1.** Theoretical (Theo) and experimental (Exp) elemental compositions, crosslinker content, and yield percentage for Ch, Ch/HA and Ch/HAMA membranes.


<sup>a</sup> Determined by elemental analysis. <sup>b</sup> Determined by ICP. <sup>c</sup> Gram of G1Phy per gram of membrane × 100.

The content of G1Phy incorporated in the membranes resulted from the ionic crosslinking between amino and phosphate groups was analyzed by ICP-OES (Table 1). The amount of the phytate crosslinker for Ch/HA membrane decreased somewhat compared to that of Ch sample what was attributed to the lower chitosan content in the former membranes (Ch:HA ratio = 75:25). However, the G1Phy crosslinked in the Ch/HAMA membranes decreased nearly to the half. This less crosslinker content in the IPN membranes in comparison to semi-IPNs maybe due to the fact that the covalently crosslinked HA could hindrance the availability of amine groups of chitosan for ionic interactions with the phosphate groups of G1Phy.

The FTIR spectra of the membranes are represented in Figure S2. The FTIR spectra of semi- and IPN samples showed the characteristic bands of Ch and HA polysaccharides. The main bands appeared between 3600 and 3200 cm−<sup>1</sup> (υ O-H and N–H associated); at 2924 and 2854 cm−<sup>1</sup> (υ C–H); at 1720 cm−<sup>1</sup> (υ C = O in carboxylic and ester groups); at 1643/1634 cm−<sup>1</sup> (υ C = O of amide, amide I); at 1579 cm−<sup>1</sup> (δ N–H); at 1420 cm−<sup>1</sup> (υ COO- and δ C–H); at 1373 cm−<sup>1</sup> (<sup>δ</sup> –CH<sup>3</sup> symmetrical); at 1333 cm−<sup>1</sup> (υ C–N, amide III); at 1258/1262 cm−<sup>1</sup> (X P = O); at 1150 cm−<sup>1</sup> (υ C–O–C asymmetric); at 1066, 1029, 995 and 984 cm−<sup>1</sup> (υ C–O alcohols, υ P–O and P–O–C, υ C–O glycosidic linkages and vibration of pyranose structure); at 893 and 721 cm−<sup>1</sup> (υ P–O and P–O–C) [33,34].

Surface morphology is a critical factor for the development of biomaterials that effectively promote cell adhesion and proliferation [35]. Figure 2 shows a detailed examination of surface topography for Ch (A), Ch/HA (B) and Ch/HAMA (C) by SEM and AFM, as well as the calculated roughness parameters (Ra, roughness average, and RMS, root mean square) for the different systems. Qualitative topographic differences among the membranes can be observed in the SEM micrographs. Particularly, Ch system showed the flattest surface followed by the semi-IPN sample, while a much more granular surface was observed in semi-IPN system. This result illustrated a topographic change due to the incorporation of HA that can be a consequence of electrostatic interactions between carboxylic groups of HA and amino groups of Ch, which leads to polyelectrolyte complex formation [32]. The Ch-HA interactions could be hindered in IPN systems by UV curing process, resulting in flatter surfaces as it is observed in their Ch/HAMA micrographs. AFM 3D images show the nano features and the roughness parameters of representative areas of the membranes. R<sup>a</sup> and RMS values correlated very well with the topography observed by SEM. As it was expected, R<sup>a</sup> and RMS values were significantly higher for Ch/HA membranes in comparison to those of Ch and Ch/HAMA that were very similar to each other. Therefore, we can conclude that covalent crosslinking of HAMA with UV-light resulted in a more compact framework in IPN systems compared to semi-IPNs, which leaded to a decrease of roughness at the nanoscale [9].

