Obtaining size-controlled microcapsules by ionic gelation with high and low acyl gellans containing Lactococcus lactis

Control the diameter of microcapsules obtained with functional biopolymer is a crucial parameter in the success of food applications, since it affects the protection of microencapsulated microorganism and also in the texture of the final product. The aim of this study was to assess the obtaining of controlled size microcapsules containing Lactococcus lactis, using mixtures of high acyl gellan (HA) and low acyl gellan (LA). A concentration of 0.2% (w/w) gellan was employed using a simple design, generating the following mixtures: 100HA/0.0LA, 0.0HA/100LA, 25HA/75LA, 50HA/50LA and 75HA/25LA. The diameter of the microcapsules, efficiency of microencapsulation and viability of the microencapsulated microorganism were studied in function of the speed of agitation (400-800 rpm) and surfactant concentration (sorbitan monooleate) (0.00.2%)v/v. The results indicated that mixtures with concentration equal or greater than 50% of HA gellan are not efficient for obtaining microcapsules, only the LA gellan and the mixture 25HA/75LA gave acceptable results. The viability of the microorganism and the efficiency of microencapsulation were descending function of the stirring speed and surfactant concentration. The microcapsules obtained had diameters not greater than 80 μm when the highest concentrations of surfactant (0.2% v/v) and stirring speed (800 rpm) were used, suggesting that the ionic gelation can be used to obtain microcapsules of controlled size (15-75 μm) containing Lactococcus lactis with high viability (83.32%) and high efficiency of microencapsulation (82.4%), which makes it feasible for use in food applications.


Introduction
Microencapsulation is a process by which certain bioactive substances are retained within a matrix or wall system in order to protect them from deleterious environmental conditions, prevent loss and gradually release them under controlled conditions (Anal and Stevens, 2005;Anal et al., 2006;Kailasapathy and Masondole, 2005;Yáñez et al., 2002).A microcapsule consists of a strong, thin, semi-permeable and spherical membrane around a solid or liquid core with a diameter that varies from a few microns to 1 mm (Goncalves et al., 1992).
There are many techniques to carry out the microencapsulation of various compounds and microorganisms, among which there may be mentioned spray drying, extrusion, fluidized bed, simple or complex coacervation, liposomes, inclusion in complexes (Gouin, 2004), spray coating, interfacial polymerization, and ionic gelation (Thies, 1996), the latter being a developed process for immobilizing a cell, which uses an anionic polymer mainly alginate as component of the membrane, in combination with divalent ions such as calcium to induce gelation (King, 1988).Some of the drawbacks that have in the production of microcapsules are the so large diameters of microcapsules that are obtained (Adikhari et al., 2003;Poncelet et al., 1992;Moslemy et al., 2004) obtained by internal gelation microcapsules of gellan with diameters between 34 and 265 µm as a decreasing function of the stirring speed, concentration of surfactant and emulsion time.
The average diameter of the microcapsules can be an important factor in the stability and efficiency in microencapsulation, i.e. large microcapsules generally provide greater protection than the smaller diameter microcapsules, but have a poor dispersion in the final food product.However, small diameter microcapsules have low microencapsulation efficiency.Therefore, there must be an optimum diameter of microcapsules, not only for a good protection of the microorganism but for a good distribution of the microcapsules in the food matrix used (Zhao et al., 2008) In the food area, lactic acid bacteria (LAB) are widely used in fermented products (Fanema, 1996).Among the LAB, Lactococcus lactis is the main microorganism used by many industries as starter culture in manufacturing a wide range of dairy products such as cheese, fermented milks and creams, which are important due to the technological, nutritional and beneficial properties to the health of consumers that they have (Stanton et al., 2001).Thus the microencapsulation of this bacterium is of great importance for food applications, since it is a technique that involves the use of non-toxic food grade components such as K-carrageenan, alginate, mesquite gum and gellans because these gums produce soft microcapsules suitable for use in food systems.(Audet et al., 1991;Kailasapathy, 2002;Özer et al., 2009).
However, the use of gellan gum as encapsulating matrix for the protection of microorganisms has been recommended in fermentation processes because of its mechanical and thermal stability.Norton and Lacroix (1990) and Camelin et al. (1993) found that gellan gels can provide mechanical stability to the matrix for the immobilization of Bifidobacterium longum (the gels were stable after 150 h of fermentation in serum), also the biocatizalyzed activity (production of lactic acid) was much greater compared with the values reported for B. longum entrapped in K-carrageenan.The gellan gum is an anionic linear polysaccharide gel-forming secreted by the bacterium Sphingomonas paucimobilis (formerly Pseudomonas elodea) (Banik et al., 2000) approved for use in foods by the FDA in 1992(Pszczola, 1993).It consists of monosaccharaides β-1,3-Dglucose, β-1,4-D glucuronic acid and α-1,4-L-rhamnose in molar ratios 2:1:1 (Chandrasekaran and Radha, 1995;Jansson et al., 1983).In the native gellan also called high acyl gellan (HA) the glucose residue A contains a substitution of a L-glycerate group at C2 and an acetate group to a substitution level of approximately 50% in the C6 (Kuo and Mort,1986).
By contrast, the low acyl gellan (LA) is produced by removal of acetate and glycerate groups with a strong alkali treatment at high temperatures.This structural difference between the two gellan molecules causes a great disparity in the rheological and functional properties of the two biopolymers (Mao et al., 2000).So they lación (82.4%), cuando se utiliza la mezcla 25HA/75LA a 800 rpm y 0.2% v/v de surfactante, lo cual la hace factible para su uso en aplicaciones alimentarias.

