Electrospinning of gelatin fibers using solutions with low acetic acid concentration: effect of solvent composition on both diameter of electrospun fibers and cytotoxicity

Gelatin fibers were prepared by electrospinning of gelatin/acetic acid/water ternary mixtures with the aim of studying the feasibility of fabricating gelatin nanofiber mats at room temperature using an alternative benign solvent by significantly reducing the acetic acid concentration. The results showed that gelatin nanofibers can be optimally electrospun with low acetic acid concentration (25% v/v) combined with gelatin concentrations higher than 300 mg/ml. Both gelatin solutions and electrospun gelatin mats (prepared with different acetic acid aqueous solutions) were analyzed by FTIR and DSC techniques in order to determine the chemical and structure changes of the polymer. The electrospun gelatin mats fabricated from solutions with low acetic acid content showed some advantages as the maintenance of the decomposition temperature of the pure gelatin (~230ºC) and the reduction of the acid content on electrospun mats, which allowed to reach a cell viability upper than 90% (analyzed by cell viability test using human dermal fibroblast and embryonic kidney cells). This study has also analyzed the influence of gelatin and acetic acid concentration both on the solution viscosity and the electrospun fiber diameter, obtaining a clear relationship between these parameters.


INTRODUCTION
In recent years, numerous reports in the field of tissue engineering have put the emphasis on the design and manufacturing of biocompatible and biodegradable supports with capacity of mimicking the structural and functional properties of extracellular matrices (ECM). [1][2][3] For an optimal biocompatibility, scaffolds used for tissue engineering should possess special characteristics of degradation, porosity, microstructure, size, etc. 1,3 These characteristics highly depend on the fabrication method and, consequently, different techniques for the production of such scaffolds have been investigated and optimized (e.g. self-assembly, phase separation). 1,2,4 More recently, the electrospinning technique appeared as a versatile technique for manufacturing nanofibers and nanofibrous arrays with dimensions and scale similar to those of the native ECM [5][6][7][8][9][10][11] , suitable for medical applications. 12 The electrospinning technique 11 allows the production of small diameter fibers (ranging from nanometers to micrometers) by applying a high voltage electrostatic field between a metal capillary syringe containing a polymer solution and a grounded collector where the fibers are deposited. During this process, as a result of solvent evaporation, electrospun fibers are deposited onto the collector in the form of nonwoven fibrous webs of high porosity. The properties of the obtained fibers depend on the operating conditions, e.g. flow rate, voltage, time, temperature and distance from the collector, as well as on the properties of the polymer solution, e.g. concentration, density, viscosity, conductivity, surface tension. 7,13 Electrospinning can be applied to both synthetic [14][15][16] and natural polymers, including polysaccharides 17 and proteins, being collagen 18,19 , silk fibroin 20 and gelatin 21 the most studied ones. Gelatin is known to have biocompatibility and biodegradability similar to collagen. 22,23 In fact, it is easily obtained by partial hydrolysis of collagen from animal tissues such as skin, muscle, and bone. Depending on the hydrolysis method, two different types of gelatin are produced: type A gelatin (acid process), type B (alkaline process). Both gelatins differ mainly in their amino acid composition, polypeptides pattern, bloom strength, turbidity and foaming properties. 24,25 From a technical point of view, the most important parameter influencing the electrospinning manufacturing process of gelatin nanofibers is the solvent selection 26 because, although gelatin is soluble in warm water, the electrospinning cannot be done at room temperature due to the gelation process that occurs between gelatin and water, which increase the solution viscosity avoiding the flow of aqueous gelatin solutions into the syringe. 27 With the aim of avoiding the gelation process and allow the electrospinnability of gelatin solutions, some complex solvent such as 1,1,1,3,3,3-hexafluoro-2-propanol (HIPF) or 2,2,2-trifluoroethanol (TFE) have been proposed for the fabrication of scaffolds made of natural polymer such as collagen. 21,[28][29][30][31][32] However, due to their highly corrosive nature, these solvents may affect the original protein structure 11,33,34 besides providing a potential cytotoxicity to the obtained scaffolds, since the presence of small amounts of residual solvent embedded on the electrospun fibers is almost unavoidable. 35,36 This fact, combined with the high cost of these solvents, promoted the search of alternative systems to electrospin gelatin such as: i) the use of gelatin aqueous solution at high temperatures 37,38 , ii) blending gelatin with another biopolymer (e.g. sodium alginate 38 , poly(ethylene oxide), poly(e-caprolactone)) 11 , PLA 39 , iii) using solvent mixtures (acetic and ethyl acetate 40 , and iiii) and using of carboxylic acid (formic acid 22,41 or acetic acid 26,[39][40][41][42][43] ).
Among these alternatives, the use of aqueous solutions of carboxylic acids has been postulated recently as a preferable option to dissolve and electrospin gelatin at room temperature. The use of these acids implies a clear advantage over HIPF and TFE solvents due to their lower cytotoxicity and their simplicity of processing compared to other alternatives. However, the concentrations of acid proposed up to now to electrospin gelatin are quite high (minimum of 60% v/v), inducing the partial decomposition of gelatin and adversely affecting the structural integrity of the nanofibers. 22 This study analyzes the feasibility to electrospin nanofiber mats of gelatin at room temperature using an acetic acid based solvent characterized by a low concentration of acid. The effect of the acidity of the solvent on both the gelatin structure and the cytotoxicity of the final mat were tested. Finally, the relationship between reagents concentration, solution viscosity and fiber diameter was studied given that the diameter of the fibers is a crucial parameter for instance, to mimic the size of the fibers composing the extracellular matrix of connective tissue.

