Preparation and Characterization of Polyamidoamine G2.0-Hematin as a Biocatalyst for Fabricating Catecholic Gelatin Hydrogel

In this study, we report that an enzyme-mimicking biocatalyst polyamidoamine (PAMAM) dendrimer G2.0-hematin (G2.0-He) was fabricated successfully. The chemical structure of G2.0-He was veri ﬁ ed by 1 H NMR and FT-IR spectroscopy. G2.0-He exhibited a size distribution from 11 : 6 ± 1 : 7 nm to 12 : 5 ± 2 : 9 nm and a zeta potential from 32.5mV to 25.6mV along with the enhancement of the hematin conjugation degree. The relative activity of G2.0-He was evaluated based on pyrogallol oxidation reactions at pH = 7 . The results showed that G2.0-He was more stable than horseradish peroxidase (HRP) enzyme in high H 2 O 2 concentrations. The HRP-mimic ability of G2.0-He was also con ﬁ rmed by the catalyzation when preparing catecholic gelatin hydrogels under mild conditions. Moreover, our results also revealed that these hydrogels performed with excellent cytocompatibility in an in vitro study and could be used as a potential sca ﬀ old for adhesion and proliferation of ﬁ broblast cells. The obtained results indicated that G2.0-He is a suitable platform for altering the HRP enzyme in several biomedical applications.


Introduction
Hydrogels are recognized as excellent 3D matrices that can mimic cell and tissue culture environments; thus, hydrogels have become a tremendous motivator in biomedical research fields [1][2][3][4]. Because of the following requirements for tissue regeneration, namely, biocompatibility, biodegradability, and versatile applications, various injectable hydrogel systems have been introduced. Recently, phenolic polymer-based hydrogels have been extensively studied [3][4][5]. An interesting aspect of a hydrogel is its injectability; a hydrogel system could be in situ formed after injecting the polymer solution into a defective site [5]. Currently, the application of a perox-idase enzyme in the preparation of injectable hydrogels is suggested as an attractive technique [5]. The peroxidase enzyme could accelerate the oxidation of phenolic hydroxyl groups to form a chemical crosslinking, subsequently resulting in the formation of a hydrogel network at the desired time that is suitable for implantation surgeries [5,6]. Among them, horseradish peroxidase (HRP) is one of the most commonly used enzymes for fabricating a phenolic gelatin hydrogel [6,7]. The change in the oxidation stage of hem groups (the active site of HRP) in the presence of an oxidant agent (H 2 O 2 ) induces the formation of free radical phenolic substrates leading to the crosslinking of the aromatic ring by C-C and C-O coupling, and ultimately leading to hydrogel formation [8][9][10][11][12]. This is the exquisite action of the HRP enzyme in the preparation of the phenolic gelatin hydrogel; however, it is a fact that the HRP enzyme becomes quickly inactivated during phenol oxidation and polymerization due to reaction conditions (e.g., an active site intermediate compound reacts with excess H 2 O 2 ) [13]. In addition, the HRP enzyme suffers from some intrinsic drawbacks, such as high cost; difficulty in mass production, purification, and storage; low stability; and sensitivity towards atmospheric conditions, which in turn result in limited availability for a wide usage in clinical applications [13,14]. Consequently, the pursuit of alternative platforms that can accomplish those challenges has brought the introduction of various enzyme mimics, also called mimicking enzymes.
It has been found that hematin, a hydroxyferriprotoporphyrin obtained through the decomposition of hemoglobin with one of the axial coordination sites of the Fe (III) center, has the analog structure of the prosthetic iron protoporphyrin IX found in the active site of HRP [15]. Various studies have explored and proved the potential of hematin as a prospective candidate for playing a catalytic role to the HRP enzyme [15][16][17][18][19][20]. However, the usage of hematin as a catalyst for phenolic hydrogelation has been restricted due to its low solubility and aggregation at neutral or acidic pH. Previously, a study had been conducted to dissolve hematin in alkaline buffer at pH = 10:0, subsequently neutralized to 7.4 for further uses; however, its applications were partially limited [17]. Therefore, different functionalization strategies, such as chitosan-g-hem [17], poly(ethylene glycol)-hematin-catalyzed (PEG-hematin) polyaniline (Pani) [18], or even its inclusion into poly(ethylene glycol)-block-poly(4-vinylpyridine) (PEG-b-P4VP) micelles [19], have been attempted.
Motivated by these studies, we aim to synthesize a soluble hematin-modified generation 2.0 polyamidoamine dendrimer (G2.0-He). In the past two decades, dendrimers have emerged as a novel class of polymeric materials, thanks to their nanosized platforms, high solubility, and monodispersed properties [20]. These characteristics give them a branched architecture that enables them to flexibly modify themselves in numerous ways [21]. It is well known that cationic PAMAM dendrimers exhibit several advantages in controlling drug and protein delivery properties, such as their high drug-loading capacity, highly electrostatic interaction with anionic bioactive molecules, enhanced hydrophobic and hydrophilic drug solubility and bioavailability, and easy functionalization of their external groups [22]. In addition, the high density of peripheral NH 2 groups in cationic PAMAM dendrimers could enhance the possibility of chemical covalent formation between dendrimers and hematin [23]. It can be expected that the combination of hematin and PAMAM G2.0 could overcome the intrinsic drawbacks of both because of their partner's outstanding properties. Therefore, we evaluated the activity of G2.0-He as an alternative catalyst to substitute the HRP enzyme in the in situ formation of a catecholic gelatin hydrogel. For the cell study, human fibroblast cells were purchased from ATCC (USA). The culture media Dulbecco's Modified Eagle's Medium (DMEM) high glucose, fetal bovine serum (FBS), trypsin-EDTA (0.25%), and penicillin-streptomycin (10,000 U/mL) were purchased from Sigma-Aldrich (Singapore). Phosphate-buffered saline (PBS, 1x) was purchased from Gibco. Other culture wells were ordered from Corning. Cell culture flasks and plates were ordered from Corning. The cytotoxicity assay was performed with sulforhodamine B (SRB) assay (Ab235935) and dual-staining Live/Dead assay via acridine orange (AO, Alfa Aesar, USA) and propidium iodide (PI, Sigma-Aldrich).