Recibido: diciembre 10 de 2012
Aprobado: noviembre 12 de 2013 are also used as thickeners or gelling agents in various food applications and biotechnology.Both gellan molecules form a three dimensional net under appropriate aqueous conditions (Chandrasekaran et al., 1988).
Gellan gum is used for the microencapsulation of viable cells at lower concentrations than those are normally employed when using alginate, K-carrageenan and agar (Buitelaar et al., 1988;Nilsson et al., 1983).It has been reported that gellan is capable of forming gels at low concentrations (0.05%); however, concentrations commonly used in the food industry are in a range between 0.2 to 0.4% (Kelco International, 1991).Therefore, the gellan gum can be an attractive alternative in the microencapsulation of bacteria of interest in food using low concentrations of the biopolymer.It is important to mention that gellans are resistant to high temperature, extreme acidic conditions (Brownsey et al., 1984;Jay et al., 1998) and has better technical properties for the microencapsulation than other biopolymers as alginate (Sun and Griffiths, 2000).
The main objective of this study was to evaluate the possibility of obtaining controlled size microcapsules containing Lactococcus lactis using binary mixtures of gellan evaluating microencapsulation parameters such as efficiency, viability of lactic acid bacteria and size of the microcapsules obtained, may be representing an useful alternative in food applications.

Preparation of dispersions
The dispersions were prepared separately in deionized water at a concentration of 0.2% w/w, using a simple design of mixtures of HA, LA gellan and the mixtures 25HA/75LA; 50HA/50LA; 75HA/25LA.Subsequently calcium was added (30 mM) (Huang et al., 2003) and they are dispersed by constant stirring at 90°C/ 10 min.

Preparation of emulsion
For the experimental design an orthogonal design L9 3 2 was used where the factors were the concentration of surfactant (0.0 -0.2% v/v) and the stirring speed (400 -800 rpm), because these parameters are those that most affect the size of the microcapsules (Krasaekoopt et al., 2003;Moslemy et al., 2004).From the dispersions of biopolymers containing the inoculum in a concentration of 10 8 cells/ml of Lactococcus lactis, emulsions were prepared with the addition of Span® 80 to the vegetable oil under constant agitation.Gluconolactone is then added until a pH of 4.2 to initiate the gelation process of the gellans.Finally the oil is removed by adsorption and the microcapsules contained in the aqueous phase are centrifuged at 5000 rpm / 10 min twice with saline solution and stored at 4 °C until its use.