Materials
Gelatin powder from bovine skin (type B with bloom ~225 g) was purchased from Sigma Aldrich (Spain) and used without further treatment or purification. Glacial acetic acid (99.99%, Panreac, Spain) and bi-distilled water were used as solvents.

Electrospinning process
Electrospinning was performed in a home-engineered device. 44,45 Each gelatin solution was loaded into a 2.5 ml syringe with a stainless steel syringe needle (0.6 mm inner diameter) connected as an anode to a high voltage power supply. About 6-10 cm below the needle, a flat copper collector (connected as a cathode to the power supply) was placed to receive the electrospun fibers. The flow rate was controlled by a pump, and set between 1-1.5 ml/h, depending on the solution requirements. The applied voltage was in the range of 15-18 kV and all solutions were electrospun at room temperature (23 ºC).
Electrospun mats were not chemically cross-linked for mechanical stabilization to avoid interferences during their chemical and structural characterization.

Viscosity measurements
The viscosity of the different solutions was determined using a viscometer (Brookfield DV-II +, USA). After mixing for 1 h, samples were stored for different times (0 h, 1 h, 3 h, 24 h) before the measure of viscosity, in order to follow the gelation process.
Each solution viscosity was measured three times and results shown a standard deviation below 2%.

Electrospun fibers characterization
The diameter and distribution of the electrospun gelatin fibers were directly examined whereas HEK 293T was purchased from the European Collection of Cell Culture.

Alamar Blue assay
Cells were seeded at a density of 4.5 x 10 4 cells/well on 96-well tissue culture-treated polystyrene plate (Nunc, Thermo Scientific, USA) the day before experiments. Then, they were exposed by indirect contact to the electrospun gelatin fibers, previously dissolved on medium (20 mg/mL in DMEM), at a final volume of 100 µL and incubated at 37 ºC in a humidified atmosphere with 5 % CO 2 . Cells were examined for signs of toxicity, using Alamar Blue assay.
Resazurin, the active ingredient of AlamarBlue ® reagent (Invitrogen, Life Technologies Corporation, Spain), is a non-toxic, cell-permeable compound that is blue in color and it can be reduced to resorufin by viable cells, developing a red color compound. After 24 h contact with cells, the solution of dissolved fiber mats was removed, the cells washed twice with PBS and stained with AlamarBlue ® reagent.

Cells morphology
Morphological changes in cells were also followed by phase contrast microscopy using an Eclipse Ti-S microscope (Nikon Instruments Inc., Netherlands), after 24 h of contact with gelatin mats.

