Preparation of chitosan matrices with ferulic acid: physicochemical characterization and relationship on the growth of Aspergillus parasiticus

ABSTRACT Antioxidant, anticarcinogenic, and antimicrobial properties have been reported for ferulic acid (FA), therefore, its application interests both food and agriculture research. FA was immobilized in different chitosan (CS) matrices, physicochemicaly characterized and the effect on Aspergillus parasiticus ecological parameters evaluated. Nanoparticles (Nparticles), microparticles (Mparticles) and microcapsules (Mcapsules) of 35–40 nm, 30–40 μm, and 20 μm, respectively were obtained; FA incorporation in matrices affected their morphology, physicochemical properties, and their fungistatic effect. The effect of the particles was dependent on the matrix exposed. Nparticles and Mparticles showed high FA immobilization efficiency as well as a good fungistatic effect against A. parasiticus: Radial growth at 168 h was 28.46 ± 1.01 and 28.84 ± 1.36 and the inhibition of spore germination at 30 h was 57.44 ± 0.22 and 55.74 ± 2.19, for Nparticles and Mparticles, respectively compared with control cultures. Abnormalities in mycelium, hyphae, and spores morphology were observed, as well as low sporulation due particle interaction with the fungus.

The study of CS has been increased in the last few years due to its ability to form Mparticles and nanoparticles (Nparticles), which has been used mainly for biomedical and pharmaceutical applications. The pharmaceutical applicability of CS is originated by its capacity to penetrate biological barriers, to protect macromolecules such as peptides, proteins, oligonucleotides, and genes from degradation in biological media and to deliver drugs or macromolecules to a target site with subsequent controlled release (López-León et al., 2005) such as antibacterial compounds, which are able to control/minimize bacterial infections (Sanpo, Ang, Cheang, & Khor, 2009). However, the studies on the antifungal activity of CS matrices in areas such as food and agriculture are still limited.
Methods for fungi control are mainly based on chemical strategies, mainly through the use of pesticides. Their excessive as many undesirable consequences such as environment pollution, the increase in the frequencies of resistant or tolerant pathogen populations, and the presence of chemical residues in food commodities which could represent a health risk (Reverberi et al., 2005). For these reasons, natural antimicrobials are being evaluated as viable alternatives; however, most of these natural compounds have limitations for their application in agricultural and food areas, so it is necessary to seek other approaches as well as their effective use.
Ferulic acid (FA) (4-hydroxy-3-methoxycinnamic acid) has low toxicity and posseses many physiological functions, including antioxidant, anti-inflammatory, anticancer (Ou & Kwok, 2004), and antimicrobial activity (Rabea et al., 2003). Recently, it has been widely used in food, pharmaceutical, and cosmetic industries; nonetheless, its use is limited by its tendency to be rapidly oxidized in the environment, therefore, viable alternatives must be sought. It has been reported that different structures from CS-based films such as Mparticles and Nparticles, are excellent vehicles for the incorporation, protection and controlled release of a variety of compounds such as antioxidants, enzymes, vitamins, minerals, and other nutrients (Nair, Reddy, Kumar, & Kumar, 2009), nevertheless, their properties (immobilization efficiency, controlled release, and biological activity) will depend on the matrix characteristics. The antifungal effect of CS matrices in toxigenic fungus species such as Aspergillus parasiticus has not been well studied; furthermore, the incorporation of FA in these matrices has been barely reported. Moreover, A. parasiticus is a highly toxigenic fungal species for both human and livestock, commonly found commodities like cereals. A. parasiticus produces different aflatoxins B1, B2, G1, and G2. Aflatoxin B1 (AFB1), among the four major types of aflatoxins, is the most toxic and the most potent carcinogen in humans and animals. AFB1 exhibits a high socioeconomic importance not only due to food spoilage and mycotoxin contamination, but also due to its until now ineffective pre-and post-harvest control. Only for the United States corn industry, Mitchell, Bowers, Hurburgh, and Wu (2016) estimate that aflatoxin contamination could cause losses ranging from US$52.1 million to US$1.68 billion annually. For these reasons, the aim of this study was to obtain and characterized physicochemically CS-tripolyphosphate (TPP) loaded FA matrices (microcapsules (Mcapsules), microparticles (Mparticles) and Nparticles), to determine the most efficient to immobilize FA and finally, to evaluate the most effective form to control the growth of A. parasiticus.

