Enhancement of Lactobacillus reuteri KUB-AC5 survival in broiler gastrointestinal tract by microencapsulation with alginate–chitosan semi-interpenetrating polymer networks




To evaluate an alginate–chitosan microcapsule for an intestinal probiotic delivery system for broilers.

Methods and Results

Lactobacillus reuteri KUB-AC5 was successfully microencapsulated with alginate and chitosan mixtures using an emulsion cross-linking method with high microencapsulation efficiency. Scanning electron microscopy revealed a large number of the bacteria entrapped in the semi-interpenetrating network. The microcapsule effectively protected the cells against strong acids. The in vitro study showed that the 8 log CFU g−1 was released at the jejunum and ileum. For the in vivo study, the number of probiotics was detected by a polymerase chain reaction-based random amplified polymorphic DNA technique. From provision of 10 log CFU, cell numbers of 5–8 log CFU were observed in the intestine.


The alginate–chitosan microcapsule can serve as a potential intestine-targeted probiotic delivery system.

Significance and Impacts of the Study

To the best of our knowledge, this is the first comparison study of the in vitro and in vivo gastrointestinal profiles of microencapsulated probiotics used as feed additives for broilers. This study reveals the similarities and differences of the in vitro and in vivo probiotic profiles and provides convincing evidence of the potential use of the alginate–chitosan microcapsule as a probiotic delivery system.


The growth in public health concerns over antibiotic resistance has resulted in an increase in the demand for suitable alternatives to antibiotics in animal production. Normally, antibiotic use has been intended for growth promotion, disease prevention and disease treatment. However, concurrent with long-term use, pathogenic bacteria have developed resistance to antibiotics leading to a serious health problem through the transference of antibiotic resistance genes to human and animal pathogens (Laxminarayan et al. 2013; Paphitou 2013). As a result, the European Union has implemented bans on the use of antibiotic growth promoters for livestock since 2006 (EC 2001, 2003) and the Center for Veterinary Medicine of the U.S. Food and Drug Administration has issued a strategic guidance to promote the judicious use of probiotics in food producing animals of antibiotics (FDA 2012). Therefore, there is an urgent need to explore alternatives to antibiotics to reduce the health and economic impacts of antibiotic misuse.

Probiotics have gained increasing attention as alternatives to antibiotics during the last two decades due to the increasing amount of evidence published in high-impact journals. The most frequently mentioned definition of probiotics is that of the Food and Agriculture Organization/World Health Organization stating that probiotics are live micro-organisms which when administered in adequate amounts confer a health benefit on the host (FAO/WHO 2002). The probiotic efficacy is mainly attributed to two mechanisms of actions, that is, providing nutritional benefits (Anadón et al. 2006) and manipulating gut microbiota (Guillot 2003). Several researchers have tried to gain more insight into the latter mechanism and demonstrated that the modulation of gut microbiota played a critical role in improving the growth performance of animals (Mountzouris et al. 2007; Giannenas et al. 2012; Han et al. 2013).

One of the important selection criteria of a preferable probiotic as a feed additive is the ability to withstand the detrimental environmental conditions of the gastrointestinal tract of animals in order to reach the intestine in a viable state and a sufficient amount. Nevertheless, numerous studies showed that probiotics could not maintain their viability during exposure to the low pH level in the upper gastrointestinal tract (Huang and Adams 2004; Iyer et al. 2005; Ross et al. 2008; Trabelsi et al. 2013) and to bile salt in the small intestine (Shima et al. 2009; Santini et al. 2010; Shi et al. 2013). Accordingly, viability improvement of probiotics for oral administration has been an extensive area of research.

Several studies have shown that microencapsulation is a promising solution to the problem of the low survival of probiotics. Microencapsulation is an emerging technology for packaging biomolecules or cells in an encapsulation matrix that can provide a desired release characteristic and a physical barrier against adverse environmental conditions (Champagne and Fustier 2007; de Vos et al. 2010). Although a variety of microencapsulation techniques and materials have been reported, the choice generally depends upon the application. Among the available techniques for microencapsulation of a probiotic additive for foods and feeds, the emulsification cross-linking technique has been frequently used because it is easy to scale up and the obtained microcapsule provides high protection ability (Chen et al. 2007). For the protection of probiotics or other bioactive compounds against gastric acids, an alginate–chitosan semi-interpenetrating polymer network (semi-IPN) has received much attention as a microencapsulating material (Sezer and Akbuga 1999; Albarghouthi et al. 2000; Lin et al. 2005).