Viscosity of solutions
Regarding the electrospinning process, besides some technical parameters such as voltage, distance and flow rate, there are several other important parameters that influence the electrospinnability of solutions such as surface tension, conductivity, viscosity and molecular weight. 11 For instance, surface tension determines the upper and lower boundaries of electrospinning window if all other variables are held constant. 46 Previous studies concluded that the increase of an acid concentration provokes a surface tension decrease 40 that benefits the electrospinnability, normally impeded by high surface tensions.
In this case, to study the effect of viscosity on the electrospinnability, firstly, the viscosity changes over the storage time were analyzed for several gelatin solutions prepared with different solvent mixtures (25 -100% v/v acetic acid). As shown in Figure 1, the viscosities of those solutions prepared with aqueous acetic acid at 50% and 75% and pure acetic acid (100%) vary depending on gelatin concentration (200 -400 mg/ml) but they were very stable with time since no significant variation could be observed up to 24 h after mixing (the maximum increment of 15% was attained for the highest gelatin concentration solution, 400mg/ml, in 100% acetic acid), contrarily that occurs for pure formic acid system, where the viscosity of gelatin hardly decreases after 5 hours of storage time. 22 In contrast, for the gelatin solutions containing 25% acetic acid, the viscosity clearly increased with time being the most important increment observed for the most concentrated solution in terms of gelatin content (about a 300% of increment). These changes of viscosity make sense taking into account the gelation phenomenon that gelatin undergoes in the presence of a high amount of water 47 , which is also proportional to the gelatin concentration in solution. 48 From a practical point of view, these results suggest that performing the electrospinning immediately after dissolving gelatin would be preferable in order to avoid the gelation process. Even so, it is important noting that sometimes a slight gelling process is unavoidable even at this moment, probably because gelation already starts during the long dissolution process due to the combination of high water content and high gelatin concentration. This is the case for those samples made 25% of HAc, which sometimes showed slightly higher viscosities (for 350 and 400 mg/ml) than the obtained for the solutions at the same gelatin concentration but with a higher concentration of HAc (e.g. 50%). The corresponding viscosities are summarized in Table 1

Electrospinnability of gelatin solutions
The electrospinnability of gelatin solutions was examined by analyzing the What is more relevant, electrospun fibers can also be obtained for low acetic acid concentrations (25% v/v) combined with high gelatin concentrations (>300 mg/ml). In this latter case, the concentration of gelatin is high enough to induce the necessary viscosity and polymer chain entanglement for adequate electrospinning. At the same time, the acetic acid content is high enough to provide electrical conductivity and, most important, to dissolve gelatin avoiding gelation (note that gelation occurs for gelatin in pure water and it absolutely impedes electrospinning). On the contrary, other combinations were not suitable for electrospinning since they produced either very thick fibers (high acetic acid concentration and high gelatin concentration: Figures 2i, 2m, 2q and 2r) or did not produce fibers at all (low acetic acid concentration and low gelatin concentration: Figures 2c, 2d and 2h). In the first case the viscosity of the solution is very high due to the high amount of gelatin and the acetic acid is only able to partially solvate it, just allowing the electrospinning of very thick fibers or microfibers 43 . In the second case the solutions did not reach the necessary viscosity and polymer chain entanglement to be electrospun. The characterization of some solutions (Figures 2b and 2g), which were partially able to produce nanofibers, also revealed the existence of beads, either as discrete beads or as beaded fibers due to fibers fusion at touching points. Similar behavior has been previously reported for gelatin solutions with a concentration between 200 mg/ml and 300 mg/ml using pure acetic acid as a solvent. 26

Effect of gelatin concentration
It is obvious from the results that the concentration of gelatin directly affects the viscosity of the mixture ( Table 1), what is in agreement with the literature. 22