Electron Microscopy
The microcapsules are immersed in a solution of 2.5%v/v glutaraldehyde in phosphate buffered saline (abbreviated PBS) (pH 7.3) for 1 hour at room temperature.Subsequently 3 rinses in PBS for 5 minutes each are performed, the samples are dehydrated with ethanol 50, 60, 70, 80, 90 and two changes of 100% for 15 minutes at each concentration.Then, the samples were dried at the critical point in the presence of carbon dioxide using a Samdri-780 dehydrator (Tousimis, USA).Subsequently, a gold coating was deposited by a DESK II gold sputter machine (Denton Vacuum, USA) over 1 min.Finally, the microcapsules were observed using a SEM-Scanning Electron Microscope, JSM-35C® (JEOL, Germany); the microcapsules are fixed with double sided adhesive tape commonly used in SEM observations.

Efficiency of microencapsulation
The suspension of microcapsules is centrifuged to separate the free cells, then the bacterial concentration in the supernatant is determined and the encapsulation efficiency is calculated as well: Where A is the total bacteria concentration in the suspension, and B is the concentration of unencapsulated bacteria found in the supernatant (Capela et al., 2007).

Determination of viability of Lactococcus lactis
Lactococcus lactis viability is determined in a flow cytometer.Suspensions of known concentrations of live and dead microorganisms were analyzed to standardize the technique as described by Molecular Probes (2001).Subsequently, is determined the viability of microencapsulated microorganisms.This technique is based on the intersection of the microorganisms and / or microcapsules with an argon laser at 488 nm, that generates fluorescence which enable to distinguish between live microorganisms and damaged microorganisms due to the reactivity of the fluorochromes.The samples were previously incubated in the dark for a time of 20 minutes in presence of fluorochromes (SYTO®9 and propidium iodide) in a ratio (1:1).Bacterial suspensions and stained microcapsules suspensions were placed in the cytometer chamber and when the laser passed through the suspension, fluorescent signals were emitted that were separated according to their wavelengths.These signals are integrated and analyzed with the computer program Summit ® WM Software Version 5.1 (Extreme Networks, Inc. California) generating percentages of live and damaged microorganisms according to the fluorescence emission from the fluorochromes, subsequently the emission response area is divided into three zones, where R1 corresponds to dead microorganisms, R2 corresponds to live microorganisms and finally R3 that represents inactive material samples to the fluorochromes.

Statistical analysis
Differences between mean values of the viability and efficiency of microencapsulation were determined with the Tukey's test (p<0.05)and one way analysis of variance (ANOVA) was performed using the software SPSS ver.13.0.All determinations were done in triplicate.

Formation of Microcapsules
The microencapsulation method used consists in forming an emulsion between two phases, one hydrophobic and one hydrophilic, where by agitation, a great number of drops are originated which are gelled by acidification with gluconolactone, since calcium is released by substitution reaction of the salt used (CaCO3).In this study microcapsules were obtained with the mixture 25HA/75LA and LA gellan with a 1:2 water/oil rate, under varying conditions of stirring speeds (400, 600 and 800 rpm) and surfactant concentrations (0.0, 0.1 and 0.2%v/v), where it was allowed the separation of the phases by applying a centrifugal force (5000 rpm/10min) at the required time.
Binary mixtures of 25HA/75LA gellan with concentrations higher than 50% of HA gellan were not efficient in obtaining microcapsules, due to the formation of a gel at high temperatures (70º C) (Matsukawa and Watanabe, 2007) which prevented the formation of the emulsion, making it easier obtaining elastic gels (San-derson et al., 1988), with which it is not possible to maintain the shape of the microcapsules; precluding the microencapsulation process.Microcapsules were obtained only with the LA gellan and the 25HA/75LA mixture because LA gellan provides gels firmer than the HA gellan (Sanderson, 1990).One possible explanation for this phenomenon is that the HA gellan needs higher temperature for the conformational order than LA gellan (Morris et al., 1996), this is probably due to the presence of the acyl groups present in the HA gellan, preventing the association between the biopolymer chains (Chandrasekaran and Thailambal, 1990).