Experimental
2.2. Synthesis of PAMAM Dendrimer G2.0-Hematin. PAMAM dendrimer G2.0-hematin was prepared using the carbodiimide coupling reagent as illustrated in Figure 1. Firstly, hematin (100 mg, 0.16 mmol) was dissolved in 50 mL DMF, and G2.0 was dissolved in 50 mL methanol according to the molar ratio: G2:0 − He = 1 − 1 ; 1 − 2. The excess EDC/NHS (molar ratio: hematin/EDC/NHS = 1/2/2) was employed to activate the carboxylic group of hematin, which subsequently reacted with the amine group of PAMAM G2.0. Then, hematin/EDC/NHS was added dropwise into the G2.0 solution under stirring. The reaction was kept at room temperature under nitrogen for 24 h in the dark. After that, the product was dialyzed against methanol using a membrane with a molecular weight cut-off (MWCO) of 1,000 Daltons until the unreacted hematin was completely removed. Vacuum distillation was used to eliminate methanol at 45°C. The products were redissolved in DI water and finally lyophilized to get G2.0-He without impurities [17]. G2.0-He was characterized by 1 H NMR and Fourier transform infrared (FT-IR) spectroscopy. Ultraviolet-visible (UV-Vis) spectroscopy was used to quantify hematin conjugated on the PAMAM G2.0 surface. Via the absorption spectrum of hematin, the λ max at 404 nm was defined and used to calculate the conjugation degree [24]: Conjugation = weight of He in G2:0-He weight of the feeding He × 100%: Furthermore, dynamic light scattering (DLS, SZ-100 nanopartica series instruments, Horiba) was measured to determine the hydrodynamic size and the zeta potential of G2.0-He. International Journal of Polymer Science [25]. Briefly, gelatin (2 g) was dissolved in 100 mL of DI water under stirring at 50°C, followed by the precise addition of dopamine solution. After 30 minutes, EDC (0.5 g, 0.00261 mol) and NHS (0.5 g, 0.00434 mol) were added to the mixture. The pH value was maintained at 5.5 during the reaction for 24 h at room temperature. The reaction solution was later dialyzed in DI water for 3 days before lyophilization. The product was verified using proton nuclear magnetic resonance spectroscopy ( 1 H NMR).