The alginate–chitosan semi-IPN is a polymer blend of alginate and chitosan whose structure is configurationally different. Alginate is an anionic polysaccharide composed of alternating blocks of 1–4 linked α-L-guluronic and β-D-mannuronic acid residues, while chitosan is a cationic polysaccharide composed of randomly distributed β-(1–4)-linked D-glucosamine and N-acetyl-D-glucosamine. Alginate and chitosan form a noncovalent network by interlacing to each other on a polymer scale and alginate is additionally strengthened by an interfacial cross-linking with calcium ions (Sperling 1977). The semi-IPN usually has a better performance compared with either of these alone owing to its double network structure (Myung et al. 2008). The alginate–chitosan semi-IPN microcapsule is less porous than the alginate microcapsule; thus, it is more resistant to acid hydrolysis in the stomach (Gombotz and Wee 1998). Likewise, the intestine-targeted delivery property of the alginate–chitosan semi-IPN microcapsule is more effective than the chitosan microcapsule as the chitosan after complexation with alginate is less soluble in the gastric acid (George and Abraham 2006).

Nonetheless, most probiotic microencapsulation studies have been carried out in vitro using simulated gastrointestinal fluids to evaluate the protection and controlled release properties of the microcapsules. The experimental conditions are basically concerned with the incubation pH and time, so they may not properly represent the situations arising inside animals where other factors, that is, feed intake, feed types, feed passage rates and transit time are also involved. Consequently, in vivo experiments are absolutely necessary to verify the successful application of the microcapsules in the probiotic delivery.

The present study was undertaken to investigate the in vitro and in vivo gastrointestinal profiles of Lactobacillus reuteri KUB-AC5, a probiotic model, microencapsulated in the alginate–chitosan semi-IPN polymer network as broiler feed additives. Lact. reuteri KUB-AC5 was originally isolated from chicken intestine. It showed antimicrobial activity against Gram-negative bacteria such as Escherichia coli and Salmonella sp. (Nitisinprasert et al. 2000) and exhibited high adhesion activity to chicken mucus (Nitisinprasert et al. 2006). In the current study, the probiotic microcapsules were prepared using an emulsification cross-linking technique, and their microencapsulation efficiency and surface morphology were then characterized. Comparisons of the survival of the microencapsulated cells and free cells under various conditions of pH, bile salts, and simulated gastrointestinal fluids were performed. Finally, the alginate–chitosan semi-IPN microcapsules were examined in vivo for their potential use as an intestine-targeted delivery system for the probiotic.

Materials and methods

Probiotic strains and culture conditions

Stock cultures of Lactobacillus reuteri KUB-AC5 were preserved in de Man, Rogosa and Sharpe (MRS; Difco, Sparks, MD) broth supplemented with 20% (v/v) glycerol at −80°C. The working cultures prepared monthly from the stock cultures were maintained at 4°C on MRS agar containing 0·6% (w/v) CaCO3. To increase the bacterial cell numbers, a single colony of the cell was dispensed in 5 ml MRS broth at 37°C for 24 h. The 10% inoculum was then transferred into 25 ml MRS broth and incubated at 37°C for 12 h. The cultures were subsequently transferred into 250 ml MRS broth and incubated under the same condition for 12 h. The cell suspensions were harvested by centrifugation at 7445 g at 4°C for 15 min. The cell pellets were collected and washed twice with 0·85% (w/v) NaCl and then resuspended in the same solution for subsequent encapsulation of the probiotics. The cell numbers were quantified by a standard plate count technique. Appropriate serial dilutions of the culture solution were plated, in duplicate, on selective medium of MRS agar containing 0·6% (w/v) calcium carbonate under anaerobic conditions (AnaeroPack Rectangular Jar, volume 2 l with AnaeroPack, Anaero, Mitsubishi Gas Chemical Co., Inc., Tokyo, Japan) at 37°C for 48 h.

Microencapsulation of Lactobacillus reuteri KUB-AC5 with alginate–chitosan semi–IPN

Lactobacillus reuteri KUB-AC5 was microencapsulated using the emulsification cross-linking method developed by Chitprasert et al. (2012) with slight modification. A 5·5 ml sample of the cell suspension was mixed with 11 ml of 3 g l−1 sodium alginate solution (Fluka, Steinheim, Germany). The cell-alginate aqueous mixture was dispersed in 11 ml of 2 g l−1 chitosan solution (MW 227 000, %DAC 87%, Seafresh Chitosan, Thailand) dissolved in 2% (v/v) acetic acid using a mechanical overhead stirrer (RW 20n; IKA Laboratory, Staufen, Germany) with a four pitched blade turbine at 800 rev min−1 for 10 min. To obtain the water-in-oil emulsion, the cell–polymer suspension was added dropwise to 68·75 ml palm oil (Lam Soon, Samutprakarn, Thailand) premixed with 0·025% (w/v) Span 60 (Fluka, Steinheim, Germany) and stirred at 400 rev min−1 for 30 min. A cross-linking agent, 0·05 mol l−1 calcium chloride solution (Ajax Finechem; Auckland, New Zealand) was dispended slowly down the side of a beaker containing the emulsion to harden the microcapsules and break the emulsion. After cross-linking for 2 h, the microcapsule suspensions were filtered through Whatman No.1 filter paper under vacuum, and the obtained microcapsules were washed twice in 0·3% (v/v) Tween 80 (Ajax Finechem) with an agitation speed of 200 rev min−1 for 10 min. The microcapsules were kept in sterile 0·1% peptone solution at 4°C until use.