Microencapsulation
The microcapsules obtained in all experiments performed in this study showed a unimodal behavior, this behavior may be explained by the slow release of calcium ions from calcium carbonate because of disruption of the gluconolactone (Pelaez and Karel, 1981), to place into the carboxyl groups of gellan, initiating the gelation process (Larwood et al., 1996;Tang et al., 1997), having sufficient time to have the size and shape of the final microcapsules.
Some parameters which are evaluated in the selection of materials used in the microencapsulation are efficiency, stability under storage conditions and the observing surface by scanning electron microscopy, among other observations (Pérez et al., 2003).Hence the microcapsules were observed in the scanning electron microscope, showing the existence of uneven and irregular structures and surfaces jagged in the microcapsules obtained with LA gellan (figure 1b),while the microcapsules obtained with the mixture of 25HA/75LA gellan (figure 1a) showed a less spherical structure with more irregular surfaces than those mentioned above; however, these results are within the afore mentioned types of microcapsules by Shahidi and Han (1993) and by Gibbs et al. (1999).Note that the microcapsules prepared with surfactant had a higher fragility (Graphs not shown) during the dehydration process with alcohol and critical point drying, which indirectly indicates a high solubility in polar substances such as alcohol and thus a weak capsular membrane (Yáñez, 2007).
Regarding the diameters of microcapsules, figure 2 shows the distributions of the microcapsules obtained with mixtures of HA/LA gellan and LA gellan under changing conditions of surfactant and stirring speed, where it is observed that the largest diameter distribution is presented in the microcapsules obtained with LA gellan without surfactant at 400 rpm (Figure 2a); where, the distribution of diameters ranges values from 15 to 250 µm, while the microcapsules obtained with LA gellan at higher stirring speeds (800 rpm) and surfactant concentrations (0.2%) showed values from 15 to 75 µm (figure 2c),observing a diminution in diameter of thereof with increasing concentration of the surfactant and the stirring speed; the average diameter of the microcapsules obtained under these conditions was 13.72 µm, being a diameter similar to that reported by Homayouni et al. (2008) who reported that microcapsules with a diameter of 17.89 µm increase survival in a 30% of Lactobacillus casei and Bifidobacterium lactis in ice cream.
Likewise the microcapsules obtained with mixtures 25HA/75LA showed the same behavior, i.e. with increasing surfactant concentration and stirring speed (Figures 2d, 2e and 2f) decreased the diameter thereof.The highest frequencies for the microcapsules obtained with mixture of 25HA/75LA were obtained for the diameters of between 15 and 80 µm (Figure 2f), which represents the ideal size for food applications, since a smaller diameter results in decreased cell protection and a larger diameter may result textural defects in the food product in which they are used (Lacroix et al., 2005).
By contrast when higher stirring speeds and surfactant concentrations were used, the highest frequencies were between 15 and 75 µm and this similar behavior reported by Kim, et al. (2008) who obtained diameters of 75 µm in microcapsules prepared with sodium alginate, note that microcapsules of diameters less than 80 µm are desired at the industrial level because they do not affect the sensory properties of foodstuffs in which they are employed (Tyle, 1993).
The microcapsules obtained in this work at lower stirring speeds have diameters between 15 and 250 µm, which are also within the optimum diameters reported by Robitaille et al. (1999) whereby the standard diameter for microcapsules should be <350 µm considering only the size of the dairy organism (0.6-0.9 µm x 1.5 µm).Similarly, in this work are reported lower diameters than those of Jankowski et al. (1997), who obtained microcapsules with diameters between 5.2 and 5.7 mm by means of the coacervation technique using sodium alginate and potato starch as covering material; also the diameters are lower than those published by Groboillot et al. (1993) who used chitosan to encapsulate Lactococcus lactis achieving an increase in the diameter of the microcapsules from 220 to 629 µm by increasing the concentration of chitosan from 1 to 4%; this difference in sizes can be explained since the concentration of gellan over microencapsulation process (0.2%) was kept constant in this work.
In general, it can be seen that the microcapsules obtained for both the mixture with 25HA/25LA and LA gellan by increasing the concentration of surfactant, the size of the microcapsules decreases (Poncelet et al., 1990) this is due to diminution in the surface tension leading to a better emulsion (Adamson, 1982), coupled with increased stirring speed, facilitates disruption of the droplets (Moslemy et al., 2004) causing small microcapsules.In this work the surfactant concentration was limited to 0.2% because of its potential toxic effect.With regard to the microencapsulation efficiency, the microcapsules obtained with the mixture 25HA/75LA showed the highest average percentages (92.5%) compared with those obtained with LA gellan (86.9%), this can be because the mechanism of gelling of the HA gellan does not require calcium ions to gel, it ne-eds only to be subjected to heating in order to initiate a cross-linking between gellans chains, which is started with the temperature drop (Banik et al., 2000).It was noted that efficiency is a descendant function of the stirring speed and surfactant concentration, since there were differences statistically significant (p<0.05) for all treatments applied (table 1), these results are contrary to those published by Gharsallaoui et al., (2007) where it is mentioned that the lack of interfacial properties leads to an increase in the percentage of microencapsulation.This decrease in efficiency can be attributed to some organisms may become trapped in the oil when preparing the microcapsules.