Materials
Low viscosity CS (Sigma-Aldrich) from shrimp head, with a degree of deacetylation <85%, molecular weight of 130 KDa and <1% protein and residual ash, was used. Sodium TPP and FA were also purchased from Sigma-Aldrich.

Microorganism and growth conditions
A monosporic culture of A. parasiticus (ATCC 16992) was grown and kept in potato dextrose agar (PDA) until its use. Inoculum was made by spreading spores on PDA and resuspended in Tween 80. The concentration of spores was determined by counting in a Neubauer chamber and adjusted to 1 × 10 5 spores/mL.

Preparation of CS-TPP matrices with FA incorporated
Three types of CS-TPP matrices with FA were elaborated by the ionotropic gelation method. To elaborate Mparticles, a 0.6% CS solution in 0.05 M acetic acid was prepared. FA (0.8 g) was dissolved in 2% TPP, (FA-TPP). CS solution was sprayed into the FA-TPP (3:4, v/v) at 50 ml/min and stirred for 15 min at 500 rpm. To elaborate Mcapsules, a 0.6% CS solution in 0.05 M acetic acid was prepared. Then, an emulsion was made by mixing CS, soy lecithin and Tween 80 (10.0:5.0:5.0, v/v, respectively). FA (0.8 g) was dissolved in 2% TPP, (FA-TPP). The emulsion was sprayed into the FA-TPP (3:4, v/v) at 50 ml/min and stirred for 15 min at 500 rpm. Finally, to elaborate Nparticles, a 0.2% CS solution in 0.05 M acetic acid was prepared. FA (0.8 g) was dissolved in 2% TPP, (FA-TPP). CS solution was sprayed into the FA-TPP (3:4, v/v) at 50 ml/min and stirred at 500 rpm for 15 min.

Characterization of CS-TPP matrices with FA immobilized
The three types of matrices were dialyzed against 0.05 M Tris buffer, pH 7.0, for 18 h in a 12 kDa membrane (Sigma-Aldrich). The pH was adjusted to 5.6 by adding 0.05 M sodium acetate buffer and lyophilized for the physicochemical characterization and bioassays on A. parasiticus. Matrices yield was obtained with the final weight after liophylization of 1000 mL of the initial solution.

Morphology and size
The average size and morphology of the Mparticle and Mcapsule matrices with immobilized FA were examined using a scanning electron microscope (SEM), JEOL brand, model DSM-54101V, equipped with an EDS. In addition, they were analyzed by transmission electron microscope (TEM), JEOL brand, model JEM-2010F operated at 200 kV (Liu & Gao, 2009;Zhou et al., 2011).

Fourier transformed infrared spectroscopy
The main functional groups in the CS-TPP matrices with immobilized FA were determined using a Fourier transformed infrared spectroscope (FT-IR) (Perkin Elmer Spectrum GX FT-IR) (Liu & Gao, 2009;Zhou et al., 2011).

Zeta potential
The zeta potential is the electric potential in the double interfacial layer; this means, is the place where the diffuse and the Stern layers are bonded. Its value is related to the stability of the colloidal dispersions, indicating the repulsion degree among adjacent particles charged in a dispersion. Likewise, this technique is amply used for electric charge quantification in the double layer. Nparticles CS-TPP (with and without FA) were suspended and washed in two different pH buffer acetate 0.02 M (pH 5.0 or pH 5.6) using a 12 KDa membrane (Sigma-Aldrich). Material inside the membrane was diluted (1:100, v/v) using distilled water to avoid particles aggregation. 2 mL of the CS-TPP matrices with immobilized FA suspension in a cell were tested for zeta potential using a Zetasizer Nano-25 (UK Malvern Instruments) equipment at 25°C (Liu & Gao, 2009).