Surface parameters such as wettability are important properties that must be studied since hydrophilic-hydrophobic balance greatly determines cell adhesion and proliferation properties of the scaffolds [36]. Surface-wetting characterization is currently carried out by WCA measurements the most common method being the sessile drop goniometry [37], which was used in this work. Measured WCA values for Ch, Ch/HA, Ch/HAMA membranes were 48.18 ± 2.71◦ , 40.97 ± 3.26◦ , and 47.73 ± 4.96◦ , respectively. All the systems showed hydrophilic surfaces (WCA < 90◦ ) as it was expected because of the characteristic water absorption nature of these polysaccharides [38]. Different WCA for Ch samples are reported in literature. For instance, Tamer et al. found higher WCA values (89 ± 0.6◦ ) for Ch surfaces [39] than those obtained in our work. However, it has been reported that Ch polarity highly depends on the type and concentration of the used neutralization solution as well as the time of washing steps. Noriega et al. [40] performed a profound study where they reported a wide range of WCA for Ch surfaces, from 45 to 65◦ , that highly depended on neutralization parameters. As expected, hydrophilicity increased as neutralization base concentration and incubation time increased [40]. In our samples, the relatively low WCA values observed for Ch could also be due to the contribution of available phosphate groups coming from the G1Phy crosslinker on the membrane surface, which exhibits high affinity to polar liquids [41]. A decrease of WCA was observed for semi-IPN due to the higher content of G1Phy in this sample, along with the presence of HA

and its polyanionic character [42]. For its part, IPNs showed WCA values similar to those of Ch membranes, which could be due to the reduction of carboxylic groups of HA after methacrylation reaction and further covalent crosslinking. Since membrane surfaces showed different WCA values, we can conclude that wettability properties seem to be an easily tunable parameter in function of composition and applied crosslinking processes in our systems. *Polymers* **2020**, *12*, x FOR PEER REVIEW 9 of 18 reaction and further covalent crosslinking. Since membrane surfaces showed different WCA values, we can conclude that wettability properties seem to be an easily tunable parameter in function of composition and applied crosslinking processes in our systems.

**Figure 2.** SEM micrographs (left) and AFM 3D perspective images with their respective calculated roughness parameters (right) for Ch (**A**), Ch/HA (**B**), and Ch/HAMA (**C**) polymeric membranes. **Figure 2.** SEM micrographs (left) and AFM 3D perspective images with their respective calculated roughness parameters (right) for Ch (**A**), Ch/HA (**B**), and Ch/HAMA (**C**) polymeric membranes.

Rheological measurements were carried out to study the viscoelastic properties of our systems. The evolution of the elastic and viscous moduli of the membranes was studied in their LVR at a constant strain of 0.1%, and it is represented in Figure 3A. All the systems exhibited a plateau in the studied frequency range, which indicated the stability of the crosslinked network. This plateau also showed a solid-like behavior of the membranes, since elastic moduli was independent on the applied frequency [43]. IPN showed higher G' values in comparison to Ch and semi-IPN systems due to the macromolecular reinforcement of the polymeric network after covalent crosslinking. IPNs have previously demonstrated to improve mechanical properties regarding semi-IPNs, because of double crosslinking mechanisms [4,9]. Finally, loss tangent (tan δ = viscous modulus/elastic modulus), which is an index of the viscoelasticity of the systems, was calculated by taking the ratio between G'' and G' at a frequency of 1 Hz. Values of 0.27, 0.17 and 0.24, were obtained for Ch, Ch/HA and Ch/HAMA Rheological measurements were carried out to study the viscoelastic properties of our systems. The evolution of the elastic and viscous moduli of the membranes was studied in their LVR at a constant strain of 0.1%, and it is represented in Figure 3A. All the systems exhibited a plateau in the studied frequency range, which indicated the stability of the crosslinked network. This plateau also showed a solid-like behavior of the membranes, since elastic moduli was independent on the applied frequency [43]. IPN showed higher G' values in comparison to Ch and semi-IPN systems due to the macromolecular reinforcement of the polymeric network after covalent crosslinking. IPNs have previously demonstrated to improve mechanical properties regarding semi-IPNs, because of double crosslinking mechanisms [4,9]. Finally, loss tangent (tan δ = viscous modulus/elastic modulus), which is an index of the viscoelasticity of the systems, was calculated by taking the ratio between G" and G' at a frequency of 1 Hz. Values of 0.27, 0.17 and 0.24, were obtained for Ch, Ch/HA and Ch/HAMA membranes, respectively.