Determination of microencapsulation efficiency

Lactobacillus reuteri KUB-AC5 entrapped in the alginate–chitosan semi-IPN microcapsules was released by dissolving in 0·1 mol l−1 sodium tripolyphosphate at pH 7 for 2 h at 37°C with horizontal shaking at 250 rev min−1. The cell suspensions were diluted to appropriate concentrations and pour plated in MRS agar supplemented with 0·6% (w/v) CaCO3. The plates were incubated anaerobically at 37°C for 48 h. The microencapsulation efficiency (ME), which is an indicator of both the efficacy of encapsulation and the survival of the cells during the encapsulation process, was calculated as:

display math

where N is the number of encapsulated cells released from the microcapsules (log CFU g−1 microcapsule) and NI is the number of cells initially added to the polymer solution (log CFU g−1).

Characterization of surface morphology

The samples were prepared in a manner similar to that described in Chitprasert et al. (2012). The microcapsules were chemically fixed in 2% (v/v) osmium tetroxide solution for 1 h, washed twice in distilled water for 10 min each time, and consecutively dehydrated in 30, 50, 70, and 90% (v/v) ethanol solution for 10 min and in 100% (v/v) ethanol solution three times. Then, they were subjected to critical point drying. The dried microcapsules were mounted on an aluminium stub using double-sided conductive adhesive tape. The internal morphology of the microcapsules was exposed by freezing in liquid nitrogen prior to fixation and cutting with a razor blade. The samples were sputter coated with gold using ion sputtering coaters. The internal and external surface morphology of the microcapsules was then examined with a scanning electron microscope (SEM; JSM-5600, JEOL, Tokyo, Japan) at an accelerating voltage of 10 kV.

Survival to pH

Free and microencapsulated Lact. reuteri KUB-AC5 were suspended in phosphate buffer saline with pH values of 1·8, 2·5, 4·0, 6·5, and 7·2 adjusted with 1 mol l−1 HCl or NaOH and filtered through a syringe filter (Nalgene syringe filters 13 mm, pore size 0·2 μm, Nalgene, New York, NY). The incubation was carried out in a rotary shaker at 100 rev min−1 and 42°C. Aliquots were taken every 10 min for pH 1·8 and every 20 min for 180 min for the other pH levels. The number of viable cells was determined according to the method described above. The results were expressed on a Log 10 scale. The experiments were carried out in triplicate, and the results were expressed as means of three observations.

Survival in bile salts

Free and microencapsulated Lact. reuteri KUB-AC5 were suspended in phosphate buffer saline (pH 7·4) supplemented with 0·3% (w/v) broiler bile salts, kindly provided by Betagro Agro Group Public Co., Ltd. (Bangkok, Thailand) and incubated at 42°C and 100 rev min−1. The samples were taken hourly for 6 h, and the cell enumeration was performed as described previously.

Survival in simulated broiler gastrointestinal fluids

Free and microencapsulated Lact. reuteri KUB-AC5 (1% (w/v)) were subjected to a model digestive system comprising a sequential incubation in broiler gastric and intestinal fluids. The gastric fluids consisted of 6·28 g l−1 of HCl, 3·86 g l−1 of NaCl, and 4·37 g l−1 of KCl. The pH levels were adjusted to 6·3, 1·8, and 2·5 with 1 mol l−1 HCl at 0, 60, and 100 min of incubation to simulate the crop, proventriculus and gizzard juices, respectively. The gizzard fluid was also supplemented with 3·2 mg ml−1 pepsin (Sigma–Aldrich, St. Louis, MO). An aliquot from each treatment was taken to determine viable cell counts after 60, 100 and 160 min incubation at 42°C and 100 rev min−1. After that, the cell suspension was filtered through a 0·22 μm membrane and the retentate was consecutively transferred to the intestinal fluids composed of 5·85 g l−1 NaCl, 4·84 g l−1 KCl and 5·88 g l−1 NaHCO3. The pH levels were adjusted to 6·4, 6·6 and 7·2 with 1 mol l−1 NaOH at 0, 40 and 80 min to simulate the duodenal, jejunal and ileal juices, respectively. The duodenal fluid was also added with 10 mg ml−1 Pancreatin (Sigma–Aldrich) and 0·3% (w/v) bile salt. For cell viability determination, an aliquot was taken at 40, 80 and 120 min.

