Production and Evaluation of Dry Alginate ... - ACS Publications

May 16, 2011 - (2, 3) These alginate microcapsules alone may be used for the microencapsulation of probiotic bacteria,(4) showing improved viability u...
0 downloads 0 Views 3MB Size
ARTICLE pubs.acs.org/Biomac

Production and Evaluation of Dry Alginate-Chitosan Microcapsules as an Enteric Delivery Vehicle for Probiotic Bacteria Michael T. Cook,†,‡ George Tzortzis,§ Dimitris Charalampopoulos,*,† and Vitaliy V. Khutoryanskiy*,‡ †

Department of Food and Nutritional Sciences, University of Reading, Whiteknights, Reading, Berkshire, RG6 6AD United Kingdom Reading School of Pharmacy, University of Reading, Whiteknights, Reading, Berkshire, PO Box 224, RG6 6AD, United Kingdom § Clasado Research Services, Ltd., Science and Technology Centre, University of Reading, Early Gate, Whiteknights Road, Reading, RG6 6BZ, United Kingdom ‡

bS Supporting Information ABSTRACT: This study investigates the production of alginate microcapsules, which have been coated with the polysaccharide chitosan, and evaluates some of their properties with the intention of improving the gastrointestinal viability of a probiotic (Bifidobacterium breve) by encapsulation in this system. The microcapsules were dried by a variety of methods, and the most suitable was chosen. The work described in this Article is the first report detailing the effects of drying on the properties of these microcapsules and the viability of the bacteria within relative to wet microcapsules. The pH range over which chitosan and alginate form polyelectrolyte complexes was explored by spectrophotometry, and this extended into swelling studies on the microcapsules over a range of pHs associated with the gastrointestinal tract. It was shown that chitosan stabilizes the alginate microcapsules at pHs above 3, extending the stability of the capsules under these conditions. The effect of chitosan exposure time on the coating thickness was investigated for the first time by confocal laser scanning microscopy, and its penetration into the alginate matrix was shown to be particularly slow. Coating with chitosan was found to increase the survival of B. breve in simulated gastric fluid as well as prolong its release upon exposure to intestinal pH.

’ INTRODUCTION Alginates are natural polysaccharides derived from brown algae comprising a linear chain of 1f4 linked β-D-mannuronic acid (M) and R-L-guluronic acid (G) residues.1 Because of the presence of carboxylic groups on both monomers, alginate exists as a polyanion in solution. These M and G residues exist within the chain as alternating (MG/GM) or homopolymeric (GG and MM) regions. The relative amount of M and G moieties in the polymer is associated with the species of algae from which the alginate is extracted. By electrostatic interaction with calcium ions, the G homopolymeric blocks within alginate form an “eggbox” structure and cross-link the polymer to form a hydrogel monolith. It is this property that has led to the alginates being commonly used as an encapsulation material by the external gelation technique.2,3 These alginate microcapsules alone may be used for the microencapsulation of probiotic bacteria,4 showing improved viability under simulated gastric conditions. Although with alginate encapsulation alone a “burst release” of large molecules at intestinal pH has been reported,5 this could be reduced by coating with chitosan. The rationale is that this way the release of bacteria would be controlled, spreading their deposition along the intestine and thus delivering probiotics to both the small and large intestine. The latter is the main area at which bifidobacteria temporarily colonize.6 r 2011 American Chemical Society

Chitosan is the N-deacetylated form of chitin, a mucopolysaccharide derived from various crustaceans and insects.7 The Ndeacetylated sections, forming ∼8590% of the polymer, have a 1f4 linked glucosamine structure ((1f4)-2-amino-2-deoxy-Dglucose).8 The amine residues in chitosan are mostly protonated below pH ∼6.5, making chitosan a polycation.9 Chitosan is also a known antimicrobial agent against various bacteria, such as Escherichia coli.1012 To form an enteric microcapsule, we can produce alginate gels by external gelation, followed by subsequent dip-coating into chitosan solution, with the polymers associating via electrostatic interactions between the acid and amine groups. These microcapsules are suitable enteric delivery vehicles because at low pHs associated with the nonfasted stomach (pH ∼25),13 the alginate microcapsules will form acid gels (if calcium ions are sequestered or replaced by monovalent salts) or be stabilized against dissolution by the layer of chitosan on the surface. However, as the microcapsule moves into the intestine, the pH will increase to ∼7, and a higher amount of phosphates will be present, thus sequestering calcium ions. This should result in the dissolution of Received: April 26, 2011 Revised: May 15, 2011 Published: May 16, 2011 2834