membranes, respectively. The mesh size is defined as the distance between crosslinking points of the membrane, which is related to the mechanical strength of the hydrogel [29,44]. As expected, the IPN membrane exhibited a lower mesh size value due to dual crosslinking, resulting in a more compact structure in comparison to Ch and semi-IPN membranes, which showed similar mesh size values. (Figure 3B) These results The mesh size is defined as the distance between crosslinking points of the membrane, which is related to the mechanical strength of the hydrogel [29,44]. As expected, the IPN membrane exhibited a lower mesh size value due to dual crosslinking, resulting in a more compact structure in comparison to Ch and semi-IPN membranes, which showed similar mesh size values. (Figure 3B) These results corroborate surface and morphology characterization illustrated in SEM and AFM images (Figure 2).

corroborate surface and morphology characterization illustrated in SEM and AFM images (Figure 2).

**Figure 3.** Evolution of elastic (G', filled) and viscous (G'', unfilled) moduli, and loss tangent (tan δ, half-filled) as a function of applied frequency at constant strain of 0.1% of Ch, Ch/HA, and Ch/HAMA polymeric membranes (**A**). The average mesh size values ξ of Ch, Ch/HA and Ch/HAMA polymeric membranes determined at a frequency of 1 Hz (**B**). **Figure 3.** Evolution of elastic (G', filled) and viscous (G", unfilled) moduli, and loss tangent (tan δ, half-filled) as a function of applied frequency at constant strain of 0.1% of Ch, Ch/HA, and Ch/HAMA polymeric membranes (**A**). The average mesh size values ξ of Ch, Ch/HA and Ch/HAMA polymeric membranes determined at a frequency of 1 Hz (**B**).

Collectively, the results described in this subsection showed some differences regarding composition, surface topography, wettability and mechanical properties for the semi- and IPN systems that are expected to also exert relevant differences on their swelling and degradation properties, as well as biological performance. Collectively, the results described in this subsection showed some differences regarding composition, surface topography, wettability and mechanical properties for the semi- and IPN systems that are expected to also exert relevant differences on their swelling and degradation properties, as well as biological performance.

### *3.2. In Vitro Swelling and Degradation Studies 3.2. In Vitro Swelling and Degradation Studies*

Swelling of the developed membranes was studied under physiological conditions and the results are represented in Figure 4A. For all membranes, a fast water uptake during the first hour was observed, reaching a stable value after 3 h, when equilibrium was attained. Ch and Ch/HA membranes showed rather similar swelling profiles giving final swelling values ~100% (Figure 4B). Due to pKa value of Ch at 6.4, its amino groups are not positively charged under physiological conditions and the repulsive forces in the polymeric backbone are not induced, not taking place this increase of the network water uptake [36]. For its part, Ch/HAMA membranes showed the highest equilibrium water absorption (up to 140%) which could be attributed to the formation of a dual crosslinked network able to locate a higher amount of water molecules in their interstices, and to retain a higher HA content in comparison to semi-IPN (Table 1). Nevertheless, all membranes showed moderate swelling that will contribute to maintain their structural stability. If necessary, swelling could be adjusted by varying the content of HAMA in the membrane [45]. Swelling of the developed membranes was studied under physiological conditions and the results are represented in Figure 4A. For all membranes, a fast water uptake during the first hour was observed, reaching a stable value after 3 h, when equilibrium was attained. Ch and Ch/HA membranes showed rather similar swelling profiles giving final swelling values ~100% (Figure 4B). Due to pKa value of Ch at 6.4, its amino groups are not positively charged under physiological conditions and the repulsive forces in the polymeric backbone are not induced, not taking place this increase of the network water uptake [36]. For its part, Ch/HAMA membranes showed the highest equilibrium water absorption (up to 140%) which could be attributed to the formation of a dual crosslinked network able to locate a higher amount of water molecules in their interstices, and to retain a higher HA content in comparison to semi-IPN (Table 1). Nevertheless, all membranes showed moderate swelling that will contribute to maintain their structural stability. If necessary, swelling could be adjusted by varying the content of HAMA in the membrane [45].