Preparation of microencapsulated probiotic supplemented feed

A sample of 5 g of the microencapsulated cells was added into 100 g of feed composed of 6·47 g fishmeal, 32·46 g soybean meal, 56·96 g cornmeal and 4·11 g palm oil and mixed thoroughly to obtain the final concentration of approx. 8 log CFU g−1. Chromic oxide (0·1 g) as an indigestible marker that can be visually detected was also added to the probiotic supplemented feed and control feed.

Animals, treatments, and sampling

In total, five Arbor Acres broilers aged 24 days were used in the study of the in vivo gastrointestinal profile of the microencapsulated Lact. reuteri KUB-AC5. They were randomly divided into two groups consisting of two in the control and three in the probiotic treated group. Each broiler was individually allocated to a stainless steel, wire-bottomed, metabolic cage and fed ad libitum. Each broiler was killed by cervical dislocation after chromic oxide was observed in the faeces. The crop, proventriculus, gizzard, duodenum, jejunum, ileum and caecum were collected and washed with 70% ethanol for 30 s and then twice with water. All samples were kept on ice and processed immediately after dissection. The mixtures of digesta and wall-associated bacterial samples from each organ were maintained at −20°C until use. Further investigations of their weights and the total viable cells of Lact. reuteri KUB-AC5 were carried out.

Determination of polymerase chain reaction (PCR) based random amplified polymorphic DNA (RAPD) fingerprinting of target bacterial strains

Eleven target strains consisting of Lact. reuteri KUB-AC5, Lact. salivarius KUB-AC21, Lact. plantarum 299V, Lactplantarun, Lact. fermentum KUB-J92, Lactococcus lactis ssp. lactis p1, Lc. lactis ssp. lactis p2, Lc. lactis ssp. lactis sb2, Lact. sakei ssp. sakei JCM 1157, Lact. reuteri KUB-D28, Enterococcus feacal 972 and Lact. salivarius KUB-I48 often found in the chicken gastrointestinal tract were analyzed for their fingerprints by a PCR-based randomly amplified polymorphic DNA method (RAPD) using the single primer OPA3 (5′-AGTCAGCCAC-3′; Truong et al. 2000). The PCR was performed by the colony PCR according to the modified method of Fukui and Sawabe (2007). A single colony of each target strain was transferred into 0·2 ml of 10× PCR buffer containing 2·5 μl of 20 mmol l−1 MgCl2, 2 μl of 1·25 mmol l−1 dNTP, 1 μl of 10 μmol l−1 Primer OPA3, 0·12 μl of 5 U μl−1 DreamTaq DNA Polymerase, and deionized distilled water was added to obtain the final volume of 25 μl. The PCR was carried out using a T Gradient Thermoblock (Biometra, Gottingen, Germany). The procedure for the reaction was as follows: initial denaturation at 95°C for 10 min, followed by 45 cycles consisting of denaturation at 94°C for 40 s, annealing at 36°C for 1 min, and extension at 72°C for 1 min. The final primer extension was carried out at 72°C for 10 min. After PCR, 5 μl aliquot of product was electrophoresed in 1% (w/v) agarose gel, followed by ethidium bromide staining and then photography under UV light. DNA molecular marker standards (100 bp Plus DNA Ladder, Fermentus, Canada) were included in each agarose electrophoresis run.

Determination of Lactobacillus reuteri KUB-AC5 concentration by PCR−based RAPD analysis

The concentration of Lact. reuteri KUB-AC5 was determined by a colony PCR and analyzed by a RAPD fingerprinting as mentioned elsewhere. A sample of 1 g was homogeneously for 2 min with a Stomacher Lab-Blender 400 (Seward Medical, London, UK) and dispersed in 9 ml of 0·85% (w/v) NaCl solution. A suitable dilution of the mixture was directly spread onto MRS agar containing 0·6% calcium carbonate and incubated under anaerobic conditions at 37°C for 48 h. A single colony presenting a clear zone on the MRS agar was transferred into 0·2 ml of 10× PCR buffer, and the PCR-based RAPD analysis was performed using a protocol mentioned elsewhere. The colonies presenting the typical DNA pattern of Lact. reuteri KUB-AC5 were counted.

Statistical analysis

All experiments were performed in triplicate unless otherwise stated. The data collected in this research were expressed as the mean ± standard deviation. Error bars in the figures represent the standard deviation.