dx.doi.org/10.1021/bm200576h | Biomacromolecules 2011, 12, 2834–2840

Biomacromolecules the microcapsule. A further reason that these polymers are particularly suited for controlled release is that both chitosan and alginate are mucoadhesives because of their high charge density, extending the capsule’s residence in the area of release.2,14,15 As a result, chitosan-coated alginate microcapsules have previously been evaluated as potent drug delivery vehicles, showing excellent mucoadhesive and controlled release properties.16 These polymers are of particular interest industrially because they are renewable, cheap and generally recognized as safe (GRAS) by the U.S. Food and Drug Administration. The aim of this work was to use these alginate-chitosan microcapsules to produce a dry enteric delivery formulation for probiotic bacteria. The formulation should both protect the bacteria used as well as provide release into the intestine. The protection given to probiotic bifidobacteria in simulated gastrointestinal (GI) conditions by alginate microcapsules alone has been shown in the literature.4 There has also been work studying the viability of bifidobacteria in alginate-chitosan microcapsules, albeit at wet state, during exposure to simulated GI conditions.17 The aim of this study was to optimize further the coating process, and, in particular, study the coating time and coating thickness. In addition to this, the effect of the type of drying on the properties of the microcapsule was studied. The behavior of the produced microcapsules was studied at a range of pH values in a series of swelling experiments to understand the likely behavior at various parts of the GI tract. Finally, the survival of the bifidobacteria was investigated in simulated GI solutions to evaluate the protective effect of the microcapsules as well as evaluate the potential for control release using a coated delivery system.

’ MATERIALS Sodium alginate (1940 kDa), chitosan (low molecular weight, 103 kDa, degree of deacetylation 85.6% determined by 1 H NMR (D2O/trifluoroacetic acid, 700 MHz, Bruker AV700)18), and fluorescein isothiocyanate (FITC) (isomer l) were purchased from Sigma-Aldrich (Gillingham, U.K.). Bifidobacterium breve NCIMB 8807 was obtained from the U.K. National Collection of Industrial and Marine Bacteria (NCIMB). Wilkins-Chalgren (WC) anaerobe agar and phosphate-buffered saline (PBS) was purchased from Oxoid (Basingstoke, U.K.). ’ METHODS Turbidity of Alginate-Chitosan Solutions with pH. Alginate and chitosan solutions (1 mg/mL) were prepared and mixed in a 2.4:2 ratio. The pH of this polyelectrolyte solution was adjusted using either NaOH or HCl (both 0.1 M). This was followed by the addition of sufficient NaCl (0.1 M) so that the total volume of 0.1 M electrolyte solution added was 1 mL. After the addition of acid or base, we measured the turbidity of the solution using a Jasco V-530 UVvis spectrophotometer (λ = 400 nm). The turbidity of the solution should provide an indication of the pH range over which the polyelectrolyte complexes are formed. Production of Unloaded Microcapsules. Unloaded (without bacterial cells) alginate microcapsules were prepared by external gelation. Sodium alginate solution (2% w/v in water, 1 mL, microfiltered using Minisart Sterile-R 0.45 μm microfilters) was extruded using a syringe and a pump (flow rate 2.0 mL/min) in CaCl2 (0.05 M, 50 mL) and left in solution to harden for 30 min. The microcapsules produced were collected by filtration. In the case of chitosan-coated microcapsules, the microcapsules were then placed in chitosan solution (0.4% w/v in 0.1 M acetic acid, adjusted to pH 6 with 1 M NaOH, 10 mL, microfiltered)