Catalytic Activity on Pyrogallol.
The catalytic activity of G2.0-He at pH = 7:0 was evaluated via pyrogallol oxidation reaction using UV-Vis spectroscopy. A mixture containing 700 mg/L pyrogallol and 50 mg/L G2.0-He, H 2 O, and H 2 O 2 for a total volume of 3.5 mL was added into a quartz cell [17]. The concentration of H 2 O 2 that was investigated varied from 0.1 to 100 mM [15]. HRP (0.01 U/mL) was used as a control. The increasing absorbance was recorded at 420 nm after 180 s since the addition of H 2 O 2 . The tracked absorbance was used to calculate the relative activity of G2.0-He and HRP at each H 2 O 2 concentration, respectively [17].
The catalytic activities were exhibited as the median values of three measurements.  [26]. Experiments were performed three times at each condition, and the data was expressed by mean ± SD. 3 International Journal of Polymer Science 2.6. Cytotoxicity Test. The toxic profiles of the G2.0-Hemediated Gel-Dop hydrogels were determined using an indirect method, which was previously described by Zhang et al. [27]. Human fibroblast cells were seeded into 96-well plates with a density of 2 × 10 4 cell/well. The cells were incubated at 37°C, 90% humidity, and 5% CO 2 condition. After an incubation time of 24 h allowing cells to be adherent, a newly completed DMEM (10% FBS, 1% penicillin-streptomycin) containing a series concentration of leachates of hydrogels (50%, 20%, and 10% v/v corresponding to 0.5 mg/mL, 0.2 mg/mL, and 0.1 mg/mL, respectively) was added, and cells were incubated for another 4 h and 24 h. Along with the extracted Gel-Dop hydrogel, the cytotoxicity of G2.0-He (with the same concentration used in hydrogelation) as well as the buffer (PBS 1x) used to extract hydrogel were also involved in this experiment. The untreated cells incubating with the completed DMEM only was used as the negative control, while the cell culture with doxorubicin (0.1 mM) was considered as the positive control. At the designed time (4 h and 24 h), the SRB kit (Ab235935) was applied in each well following the guidance of the manufacturers. The percentage of viable cells was calculated regarding untreated cells. All the experiments were repeated with three independent replications.

Gelation
We also performed another cytotoxicity assay for validation using AO/PI dual staining. In this study, the fibroblast cells (2 × 10 5 cells) were cultured in a 35 mm culture dish (Thermo Fisher Scientific™ Nunc). After 24 h of incubation time similar with the previous condition, the culture medium was removed and the new culture medium containing the leachates of hydrogels (50% v/v), G2.0-He (100 ppm), and PBS 1x was then added into each culture dish. These culture dishes were incubated at 37°C, 90% humidity, and 5% CO 2 for a further 24 h. Afterward, 2 μL working solution (AO/PI) was added to each dish and incubated for 15 minutes in a CO 2 incubator. PBS 1x was used to remove the free dye in the media, and the newly completed media was supplied prior to microscope observation. Confocal (Dragonfly, Oxford Instrument, England) was used to observe the morphology and evaluate the live and dead cells through dual channels (525 nm and 617 nm).  [28]. The spectrum shows the signal at δ~1:877 ppm belonging to the methyl protons in the hematin structure (c) [29]. In addition, ring stretching vibrations mixed strongly with C-H in-plane bending at 1031 cm -1 and 1439 cm -1 . The C-N stretching mode of amide III is at 1255 cm -1 , and for CH 3 , the deformation of C-H and N-H is at 1384 cm -1 . Amide II bands arise from C-N stretching and CHN bending vibrations at 1544 cm -1 [30,31]. For amide band I, the "footprint" for the generation of G2.0-He, the C=O stretching vibration of the amide C=O is predominantly shown at 1651, 1661, and 1727 cm -1 . C-H stretching vibrations of methyl (CH 3 ) and methylene (CH 2 ) groups are exhibited at the range of wavenumbers from 2800 to 3100 cm -1 . Also, the stretching O-H asymmetric structure in the remaining carboxylate groups of hematin appears at 3454 cm -1 [31]. Besides, the FT-IR spectrum of G2.0-He (1-2) displays the correlation to G2.0-He (1-1) as well. Generally, the 1 H NMR and FT-IR data are consistent in consolidating the successful synthesis of PAMAM dendrimer G2.0-hematin.  [32]. Besides, the spectrum exposes the resonance peaks of aromatic protons of phenylalanine and other typical protons of amino acids in gelatin at 7.340-7.406 ppm, and the signals at 0.9-4.6 ppm are assigned to the alkyl protons of gelatin [33]. Regarding the results, the successful conjugation of dopamine onto the gelatin backbone was verified for use in preparing catecholic gelatin hydrogels. Figure 3(a) displays the UV-Vis spectra of 3 ppm hematin, G2.0-He (1-1), and G2.0-He (1-2) in DMSO with the absorbance peak of 404 nm; thus, the conjugation degree of hematin was measured and calculated at this absorbance peak. The conjugation of G2.0-He (1-1) reaches 86%, which is about 22% higher than that of G2.0-He (1-2) (Figure 3(b)); hence, G2.0-He (1-1) was preferred to investigate the effect of catalytic ability when comparing to the HRP enzyme in forming catecholic gelatin hydrogels. The difference in the conjugation of G2.0-He (1-1) and G2.0-He (1-2) can be explained by the impaction of the steric effect of loaded hematin and the effect of hematin on the entrapment efficiency.