Viability of Lactococcus lactis
A method for distinguishing and quantifying live and dead microorganisms was employed with the aid of a flow cytometer, even in mixtures containing both types of microorganisms.This method uses a mixture of two fluorochromes that stain the nucleic acids and which are excited when exposed to an argon laser at 488 nm which is the cytometer (Molecular Probes, 2001), this permits the analysis of the bacteria on the surface and inside the microcapsules (Demirci et al., 2003;Yáñez et al., 2008).Figure 3 shows typical graphs obtained by the red fluorescence versus green fluorescence.
After microencapsulation the number of live microorganisms of Lactococcus lactis decreased approximately 11%, taking into account all the treatments applied at the time of microencapsulation.This result probably took place because of stirring speeds used; suggesting the effect of shear rate on the survival of the microorganism (Arnaud et al., 1992;Capela et al., 2007).However, the number of microencapsulated microorganisms can increase compared to free microorganisms when used in food fermentations (Kailasapathy, 2006;Krasaekoopt et al., 2006).
The samples which contained only Lactococcus lactis showed 97.32% of viability (figure 3a).With respect to the results of microencapsulation and after applying the ANOVA (one way) there were differences statistically significant (p<0.05)among all the viabilities obtained, highlighting the effect that both the stirring speed as the presence of surfactant have on this parameter.
Viability percentages obtained showed a decrease in bacterial viability of 10.32% (figures 3b and 3c) in the microcapsules of LA gellan and 7.91% (figures 3d and 3e) for microcapsules obtained with the mixture HA/ LA, suggesting an inverse relationship between increasing the stirring speed and concentration of the surfactant with microbial viability.
The cellular microencapsulation has a protective effect on bacteria when they are introduced in food systems, so it is possible to apply the microcapsules obtained in this work to food systems because high percentages of microencapsulation were obtained (> 80%), which are larger to those reported by Yáñez et al. (2008) who obtained viabilities of 46.7% in the microencapsulation of Lactobacillus spp.using a mixture of gellan and Columns with no common letter differ significantly at a confidence level of 95% mesquite gum; this marked difference in the viabilities is due to a technique using a cross-linking agent such as glutaraldehyde, which may affect the viability of the microorganism.However, ionic gelation is a technique originally designed for the cells microencapsulation, thus results with high viability are obtained in comparison with those obtained by other authors (Champagne et al., 1992).

Conclusions
Under the conditions studied it was not possible to obtain microcapsules with HA gellan and/or mixtures with concentrations equal to or greater than 50%.The microcapsules of 25HA/75LA mixtures showed higher average microencapsulation efficiency (92.5%) and percentage of viability (86.84%) than those obtained with low acyl gellan (86.9 and 84.98%, respectively).With regard to the parameters evaluated, both the stirring speed as the concentration of surfactant showed to be descending functions on the size of the microcapsules, microencapsulation efficiency and microbial viability.
It is possible to obtain microcapsules with mixtures of 25HA/75LA gellan and LA gellan with diameters between 15 and 80 µm using stirring speeds of 800 rpm and surfactant concentrations of 0.2%v/v, with a size suitable for food applications; therefore, results of this study suggest that the ionic gelation can be used to provide microcapsules of controlled size containing Lactococcus lactis with viabilities and microencapsulation efficiencies greater than 70% which makes it ideal for food applications.Experiments using dairy product are currently underway in our laboratory.

Table 1 .
Efficiency of gellans microcapsules obtained at different concentrations of surfactant and stirring speeds.