Effect of acetic acid concentration
It is well known that the presence of acetic acid influences the surface tension of solutions in a way that the surface tension could be reduced by increasing acetic acid concentration. 22 In this sense, one may expect that viscosity is also affected in a similar way but it is necessary to establish the exact relationship between both parameters, which concern electrospinnability. Taking into account the experimental findings, at low concentrations of gelatin the acetic acid content did not affect significantly the viscosity. Accordingly, in these cases the electrospinning process is dominated by the surface tension. It is also important to note that the acetic acid content influences, in turn, the water content and thus, the gelling process. Performing the electrospinning immediately after dissolving the gelatin would avoid the mostly of the spontaneous increase of viscosity, especially for solutions with high content of water (25 % acetic acid).
Similarly to the gelatin concentration, the effect of the acetic concentration was correlated to the fiber diameter (Figure 4), corroborating the same behavior which has been observed for gelatin in other solvents. 39,40 In

FTIR
Samples prepared with a fixed gelatin concentration were analyzed by FTIR to  FTIR might not be accurate enough to confirm the latter conclusion, as it was suggested by Chang et al. 22 and, therefore, we carried out a DSC analysis to check whether the chemical structure of gelatin was affected by acetic acid in solution.

DSC
DSC thermograms of electrospun fibers prepared with solutions containing 25,50,75 and 100 % v/v acetic acid and 300 mg/ml gelatin are plotted in Figure 7, together with the data corresponding to pure powder gelatin. The peak found about 230 ºC for pure gelatin (Figure 7a) agree with the reported value for gelatin decomposition. 22,40,41 This peak was also found (although slightly shifted) for the nanofiber mats prepared with the lowest acetic acid concentration (25%, Figure 7b)).
this peak was not detected, and a wider and shorter peak appeared offset to 200 °C (more deflected at higher acetic acid content). These changes suggest an increase in the amorphous part of the gelatin structure, i.e a decrease in its crystallinity. On the one hand it could be simply explained by the nanoscopic size of fibers, but according with the literature 22 , the changes are attributed to alterations of the random coil conformations of the protein. Thus, despite FTIR spectra did not show many differences for the electrospun fiber mats prepared with different acetic acid concentrations, indicating that the chemical structure of gelatin is not affected, the tertiary structure of the protein is certainly altered causing significant differences in the DSC thermograms. Accordingly, it seems necessary to reduce the acid concentration as much as possible in order to produce nanofibers more analogous to the pristine gelatin.

Cytotoxicity evaluation
The culture medium used to dissolve the mats of electrospun gelatin fibers contains Phenol Red, a pH indicator frequently used in cell biology that allows for detecting any chemical or microbiological contamination in the medium, which could affect the cells, basing on the color changes. This indicator spans the pH range from 6.8 (yellow) to 8.4 (purple) 58 and is useful to detect any possible trace amounts of acetic acid in fiber mats (Figure 8).

Cell morphology
The morphology of BJ-5ta fibroblast cells after indirect contact with the different electrospun gelatin fibers was examined after 24 h by phase contrast microscopy

CONCLUSIONS
The feasibility of electrospinning gelatin nanofibers from solutions with different concentrations of acetic acid and gelatin at room temperature was tested.
The results showed the viability to obtain electrospun gelatin nanofibers at low acetic acid concentration (25%) combined with gelatin concentration of 300 mg/ml or higher.
Both acetic acid content and gelatin concentration exhibited a clear influence on the viscosity solution, which trend was directly correlated with electrospun fiber diameter.
Moreover, the study of viscosity solution in front of time determined that the solutions with low acetic acid and high gelatin concentration were those showed the higher rheology instability, due to the gelation process, suggesting the importance to develop the electrospinning just after 1h of stirring the solution.
Although the FTIR spectra did not show many differences on the electrospun gelatin mats in function of acetic content, the DSC analysis allows to determine the benefit to work at low acetic acid concentration, being the electrospun mat from 25% of acetic acid the only sample that keeps showing the characteristic degradation peak of pure gelatin at 230ºC, related with the crystallinity conformation of polymer.
Finally, the indirect cytotoxicity assay demonstrated the direct relationship between the acetic concentration of the solution and the acid traces found in the final mats revealed by the pH indicator changes. Also, the greatest cell viability (upper than 90%) was achieved for mats from solutions at 25% acetic acid concentration.