Microencapsulation efficiency and surface morphology of microencapsulated Lactobacillus reuteri KUB-AC5

Primary characterization of the microcapsules, that is, microencapsulation efficiency and surface morphology, was performed. The encapsulation efficiency calculated based on the ratio of the number of live cells released from the microcapsules over the number of cells initially mixed with the encapsulating polymers was 95·84 ± 0·39%. The information of surface morphology provided by the SEM is shown in Fig. 1. The microcapsules were almost spherical in shape with wrinkled surfaces (Fig. 1a). Distinct wrinkles with some porosity were observed at a high magnification (Fig. 1b). In addition, there were no probiotic cells detectable on any of the external surfaces. The cross-section image showed a large number of the short rod-shaped bacteria entrapped in the alginate–chitosan semi-IPN at the inner edge of the microcapsule (Fig. 1c). At a high magnification, the porous (interstitial) fluid spaces between the polymeric chains were visible (Fig. 1d). The embedded cells were also observed in the matrix at the centre of the microcapsule (Fig. 1e), but were less crowded than those at the inner edge. A high-magnification image recorded at the centre of the microcapsules displayed distinct cellular distribution in the polymer network and the morphological defects, which were possibly due to the razor cut (Fig. 1f).

Figure 1.

SEM of whole and fractured alginate–chitosan semi-IPN microcapsules loaded with Lactobacillus reuteri KUB-AC5. (a) Whole microcapsules; (b) outer surface area of microcapsules; (c) inner edge of microcapsules; (d) higher magnification view of inner edge of microcapsules; (e) centre of microcapsules; (f) higher magnification view of centre of microcapsules.

Survival at pH levels and in bile salts

The survival of free and encapsulated probiotic during exposure to different pH levels is illustrated in Fig. 2. It was found that under acidic conditions, the free cells declined more readily than the encapsulated cells. The lowest cell viability was observed in the free cells at pH 1·8 (Fig. 2a). From the initial cell number of approx. 8 log CFU ml−1, the number of viable cells was reduced to <4 log CFU ml−1 at 40 min and to an undetectable level after 60 min. In contrast, the encapsulated cells had only 1 log CFU g−1 reduction within 180 min. At pH 2·0, the viable counts of free cells became undetectable after 140 min, whereas the counts of viable encapsulated cells remained stable. At pH 4·0, a slight decrease was found for the free cells, while at pH 6·5 and 7·2, a slight increase was observed for both free and encapsulated cells after 180 min (Fig. 2b). The survival of free and encapsulated cells exposed to bile salt stress is shown in Fig. 2c. There was a slight increase in the viable counts (around 8·2–9·37 log CFU ml−1) during the test.

Figure 2.

Survival of free (FC) and microencapsulated (EC) Lactobacillus reuteri KUB-AC5 after exposure to different conditions. (a) pH 1·8, 2·5, and 4·0; image_n/jam12517-gra-0001.pngFC pH 1·8, image_n/jam12517-gra-0002.pngFC pH 2·5, image_n/jam12517-gra-0003.pngFC pH 4·0, image_n/jam12517-gra-0004.pngEC pH 1·8, image_n/jam12517-gra-0005.pngEC pH 2·5, image_n/jam12517-gra-0006.pngEC pH 4·0. (b) pH 6·5 and 7·2; image_n/jam12517-gra-0007.pngFC pH 6·5, image_n/jam12517-gra-0008.pngFC pH 7·2, image_n/jam12517-gra-0009.pngEC pH 6·5, imageEC pH 7·2. (c) 0·3% bile salt; image_n/jam12517-gra-0009.pngFC, image_n/jam12517-gra-0010.pngEC.

In vitro profiles of Lactobacillus reuteri KUB-AC5 in simulated broiler gastrointestinal fluids

The change in probiotic numbers during exposure to the simulated gastrointestinal fluids is displayed in Fig. 3. The viable cell counts of free cells remained constant from the initial level to the crop at approx. 8 log CFU ml−1. After that, the cells decreased dramatically to approx. 4 log CFU ml−1 in the proventriculus and to an undetectable level in the gizzard. On the other hand, the number of encapsulated cells detected in the crop was only 3 log CFU g−1, and then, it gradually increased in the stomach (proventriculus and gizzard) and the first section of the small intestine (duodenum) to 5 log CFU g−1. Finally, a sudden increase in the cell counts to 8 log CFU ml−1 was observed in the jejunum and ileum.

Figure 3.

In vitro broiler gastrointestinal profiles of free (FC) and microencapsulated (EC) Lactobacillus reuteri KUB-AC5. image_n/jam12517-gra-0004.pngFC, image_n/jam12517-gra-0010.pngEC.