Total immobilization of FA
The total amount of immobilized FA in the different matrices was determined as free and linked FA. Free FA is the one that is not interacting with the CS-TPP and linked FA is the one that is interacting with the matrix. Free and linked FA in each of the CS-TPP matrices was determined according to the Folin-Ciocalteu method for total phenols. To quantify the free FA, 50 mg of each matrix was pulverized, mixed with 1.0 ml of methanol and vigorously stirred for 30 min (Vortex Fisher Scientific). They were sonicated (Fisher Scientific model 500) at 10% amplitude for 10 min and the amount of free FA in the supernatant was quantified. To determine the amount of linked FA, the same technique as above was employed, except that the sample was sonicated, centrifugated at 10,000 rpm (Hermle Z 326 K), and washed twice with 1.0 ml methanol to remove the free FA. The linked FA was determined in the precipitate (5 mg/mL). A standard curve (R 2 = 0.9967) was prepared using different concentrations of gallic acid (0.02, 0.10, 0.20, 10, 25, 50, 75 mg/mL). Absorbance was measured at 750 nm using a UV-VIS spectrophotometer (Cary 5000 UV-Vis) (Robles-Sánchez, Rojas-Graü, Odriozola-Serrano, González-Aguilar, & Martín-Belloso, 2009). Results were reported as mg of gallic acid equivalents (GAE) per 100 g of sample. FA immobilization efficiency was determined according to the equation of Hosseini, Zandi, Rezaei, and Farahmandghavi (2013): where EE is the Immobilization efficiency, FA T is the total immobilizated of FA (Table 1) and FA I is the initial amount of FA (1433 mg GAE/100 mg).
Antifungal activity of CS-TPP matrices with FA immobilized on A. parasiticus

Radial growth
In order to measure radial growth, the deposition technique was used. When Petri dishes containing 7 mL of agar Czapek dox (solid medium) were at 50°C, 3 mL of a solution containing 15 mg of the CS-TPP matrices with immobilized FA (7:3, v/v) at pH 5.6 was added, mixed, and allowed to cool. A hole of 0.5 cm in diameter was made in the center of Petri dishes.
An inoculum of 1 × 10 5 spores/ml of A. parasiticus was deposited in each hole and was incubated at 28°C. Radial extension growth of the colonies was measured every 12 h and compared with control media until control colony reached the plate border. Controls consisted of agar Czapek dox at pH 5.6 using acetate buffer, CS 0.5% at pH 5.6, TPP 2% at pH 5.6, and matrices without FA at pH 5.6. The rate percentage of growth inhibition (GI)and the radial expansion of the colony (cm h −1 ) were determined from the slope resulting from the radius versus time graph as described by Cota-Arriola et al. (2011) using the acidified agar Czapek dox control as reference. The GI was calculated as follow: where Rc is the mean value of the colony radius in the 5.6 pH agar Czapek control media, and Ri is the colony radius in presence of the different treatments (matrices with and without FA, CS pH 5.6, and TPP pH 5.6).

Spore germination
Four coverslips were placed inside of a 8-cm glass Petri dish with 21 mL of Czapek liquid medium and 9 mL of a solution containing 15 mg of CS-TPP matrices with immobilized FA (proportion 7:3, v/v, respectively) at pH 5.6. Plates were inoculated with 1 × 10 5 spores/mL of A. parasiticus and manually agitated for homogenization. The plates were incubated at 28°C and every 6 h coverslips containing spores was removed and germinated and non-germinated spores were directly counted using an optical microscope (Olympus CX31, Japan) with 40X objective. A spore was considered germinated when the length of its germinal tube reached one-half of the spore diameter (Plascencia-Jatomea, Viniegra, Olayo, Castillo-Ortega, & Shirai, 2003). Controls consisted of agar Czapek dox at pH 5.6 with and without FA at pH 5.6, CS 0.5% pH 5.6, and TPP 2% pH 5.6. The number of germinated spores, the percentage of spore germination inhibition, and the germination rate (spores h −1 ) were determined as described by Cota-Arriola et al. (2011). Spore germination inhibition (FI) was calculated as follow: where Sc was the percentage of spores germinating in the control (agar Czapek dox at pH 5.6) and Si was the percentage of spores germinating in the different treatments and controls. Datos seguidos de su desviación estándar, de medias de cinco experimentos. Valores de tratamientos con letras distintas son significativamente diferentes (P ≤ 0,05).