Results and Discussion
The combination of PAMAM G2.0-hematin brings benefits in two directions: improvement in hematin solubility and the toxicity minimization of PAMAM G2.0. Given the presence of amine groups on the PAMAM G2.0 surface, hematin-covered PAMAM G2.0 yields higher solubility compared to pure hematin. In addition to this, hematin-covered PAMAM G2.0 also solves the membrane disruption problem caused by the positive charge of these amine groups. Here, we investigated G2.0-He with four molar ratios (1-1, 1-2, 1-4, and 1-6). According to our results, G2.0-He (1-1) and (1-2) were well soluble, while the others were precipitated. Thus, G2.0-He (1-1) and (1-2) were selected for further observation and investigation. As shown in Figure 4, generally, the more    Figure 5(d). The relative activity of the HRP enzyme increased from 82% to 100% along with the increasing H 2 O 2 concentration in the range from 2 to 10 mM, while it significantly decreased when the H 2 O 2 concentration was higher than 10 mM. These results express that the HRP enzyme suffers inactivation when the concentration of H 2 O 2 is further increased up to 30 mM, which is consistent with previous studies [15,17]. G2.0-He was found to possess outstanding H 2 O 2 stability over a wide range of peroxide concentrations from 2 to 100 mM. The relative catalytic increase from 19% in 2 mM H 2 O 2 to 85% in 50 mM H 2 O 2 then reaches approximately 100% in 100 mM H 2 O 2 . The reason for this could be derived from the intrinsic peroxidase property of hematin molecules. Consequently, G2.0-He is available in a wide range of applications given its high stability.

G2.0-He-Mediated Gelation of Catecholic Gelatin
Hydrogel. Gel-Dop, a polymer with phenol functionalities, can be crosslinked by HRP/H 2 O 2 or G2.0-He/H 2 O 2 systems. As a result of assuming these systems, phenol radicals will be produced, then they will be coupled with each other, either at the C-C or C-O positions, and crosslinks will be formed between the polymer chains ( Figure 6(a)). Figure 6(a) shows . According to previous studies [15,16], a higher amount of H 2 O 2 (above 30 mM) was used to boot the catalytic activity of hematin. It has been noted that using a higher amount of H 2 O 2 to control the gelation time of phenol motif polymers such as Gel-Dop brings with it a remarkable risk to the health of the cells, thereby, limiting the widespread use of hematin in biological hydrogels [4]. In this study, aside from controlling gelation time by adjusting H 2 O 2 concentration, the gelation time could also be controlled by varying G2.0-He concentrations. The maximum concentration of hematin in the same study was around 0.08% w/w, whereas, in this study, the concentration of G2.0-He could be up to 1.2 wt%. Hence, our findings support the application of hematin as HRP mimics in biotechnology, specializing in biohydrogelation.
To prove the potential application of Gel-Dop hydrogel formation with the help of G2.0-He as a scaffold for cell stud-ies, the cytotoxicity of the obtained Gel-Dop was investigated using human fibroblast cells by SRB assay and Live/Dead staining assay. Human fibroblast cells were incubated with different concentrations of Gel-Dop hydrogel extract (10%, 20%, and 50% v/v), and their cytotoxicity was then compared to G2.0-He, positive control (doxorubicin, 0.1 mM), and PBS 1x (a solution used to extract hydrogels for the first time). After the first 4 h of coculture with the positive control, the viability of cells were reduced to half as compared to untreated cells, about 53:39 ± 6:01% (Figure 7(a)). As the culture time was prolonged, the viability of fibroblast cell culturing with doxorubicin plunged to 12:14 ± 2:66%. On the contrary, there was no significant difference observed in % viability of cells with the Gel-Dop hydrogel extract (10%, 20%, and 50% v/v), G2.0-He, and PBX 1x (Figure 7(a)); levels of the viability of fibroblast cells were approximately 90% for untreated cells after 4 h or 24 h culture (p > 0:01). Likewise, the data from the Live/Dead staining assay with AO/PI was consistent with the SRB assay (Figure 7(b)). Fibroblast cells were alive (green cells, >90%), and an unremarkable red fluorescent signal (the death marker) was observed in 50% extracted hydrogel, G2.0-He, and PBS 1x. Altogether, the Gel-Dop hydrogelation using G2.0-He has almost no adverse impact on cell viability suggesting that it can be potentially used as a scaffold in tissue regenerated application.

Conclusion
In conclusion, this study was successful in preparing PAMAM dendrimer G2.0-hematin, which can overcome the poor aqueous insolubility of hematin. Besides, our synthesized G2.0-He exhibits the same catalytic activity as the HRP enzyme but is more stable in a high H 2 O 2 concentration condition. Therefore, G2.0-He could be employed in fabricating hydrogels based on its catalytic ability, similar to the HRP enzyme, and these hydrogels are highly cytocompatible and able to serve as an excellent scaffold. Overall, according to our findings, G2.0-He could be considered an efficient horseradish peroxidase mimetic catalyst for biochemical analysis and fabrication of biomedical materials.

Data Availability
The data used to support the findings of this study are included within the article.