ARTICLE

before being subjected to fluid bed drying (30 °C, air flow: 50% of full power, 15 min, described in detail below). Drying of Microcapsules. Initially, a suitable drying method was established. Four different methods were used to dry the alginate microcapsules produced, namely, air drying, vacuum oven drying (Gallenkamp vacuum oven), freeze-drying (Thermo LL3000), and fluid-bed drying (Retsch TG200). Fluid-bed drying was chosen (as described in the Results and Discussion), and the microcapsules were dried (15 min, 30 °C, air flow: 50% of full power, idle flow: 183 m3 h1). Production of Bacteria-Containing Microcapsules. To produce bacteria-containing microcapsules, we grew B. breve (37 °C, 72 h) in an anaerobic cabinet on Wilkins-Chalgren anaerobe agar before inoculation into trypticase-phytone-yeast (TPY) broth (10 mL). The inoculated cell suspension was grown (37 °C, 22 h) before centrifugation (3200 rpm, 15 min, 4 °C). The supernatant was removed, and the cell pellet was resuspended in PBS to OD600 ≈ 2.0. This cell suspension was then mixed with alginate (2 g in 90 mL of solution) in a 1:9 volume ratio, and the microcapsules were prepared and dried as above.

Swelling and Dissolution under Simulated Gastrointestinal Conditions. Simulated gastric fluids were made containing 0.2% w/v NaCl and adjusted to pH 2, 3, 4, and 5 with 1 M HCl, a suitable pH range for mimicking the conditions in the nonfasted stomach. Simulated intestinal fluid was made by dissolving KH2PO4 (6.8 g) in 750 mL of deionized (DI) water. To this, NaOH (0.2 M) was added to adjust pH to 7.2 ( 0.1, and the solution was made up to 1 L. Unloaded microcapsules (20 mg) were placed in a vial containing simulated gastric or intestinal juice (5 mL) and placed in a water bath (37 °C, with shaking, 100 rpm). For simulated gastric fluid, the weight of the microcapsules was measured every 30 min for 180 min; the microcapsules in the simulated intestinal fluid were measured every 10 min until dissolution. The swelling degree was then calculated by normalizing with respect to the dry weight using eq 1. swelling degree ¼

Wt W0

ð1Þ

where Wt is the weight of the microcapsules at time t and W0 is the weight at time 0. Each swelling experiment was repeated three times, and the mean ( standard deviation is calculated. Synthesis of FITC-Labeled Chitosan. The fluorescein isothiocyanate (FITC)-labeled chitosan was synthesized by modification of a known method.19 Chitosan solution (1% w/v in 0.1 M acetic acid, 100 mL) was prepared, followed by the addition of dehydrated methanol (100 mL) and FITC (2 mg/mL in methanol, 50 mL). The reaction was carried out in the dark at room temperature for 3 h before precipitation in NaOH (0.1 M, 1 L). The resulting precipitate was filtered and dialyzed in deionized water (4 L, replaced daily) until FITC was not present in the dialysis jar, as checked by UV-spectrofluorometry (Jasco FP-6200, λexc: 488 nm, λemi: 515 nm). The dialyzed product was then freeze-dried. The labeling efficiency was established by UV spectrofluorometry (Jasco FP-6200, λexc: 488 nm, λemi: 500600 nm) using standards in a 50:50 water/propan-1,2-diol mixture (a mixture known to be suitable for dissolving FITC),20 in which the deionized water was first adjusted to pH 6 with 0.1 M AcOH. From these data, a calibration was constructed of Imax (the maximum fluorescence intensity) against FITC concentration, and the molar ratio of free amine to FITC-labeled residues was determined to be 7.5:1.