Morphometric parameters of hyphae and spores
Samples from the spore germination test were taken at 12 and 18 h. The hyphae and spore average diameter were measured with an image analyzer Pro Plus Version 6.3 software (2008 Media Cybernetics Inc., USA) using an optical microscope with a 40X objective (Olympus CX31, Japan) connected to an Infinity 1 camera (Media Cybernetics, USA).

Statistical analysis
Statistics on a completely randomized design were determined using a one-way analysis of variance (ANOVA). JMP software version 5.0 (SAS Institute Inc., USA) with a significance level of P = 0.05, was used. Means between homogenous groups were separated using Tukey's multiple comparison test (Tukey's post hoc test) with a confidence interval of 95%.

Results and discussion
Acquisition and morphology of CS-TPP matrices with immobilized FA Poncelet (2006) (Figure 1) with a yield of 37.9 g/L. They exhibit a yellow color due to the presence of soy lecithin. The irregular shape of the Mparticles and Nparticles may be due to competition between the carboxyl group of the FA and the phosphate group of TPP to interact with the CS amino group. The spherical shape of the Mcapsules could have been due to less interaction among the components in the presence of soy lecithin, which forms a layer between the CS-TPP matrix and the FA.
Infrared spectroscopy of CS-TPP matrices with immobilized FA Infrared spectra from the main components of the matrices with and without FA are presented in Figure 2. The characteristic peaks of CS, TPP, and FA were observed as well as their interactions. Concerning Mparticles and Nparticles without FA (Mparticles and Nparticles), the presence of the characteristic band of the primary amine of CS to 1644-1646 cm −1 was observed (Woranuch & Yoksan, 2013). Also, the band of TPP (P=O) was present, 1103 and 889-750 cm −1 , corresponding to the stretching vibrations of the P=O group, and a band at 540 cm −1 corresponding to the deformation vibration of P=O, which indicated the presence of CS-TPP in both matrices (Ibezim et al., 2011). Similarly, in the spectrum of the Mcapsule without FA (Mcapsule), the characteristic bands of CS and TPP mentioned above were observed, and another band around 1739 cm −1 was observed, which corresponded to the C=O group of lecithin (Wang, Luo, & Xiao, 2014). In the spectra of Mparticle/FA, Nparticle/FA, and Mcapsule/FA, the appearance of a new peak at 1596-1529 cm −1 was detected. It corresponded to the characteristic band of the C=C from the FA aromatic ring, which indicated its immobilization in the matrices (Sun, Wang, Kadouh, & Zhou, 2014). Also, a frequency shift and a decreased intensity of the peaks corresponding to CS and TPP were observed, compared with the spectra of the matrices without FA, indicating an ionic interaction between the CS-TPP and a possible ionic interaction between the carboxyl group of the FA with the CS amino groups.
Total immobilization and immobilization efficiency (EE) of FA in matrices of CS-TPP Table 1 shows data for the total immobilization of FA and the efficiency of immobilization (EE) by the matrices. The immobilization efficiencies of Nparticles and Mparticles were statistically similar. Mcapsules had the lowest immobilization efficiency but were the most efficient to encapsulate free FA because of its larger size, however, they had less linked FA (P ≤ 0.05). This may have been due to the presence of lecithin and Tween that can form an interlayer between CS-TPP and the FA, affecting the ionic interaction between the COO − and the NH 3 + groups of FA and CS, respectively. The immobilization efficiencies of Nparticles and Mparticles were statistically similar, but Mparticles showed higher total FA encapsulation. In addition, Nparticles showed high amount of linked FA, which may have been due to the compact nature and small size of the structure. This brought the carboxyl group of FA and the amino group of CS close enough to favor an ionic interaction. Furthermore, in the Mparticles, significant amounts of linked FA were detected, which may have been due to the concentration of CS used, making the amino groups more available to interact with the FA carboxyl group. Several studies have reported variable values of immobilization efficiency, due to many intervening factors such as concentration, polymer properties, crosslinking agent, degree of crosslinking, size and morphology of the matrix, and others (Agnihotri, Mallikarjuna, & Aminabhavi, 2004).