Random amplified PCR products from various lactic acid bacteria

Eleven lactic acid bacteria (LAB) strains comprising Lact. reuteri KUB-AC5, Lact. salivarius KUB-AC21, Lact. plantarum, Lact. plantarun KUB-D73, Lact. fermentum KUB-J92, Lc. lactis ssp. lactis p1, Lc. lactis ssp. lactis p2, Lc. lactis ssp. lactis sb2, Lact. sakei ssp. sakei JCM 1157, Lact. reuteri KUB-D28, E. feacal 972 and Lact. salivarius KUB-I48 showing similar colony morphology on MRS agar were determined by RAPD as shown in Fig. 4. Lactobacillus reuteri KUB-AC5 had a DNA banding pattern at 900, 1300 and 1700 base pairs which was different from other LAB species and Lact. reuteri strains. Therefore, the typical DNA banding pattern of Lact. reuteri KUB-AC5 was used to evaluate its concentration in each organ.

Figure 4.

Random amplified polymerase chain reaction products on 1·5% agrose gel electrophoresis. Lane M:100 bp Plus DNA Ladder; lane 1: Lactobacillus reuteri KUB-AC5; lane 2: Lact. salivarius KUB-AC21; lane 3: Lact. plantarum 299V; lane 4: Lact. plantarun KUB-D73; lane 5: Lact. fermentum KUB-J92; lane 6: Lc. lactis ssp. lactis p1; lane 7: Lc. lactis ssp. lactis p2; lane 8: Lc. lactis ssp. lactis sb2; lane 9: Lact. sakei ssp. sakei JCM 1157; lane 10: Lactobacillus reuteri KUB-D28; lane 11: E. feacal 972; lane 12: Lact. salivarius KUB-I48.

In vivo profiles of Lactobacillus reuteri KUB-AC5 in gastrointestinal tracts of broilers

Two and three broilers were used to investigate the gastrointestinal profile of Lact. reuteri KUB-AC5 in the control and probiotic-treated groups. When chromic oxide used as an indicator appeared in the faeces (which took about 4 h), the broilers were killed. Feed intake and the weight of digesta and wall-associated bacteria from each organ of the gastrointestinal tract were recorded as shown in Table 1. After 4 h of feeding, the feed intake of each chicken in both the control and probiotic treatments varied over a wide range from 41·16 to 73·16 g. In addition, the weight of digesta and wall-associated bacterial samples from the crop, proventriculus, gizzard, duodenum, jejunum, ileum and caecum also showed high variation. Lactobacillus reuteri KUB-AC5 could be detected only from the gastrointestinal tract of the probiotic treatment. From the initial 10 log CFU, cell numbers of 6–10, 4–5, 6–8, and 5–8 log CFU were observed in the crop, proventriculus, gizzard and intestine, respectively (Fig. 5).

Table 1. Feed intake and digesta and wall-associated bacteria from each organ of the gastrointestinal tract
CharactersDigesta and wall-associated bacteria (g)
ControlProbiotic treatment
Feed intake50·1773·1641·1658·7653·56
Figure 5.

In vivo broiler gastrointestinal profiles of microencapsulated Lactobacillus reuteri KUB-AC5. image_n/jam12517-gra-0011.pngNo. 1, imageNo. 2, image_n/jam12517-gra-0006.pngNo. 3.


Extensive research has revealed that probiotics are considered as a potential alternative to antibiotic growth promoters for chicken (Santini et al. 2010; Giannenas et al. 2012; Tellez et al. 2012). The number of probiotics that reach the intestine in a viable state is one important factor influencing their health benefits for the host (Kailasapathy and Chin 2000; Ross et al. 2005). Therefore, a delivery system for the probiotics is required to protect the cells against the harsh gastrointestinal environment and release the cells at target sites. In this study, an intestine-targeted delivery system for a potential probiotic (Lact. reuteri KUB-AC5) was developed using an emulsification external gelation technique. It is a widely used technique for the microencapsulation of probiotics because it is uncomplicated and customizable for large-scale production (Krasaekoopt et al. 2003). Moreover, it is a noninvasive and effective cell loading technique (Sultana et al. 2000; Chitprasert et al. 2012), so that high microencapsulation efficiency indicating high cell viability during the microencapsulation process combined with high cell retainability can be achieved. In this study, the alginate–chitosan blend was selected as the microencapsulating material. It is a semi-interpenetrating polymer network cross-linked with calcium chloride. The network occurrs due to the electrostatic attraction between the negatively charged carboxyl group of alginate and the positively charged amino group of chitosan. After the complexation between these two polymers, the carboxyl group of alginate was additionally cross-linked with calcium ions. Our results showed that microencapsulation of Lact. reuteri KUB-AC5 with the semi-interpenetrating polymer network of alginate–chitosan using the emulsification cross-linking technique yielded a high microencapsulation efficiency of more than 95%. This is a desirable characteristic of the microcapsule because the high number of initial cells potentially increases the number of viable cells after passage through the digestive system of broilers.