In vitro degradation of all hydrogel membranes immersed in a PBS solution at 37 ◦C was below 10% over 14 days (Figure 4B) and it slightly increased after 28 days. However, for longer incubation time (~2 months), degradation of Ch and Ch/HAMA was maintained while Ch/HA membranes suffered further degradation (~16%). The initial membrane weight loss could be attributed to the progressive breaking of the ionic bonds formed between the phosphate and amino groups, what produces release of the G1Phy crosslinker and consequently, dissolution of HA and Ch polymeric chains. Similar results were reported for Ch/HA tissue engineering porous scaffolds where it was suggested than the degradation in PBS is only because of polymeric dissolution [46]. The higher weight loss of Ch/HA versus Ch membranes could be associated to the presence of entangled HA within the Ch

network, favoring its dissolution. Accordingly, Ch/HAMA membranes, displayed the highest stability. This fact could be explained because the covalently crosslinked HA prevents its dissolution and the hydrolytically degradable ester bonds in the HAMA are sterically hindered [ *Polymers*  47]. **2020**, *12*, x FOR PEER REVIEW 11 of 18

**Figure 4.** Effect of hydrogel membrane composition on swelling after incubation in PBS 7.4 at 37 °C for different periods of time (**A**). Weight loss of Ch, Ch/HA and Ch/HAMA membranes at different time points after soaking in PBS 7.4 at 37 °C under static conditions (**B**). Data represented the mean ± **Figure 4.** <sup>E</sup>ffect of hydrogel membrane composition on swelling after incubation in PBS 7.4 at 37 ◦C for different periods of time (**A**). Weight loss of Ch, Ch/HA and Ch/HAMA membranes at different time points after soaking in PBS 7.4 at 37 ◦C under static conditions (**B**). Data represented the mean ± SD.

### SD. *3.3. G1Phy Release*

In vitro degradation of all hydrogel membranes immersed in a PBS solution at 37 °C was below 10% over 14 days (Figure 4B) and it slightly increased after 28 days. However, for longer incubation time (~2 months), degradation of Ch and Ch/HAMA was maintained while Ch/HA membranes suffered further degradation (~16%). The initial membrane weight loss could be attributed to the progressive breaking of the ionic bonds formed between the phosphate and amino groups, what produces release of the G1Phy crosslinker and consequently, dissolution of HA and Ch polymeric chains. Similar results were reported for Ch/HA tissue engineering porous scaffolds where it was suggested than the degradation in PBS is only because of polymeric dissolution [46]. The higher weight loss of Ch/HA versus Ch membranes could be associated to the presence of entangled HA within the Ch network, favoring its dissolution. Accordingly, Ch/HAMA membranes, displayed the highest stability. This fact could be explained because the covalently crosslinked HA prevents its dissolution and the hydrolytically degradable ester bonds in the HAMA are sterically hindered [47]. *3.3. G1Phy Release*  The release profile of G1Phy from the different hydrogel membranes is showed in Figure 5. All membranes showed a fast release (~72%) during the first 24 h which correspond to a G1Phy concentration of 0.1 ± 0.003, 0.09 ± 0.005 and 0.04 ± 0.001 mg/mL for Ch, Ch/HA and Ch/HAMA membranes, respectively. It is worth mentioning that at 24 h the G1Phy concentration of Ch/HAMA membrane is nearly half to that observed for Ch and Ch/HA membranes and this fact is in agreement with the lower initial content of G1Phy incorporated in the Ch/HAMA membranes, data previously described in Table 1. A plateau in the release profile is observed for the three systems after 7 days, The release profile of G1Phy from the different hydrogel membranes is showed in Figure 5. All membranes showed a fast release (~72%) during the first 24 h which correspond to a G1Phy concentration of 0.1 ± 0.003, 0.09 ± 0.005 and 0.04 ± 0.001 mg/mL for Ch, Ch/HA and Ch/HAMA membranes, respectively. It is worth mentioning that at 24 h the G1Phy concentration of Ch/HAMA membrane is nearly half to that observed for Ch and Ch/HA membranes and this fact is in agreement with the lower initial content of G1Phy incorporated in the Ch/HAMA membranes, data previously described in Table 1. A plateau in the release profile is observed for the three systems after 7 days, where Ch and Ch/HA membranes reached an 85% release of the initial G1Phy membrane content, giving a final G1Phy concentration of 0.11 ± 0.004 and 0.10 ± 0.0005 mg/mL, respectively. On the other hand, Ch/HAMA membranes showed a G1Phy release of ~90%, corresponding to a final concentration of 0.05 ± 0.005 mg/mL. The slightly higher G1Phy release in the latter membrane could be due to the higher water uptake of Ch/HAMA membranes what can favor ion diffusion and subsequently ion exchange between G1Phy anions and negative anions present in the Tris buffer. Electrostatic interaction between G1Phy ions (PO<sup>4</sup> <sup>2</sup><sup>−</sup> or HPO<sup>4</sup> <sup>−</sup>) and the amino groups of Ch and breaking of links with incubation time could account for these release profiles. However, no complete release of initial G1Phy content was achieved after 14 days. Physical mixture of Ch with phytic acid has been reported by Barahuie et al. [48] showing a complete release after 60 s. Finally, it is worth mentioning that the release pattern of our systems is in agreement with that of Ch/phytic acid systems reported in the literature [48,49]. *Polymers* **2020**, *12*, x FOR PEER REVIEW 12 of 18