Zeta potential of CS-TPP matrices with immobilized FA
The zeta potential or surface charge is an essential parameter in the characteristics of a particle, mainly through its influence on stability (Du, Niu, Xu, Xu, & Fan, 2009), and determines the antimicrobial potential on fungi and bacteria (Du, Xu, Xu, & Fan, 2008). Table 1 shows the zeta potential of the elaborated matrices at two pH levels, 5.0 and 5.6. The Nparticle and Mparticle at pH 5.6 had low surface charges (+), which may indicate that not all of the amino groups from CS were loaded and could be due to the excess of TPP added for the crosslinking (Liu & Gao, 2009). However, the immobilization of FA in the Nparticles and Mparticles at pH 5.6 increased their surface charge (+), which may be due to the competition between the FA and TPP for linkage to the amino group of CS or to structural rearrangement of the particle by the immobilization of FA, exhibiting more free amino groups on the surface (Liu & Gao, 2009). In addition, the Nparticles and Mparticles with or without immobilized FA at pH 5.0 had increased surface charges (+) compared with those at pH 5.6. This may have been due to a high protonation of the amino groups, or possibly to a rearrangement of the particles by lowering of the pH, which is in agreement with Liu and Gao (2009). The Mcapsules with or without immobilized FA showed a negative surface charge; this may have been caused by the addition of soy lecithin, which increased the presence of phosphate groups on the surface of the Mcapsule, affecting its electrostatic properties. Zeta potential is a crucial parameter for stability of aqueous nanosuspensions. For a physically stable nanosuspension solely stabilized by electrostatic repulsion, a zeta potential of ±30 mV is required as a minimum (Müller, Jacobs, & Kayser, 2001). All these data suggested that CS Nparticles and CS Nparticles loaded metal ions prepared here were stable. Several studies have reported that the antimicrobial activity of CS under acidic conditions is due to the protonation of -NH 2 on the C-2 position of the D-glucosamine repeat unit (Ali, Joshi, & Rajendran, 2010). Positively charged CS binds to the bacterial cell wall surface which is negatively charged and disrupt the normal functions of the membrane affecting the nutrient transport into the cell or by promoting the leakage of intracellular components. Du et al. (2009) found that antibacterial activity of CS Nparticles loaded metal ions was directly proportional to potential zeta.

Transmitance (%)
Wavenumber (cm -1 ) Figure 2. FT-IR spectra of CS-TPP matrices with immobilized FA. Blue, red, green, and black lines correspond to the peaks of the main functional groups of CS, TPP, FA, and lecithin, respectively.
Antifungal activity of CS-TPP matrices with immobilized FA on A. parasiticus