The morphological analysis using the SEM revealed that the probiotics were completely enclosed within the microcapsule; no cells were found on the external surface. No distinct macro and micropores were observed on the outer surface or skin of the microcapsule. The dense surface was possibly due to the strong complex formation between alginate and chitosan, and cross-linking by the calcium ion. The internal morphology analysis revealed that the cells were more crowded at the edge than further into the core of the microcapsule. This might have resulted from higher amounts of the calcium ion cross-linking at the edge during the formation of the microcapsule, leading to a higher density of the polymer matrix (Araki et al. 2005; Cook et al. 2011) and higher cell loading. This is a typical characteristic of a microcapsule produced by an external cross-linking method. In addition, during the microencapsulation process, centrifugal forces drew the probiotics away from the centre of rotation causing greater crowding of cells in the peripheral area of the microcapsule.

The study of the probiotic ability to survive harsh conditions showed that the free cells could not survive in the low-pH environment, but the encapsulated cells exhibited acid tolerance. During the passage through the digestive tract, the probiotic encounters various stresses. Acidity in the stomach is considered as one of the important factors detrimental to the viability of the probiotic (Papagianni and Anastasiadou 2009; Sahadeva et al. 2011; Trabelsi et al. 2013). Hydrochloric acid naturally present in the proventriculus and gizzard is a strong oxidizing agent that can diffuse through cell membranes and destroy important cellular components such as proteins, fatty acids and nucleic acids (Pan et al. 2009). In the proventriculus of broilers, a strongly acidic pH of 1·8 can be found during feed ingestion; therefore, this pH was chosen as the lowest pH for the acid tolerance test. After 40 min incubation at pH 1·8, the number of viable cells dramatically decreased to <4 log CFU ml−1. As the feed retention time in the proventriculus can be up to 40 min, these data suggest that the number of viable cells is possibly not adequate to provide health benefits. On the other hand, only 1 log CFU g−1 reduction in encapsulated cells was observed during 3 h of incubation under the same conditions; thus, the alginate–chitosan microcapsule showed a protective effect against the strong acid. The protective ability of the alginate–chitosan semi-interpenetrating polymer network is attributed to its structural stability, which is greatly influenced by an ambient solution pH. When the pH value of the ambient solution is lower than the pKa of mannuronic acid (3·38) and guluronic acid (3·65) residues of alginate (Li et al. 2009), these groups protonate, resulting in the precipitation of the alginate hydrogel into the insoluble alginic acid. Under the same pH conditions, chitosan whose pKa is 6·5 (Sogias et al. 2010) is still soluble owing to the quaternization of the amino group on D-glucosamine residues. Therefore, the ion–ion interaction between the carboxylate ion of alginate and the ammonium ion of chitosan was destroyed, while the intermolecular hydrogen bond remained. Additionally, a lower strength ion–dipole interaction between the ammonium ion and the carboxyl group was formed. Consequently, lower binding between chitosan and alginate was obtained at a pH less than the pKa of alginate, leading to the swelling of the microcapsule and the penetration of the acid (Dai et al. 2008a). However, only a slight decrease in the cell viability was observed, which indicated the structural stability of the microcapsule against disintegration.

The study of the tolerance of the probiotic to bile salt revealed that bile salt had no effect on the survival of this bacterial strain. The ability to tolerate bile salt in the duodenal loop is considered as one of the major criteria influencing the effectiveness of probiotics (Shima et al. 2009). Bile salts can emulsify lipids, the main components of cellular membranes, facilitating lipid digestion by lipase. Thus, bile salt can cause leakage of the cell membranes and ultimately cell death. However, some strains of Lact. reuteri are able to excrete bile salt hydrolase, which cleaves the amide bond of bile salt releasing free bile acid, thereby reducing its bactericidal effects (Taranto et al. 1999). Even though our experimental results proved that Lact. reuteri KUB-AC5 possessed bile tolerance, the microencapsulation of this acid sensitive probiotic is still necessary to protect the cell from the gastric acid and promote its intestinal colonization.