where Ch and Ch/HA membranes reached an 85% release of the initial G1Phy membrane content,

**Figure 5.** Release profiles of G1Phy from the Ch, Ch/HA and Ch/HAMA membranes in 0.1 M Tris buffer (pH 7.4) at 37 °C. Data represented the mean ± SD. **Figure 5.** Release profiles of G1Phy from the Ch, Ch/HA and Ch/HAMA membranes in 0.1 M Tris buffer (pH 7.4) at 37 ◦C. Data represented the mean ± SD.

which also confirmed the biocompatibility of the membranes [51,52].

3.4.2. Cell Viability and Proliferation Assays

To characterize the microstructural architecture of Ch, Ch/HA and Ch/HAMA membranes and observe the morphology of hMSCs cultured on them, an ESEM analysis was carried out on day 21. It is known that the surface roughness is an important factor in promoting cell attachment [50]. ESEM images of Ch, Ch/HA and Ch/HAMA membranes (Figure 6A) revealed a rougher surface for the Ch/HA membrane compared to those of Ch/HAMA and Ch, corroborating the SEM observations for dried samples (Figure 2) but surfaces of all membranes were able to support cell growth (Figure 6B). ESEM images evidenced the cells covering the surface of Ch, Ch/HA and Ch/HAMA membranes, with good adhesion, spreading, and a homogenous distribution throughout the entire surface. Moreover, ESEM images showed an interconnected cell community that attached to the scaffold

In order to evaluate the feasibility of Ch, Ch/HA and Ch/HAMA hydrogel membranes as an adequate support for cell survival, the viability of the seeded hMSCs on the top of the membranes was evaluated. The live/dead assay was employed to visualize the presence of living and dead cells after 1, 7 and 21 days in the hydrogel membranes (Figure 7A). Confocal images showed hMSCs growing on all the membrane surfaces at days 1 and 7. The number of living cells was much higher at day 21 and cells appeared covering the hydrogel membranes with few dead cells. These results indicated that Ch, Ch/HA and Ch/HAMA hydrogel membranes can provide an amenable environment that supports hMSCs growth and confirmed the cell viability with no cytotoxic effects.

*3.4. Biological Evaluation* 