Radial growth
In fact, the study of Mparticles and Nparticles of CS with antimicrobial potential is having a strong impact, since they have greater effect than CS in solution. For this reason, alternatives were needed to increase their antimicrobial activity by incorporating natural compounds.  -Oliveira et al., 2011). In addition, the fungistatic effect of Nparticles may be related to their small size, having larger contact surface with the cell membrane (Ali et al., 2010). Furthermore, it was found that A. parasiticus growth and spore germination was greater in the presence of Mcapsules and Mcapsules/FA, compared with those in buffer acetate control. This effect can be attributed to the presence of lecithin in the capsules, since it is reported that lecithin accelerates the growth of fungus, such as in Anthracophyllum discolor, which uses it as a carbon source (Bustamante, Rubilar, & Diez, 2014). The addition of FA in Nparticles and Mparticles potentiated its inhibitory effect, compared with the matrices without FA. This effect may be due to an increase in the zeta potential (+) of such matrices for the presence of immobilized FA (Table 2), indicating that more amino groups are loaded, resulting in an increased interaction with the phospholipids of the cell membrane, thereby affecting cell permeability and the development of the fungus (Ali et al., 2010;Saharan et al., 2013). Furthermore, the increase in the fungistatic potential of the Mparticles/FA and Nparticles/FA, can also be attributed to the presence and release of FA (Free and linked FA in CS-TPP matrices). It has been reported that phenolic compounds exhibit antimicrobial effects against bacteria and fungi, due to their ability to interact via hydrogen bonds with proteins and amino acids from the cell wall and cell membrane, and even from cytoplasm, damaging cell permeability (Yi et al., 2014;Nakayama et al., 2013;Bossi et al., 2007;Couzinet-Mossion et al., 2010). It was also mentioned that these phenolic compounds decrease or inhibit the activity of essential enzymes for fungal growth (Bossi et al., 2007;Kandil, Li, Vasanthan, & Bressler, 2012;Yi et al., 2014). Also, phenolic compounds exhibit antimicrobial activity in their action against free radicals, generating a high amount of hydrogen peroxide, causing irreversible damage to microbial cell (Yi et al., 2014). The effects on the morphology of A. parasiticus in the presence of CS-TPP matrices with immobilized FA are shown in Figure 3. It was observed that morphological changes in A. parasiticus were more evident in the presence of the Mparticles and Nparticles of CS-TPP with immobilized FA (Nparticles/FA and Mparticles/FA). Colonies presented white color and cottony mycelia growth, low sporulation and hypha apical growth (Figure 3), compared with those in the controls and matrices without FA, in which the mycelium was yellowish-green with high sporulation and longitudinal growth of the hyphae. Mycelium morphological changes observed in presence of Nparticles/FA and Mparticles/FA may be due mainly to the presence of CS and FA. This effect also can be attributed to the ability of FA to interact with amino acids and enzymes that are important for the development and sporulation of A. parasiticus (Holmes, Boston, & Payne, 2008;Kandil et al., 2012;Nakayama et al., 2013;Yi et al., 2014). In addition, a greater effect on the mycelium due to interactions between the Nparticle or Mparticle with the membrane could be caused by the increase in zeta potential (+) elicited for the addition of FA (Du et al., 2009). It was also observed that the presence of Nparticles/ FA resulted in most significant, exhibiting a completely white and cottony mycelium with little sporulation, which could be due to the small size of the particles, causing greater interaction with the fungus membrane and thereby, greater effect on its morphology and growth. It was also noted that there were no obvious changes in the mycelium and sporulation of the fungus by the Mcapsules, and this was attributed to the presence of lecithin. However, in the Mcapsules/FA, few white and cottony mycelia were observed, probably due to the presence of the FA in the matrix.

Spore germination
Spore germination is the first stage of fungal growth and is indicative of adaptation to the medium, because at this stage, the fungus produces the necessary compounds for its development (Cota-Arriola et al., 2011). It has been reported that calcium (Ca + ) and some enzymes, such as chitin synthase among others, play a fundamental role in the germination, mainly in the development of hyphae  et al., 2003). Table 3 shows the percentages of inhibition of spore germination of A. parasiticus in the presence of CS-TPP matrices with immobilized FA, and in control treatments. At 12 h, it was found that the presence of TPP, Mcapsules, and Mcapsules/FA, did not inhibit the germination of spores. It was also observed that at 30 h, all treatments showed inhibitory effects compared with the control acetate buffer, with a greater effect in the presence of Mparticle/FA and Nparticle/FA. The inhibitory effect presented by CS on the germination of spores of A. parasiticus was attributed to its calcium (Ca + )-chelating potential, the electrostatic interaction between protonated amino groups (NH 3 + ), and the ability of the phosphate ion to electrostatically interact with major ions (Ca + ) for spore germination (Palmeira- de-Oliveira et al., 2011). It was generally observed that Mparticles and Nparticles of CS-TPP potentiated the inhibitory effect of germination compared with TPP and CS solutions. This may have been due to the aforementioned mechanisms (Palmeira-de-Oliveira et al., 2011) and to the formation of the particle (smaller), presenting a greater effect on fungal germination (Ali et al., 2010;Du et al., 2008;Yien, Zin, Sarwar, & Katas, 2012). However, the incorporation of FA into Mparticles and Nparticles potentiated the inhibitory effect on spore germination, and was attributed to the increase of the zeta potential of the particles (Ali et al., 2010;Chowdappa, Gowda, Chethana, & Madhura, 2014;Saharan et al., 2013). In addition, this effect may be due to the capacity of FA (free and bound to the matrix), to inactivate enzymes that are important for fungal development, as occurred in the radial growth phase (Kandil et al., 2012;Nakayama et al., 2013;Yi et al., 2014).