Besides acid and bile salt, many kinds of enzymes in the digestive system such as pepsin and pancreatin exhibit antimicrobial activities (Priya et al. 2011). Therefore, the cell viability in the simulated gastrointestinal fluid in the presence of these enzymes was also assessed. It was observed that free cells were unable to withstand the acid and pepsin in the upper gastrointestinal fluid, whereas most of the microencapsulated cells safely passed through the gastric parts and reached the jejunum and ileum. After 40 min incubation at pH 1·8 and in the proventriculus, the decrease in the viable count of free cells was approximately the same (4 log CFU). Hence, the deleterious effect of pepsin was not distinct. The decline in cell survival to an undetectable level in the gizzard indicated that a delivery system for this probiotic strain was required and the alginate–chitosan microcapsule was suitable for this purpose. A small number of 3–5 log CFU g−1 of microencapsulated cells was detected in the gastric parts and the duodenum, whereas 8 log CFU g−1 was released in the jejunum and ileum. The release of the probiotic was controlled by the pH sensitive property of the microcapsule. At a pH level less than the pKa of alginate as found in the proventriculus and gizzard, only chitosan was ionized and a partial release of the cell was observed. At a pH level >6·5 as found in the jejunum and ileum, the swelling of the hydrogel network increased considerably as a result of the strong interaction between the carboxylate ion of alginate and water molecules as previously reported by Dai et al. (2008a,b) and Martins et al. (2013). Furthermore, the alginate hydrogel was susceptible to disintegration in the presence of excess monovalent ions (Smidsrød and Skjåk-Braek 1990) as found in the intestine. As a large amount of the viable probiotic should release in the distal parts of the intestine to achieve a beneficial health effect on the host, the alginate–chitosan microcapsule proved to be a suitable intestinal delivery system for the probiotic;nevertheless, the simulated gastrointestinal conditions may not completely correspond to the precise gastrointestinal condition of broilers. Accordingly, the in vivo study was carried out to gain more conclusive insights.

The goal of this research was to investigate the role of alginate–chitosan microcapsules in the protection of Lact. reuteri KUB-AC5 feed additive against the harsh environment in the stomach of broilers and delivery of the live probiotics to the intestine. In the preliminary experiment, the detection of Lact. reuteri KUB-AC5 in the control group suggested that this probiotic strain was neither a normal part of the bacterial flora of the tested broilers nor a contaminant from feeds. Thus, the presence of this bacterial strain in the gastrointestinal tract of the probiotic-treated group was due to its release from the microcapsules. The numbers of detectable probiotics in different organs of the gastrointestinal tract are mainly governed by the feed intake behaviour, feed passage rate, protective efficacy, and release characteristic of the microcapsule and by probiotic stability under gastrointestinal conditions. Owing to the difference in the feed intake behaviour and feed passage rate, all broilers displayed considerable variation in the weight of ingesta obtained from various digestive organs. The crop showed a high cell number of 6–10 log CFU was detected, while only 3–4 log CFU was found in the simulated crop fluid. The unexpectedly high release observed in the in vivo test was likely due to the swelling of the microcapsule accompanying its loss of integrity. The feed stored in the crop may be partially soluble generating a large number of ions that diffused into the microcapsule and destroyed the ionic cross-link between the polymer network, giving rise to the polymer chain relaxation and cell release. In addition, the cell release was triggered by the near neutral pH of the crop fluid. When the released cells in the crop reached the proventriculus, these cells were unable to survive in the highly acidic fluid as shown in the earlier study. Moreover, the low pH level of the proventriculus caused the precipitation of alginate leading to the retardation of the cell release. Because of these two reasons, cell numbers of 4–5 log CFU were observed in the proventriculus. The feed with the microcapsules and released cells then encountered less acidic fluids in the gizzard where it was ground with the aid of swallowed stones and grits; thus, a high cell viability of 6–8 log CFU was recorded. After leaving the gastric part of the system, the nonencapsulated and microencapsulated cells moved into the lower parts of the digestive tract, where cell release from the microcapsule was promoted. Cell number of 5–8 log CFU were detected in the proximal and distal intestine from the initial provision of 10 log CFU. As Lact. reuteri KUB-AC5 is present in the intestine in high numbers, the suppression of nonbeneficial groups of bacteria, that is, Klebsiella, Chryseobacterium, Citrobacter, Aeromonas, Acinetobacter and Campylobacterales is possible as demonstrated by Nakphaichit et al. (2011).

We discussed here the potential use of the alginate–chitosan microcapsule as an intestine-targeted delivery vehicle for the acid sensitive probiotic Lact. reuteri KUB-AC5. Both in vitro and in vivo studies proved that the microcapsule could protect the cell from acid-induced cell death in the upper digestive tract of broilers and deliver the cells in sufficient numbers in the intestine. As the feed retention and pH of the crop of broilers promotes cell release, the results suggest a better chance of success in applying the developed microcapsule with monogastric animals. However, a possible solution to retard the cell release in the crop might be coating the microcapsule with hydrophobic materials that can effectively prevent the penetration of ions and water.


This research was co-funded by the Kasetsart University Research and Development Institute (KURDI) and Betagro Science Center Co., Ltd., Thailand.

Conflict of interest

The authors have no conflict of interest.