Morphometric parameters and morphological changes of spores and hyphae of A. parasiticus
Changes in morphometric parameters indicated the morphological changes undergone by the fungus in the presence of adverse factors in their adaptation and development. In Table 4, the diameters of spores and hyphae of A. parasiticus in the presence of CS-TPP matrices with embedded FA and control treatments are shown. The spore diameter at 12 and 18 h increased due to the effect of CS in solution and to the CS-TPP matrices with immobilized FA (Nparticles/FA, Mparticles/FA and Mcapsules/FA), compared with other treatments. In addition, the length of hyphae was shorter, with    (Deacon, 1993). Shorter and irregular hyphae in the presence of CS were observed, attributed to alterations in the cell wall or to changes in pressure and wall tension (Cota-Arriola et al., 2011). This can also be due to the lack of calcium (Ca + ), which is important for the formation and elongation of the hyphae (Plascencia-Jatomea et al., 2003). Figure 4 shows the morphology of spores and hyphae of A. parasiticus. It was observed that the presence of CS (Figure 4(b)) caused an increase in the diameter of spores and shorter hyphae compared with control and TPP (2.0%), in which average morphology was observed (Figure 4(a,c)). The morphology of A. parasiticus in the presence of Nparticles and Mparticles (Figure 4(d,e)) showed more noticeable changes, displaying shorter and irregular hyphae, relative to the control. In addition, a possible interaction between Nparticles and Mparticles with spores and hyphae was observed. However, in the presence of Mcapsules (Figure 4(f)), no obvious change was observed in hyphae or spores, nor in the interactions between the Mcapsule and fungus. This was likely because on the surface of the capsule, there are no cationic amino groups (NH 3 + ) present to interact with the fungus membrane, being consistent with the results of Table 1 (zeta potential negative). The most obvious changes in morphology were presented in the fungus exposed to the presence of matrices with incorporated FA (Figure 4(g-i)), where deformations in spores and hyphae were observed. Morphological changes were due to the aforementioned mechanisms in the radial growth phase Datos seguidos de su desviación estándar, de medias de cincuenta mediciones. Valores de tratamientos con letras distintas son significativamente diferentes (P ≤ 0,05).  and spore germination. Moreover, they were due to the increase in the zeta potential of the particle, causing greater interaction and to the presence of FA (free and bound to the matrix of CS-TPP), which can inhibit enzymes that are important for fungal development (Kandil et al., 2012;Yi et al., 2014), such as chitin synthases, responsible for catalyzing the polymerization of chitin. Furthermore, it has been reported that when the activity of chitin synthase is interrupted to form chitin in Aspergillus species, hyphae cell walls swell, appearing abnormal conidiophores (McIntyre et al., 2001).

Conclusions
The results of this study indicate that the CS-TPP matrices are effective in the immobilization, protection, and release of FA. The immobilization efficiency of FA depends on the matrix obtained and how it is in the matrix (free or linked), as we found that Mparticles had higher total immobilization of FA. Mparticles exhibited higher total immobilization of FA, and the interaction with CS matrices had effect on matrix morphology as well as physicochemical properties, depending on the matrix, increasing zeta potential (+) of Mparticles and Nparticles. These changes affected the fungistatic properties against A. parasiticus; moreover, the incorporation of FA (free and linked) into Mparticles and Nparticles significantly enhanced its fungistatic activity on growth, spore germination, and morphology of A. parasiticus; however, Mcapsules enhanced fungus growth. These results indicate that the application of CS matrices to control toxigenic fungi will depend on the composition, structure, and FA immobilization in the matrices. In addition, this study suggests that the development of CS Mparticles and Nparticles by incorporating natural antimicrobial compounds as FA is a promising technology for future application in food and agriculture areas, mainly for control of socioeconomic important toxigenic fungi.