Determination of Coat Thickness by Confocal Microscopy. Wet alginate microcapsules were placed in FITC-labeled chitosan (0.4% w/v in 0.1 M acetic acid made up to pH 6 with 0.1 M NaOH). At time intervals of 1, 10, 20, 30, 1080, 1440, and 2400 min, the microcapsules were removed and examined on a Leica SP2 confocal microscope (λexc: 488 nm). From these images, the coating thickness was determined using the image analysis software ImageJ. Each experiment was repeated in triplicate, and the mean was taken as the coating 2835

dx.doi.org/10.1021/bm200576h |Biomacromolecules 2011, 12, 2834–2840

Biomacromolecules thickness. The images indicated that the surface of the microcapsules is homogeneously coated with chitosan. Viability of Free and Encapsulated B. breve. For nonencapsulated bacteria, B. breve was inoculated in 10 mL of TPY broth,21 and incubated at 37 °C for 22 h. The culture was then centrifuged (3200 rpm, 15 min, 4 °C), the supernatant was removed, and the cells were resuspended in PBS (1 mL) before enumeration. The cell suspension (1 mL) was then added to simulated gastric juice (pH 2, 9 mL) and placed in a water bath (37 °C). At half hour intervals, and up to 2 h, an aliquot (1 mL) of the simulated gastric juice was removed and centrifuged at 10 000 rpm for 10 min. The supernatant was removed, and the cells were resuspended in PBS (1 mL), diluted in PBS, and enumerated using serial dilution and viable counts on WC (Wilkins-Chalgren) agar plates. For microencapsulated bacteria, three samples of bacteria-loaded microcapsules (20 mg) were suspended in simulated gastric juice (10 mL) and placed into a water bath (37 °C). At 0, 1, and 2 h, all of the microcapsules from one sample were removed from solution and placed in PBS (99 mL). After 60 min, the PBS solution containing microcapsules was placed into a stomacher (Seward stomacher 400 circulator) for 30 min. The resulting cells in the suspension were enumerated using serial dilution in PBS and viable counts on WC agar plates. Release of Cells from Microcapsules. We determined the release of cells from the microcapsules by placing the microcapsules in simulated gastric juice (pH 2, 1 h), followed by simulated intestinal juice (pH 7.2, 2 h). During this process, aliquots (1 mL) were taken from the liquid medium at the start of exposure as well as at half hour intervals thereafter and were enumerated using serial dilution and viable counts on WC agar plates. In addition, we estimated total cell counts by counting on a hemocytometer using a Nikon Microphot-SA microscope set to phase contrast mode.

’ RESULTS AND DISCUSSION Drying of Microcapsules. Among the drying methods tested, it was observed that fluid bed drying and freeze-drying produced granular products, as opposed to air drying or vacuum oven drying, which had a tendency to form large clusters unless manually separated (See images in Supporting Information). Freeze-drying produced large, brittle granules, whereas fluid bed drying gave smaller, harder capsules, with a much quicker drying time. It was for the reduction in size and decrease in drying time that the fluid bed drying method was chosen. An example microcapsule is shown in the scanning electron microscope (SEM) image in Figure 1. The irregular structure is most likely a consequence of a diffusion setting process during gelation, which results in a higher polymer concentration on the surface of the capsules. As the capsule dries, the interior (with a lower polymer concentration) shrinks to a greater extent than the exterior regions, giving this kind of “crumpled” structure with decreased size but relatively unreduced surface area. The optimal drying time used for the microcapsules was determined by drying at a relatively low temperature until the weight was constant (15 min at 30 °C was selected, data not shown). To our knowledge, this is the first application of fluid bed drying to this type of system. Turbidity of Alginate-Chitosan Solutions with pH. Mixing aqueous solutions of alginate and chitosan results in the spontaneous formation of polyelectrolyte complexes, which form turbid solutions. The dependence of turbidity on the solution pH was determined (Figure 2). The data show that the solutions are mostly turbid over the region of pH 27, indicating the presence

ARTICLE

Figure 1. SEM image of a fluid bed dried alginate microcapsule.

of polyelectrolyte complexes. On either side of this range, the turbidity decreases suddenly, indicating possible decomplexation of polyelectrolytes. The implication of these data is that the electrostatic interactions between alginate and chitosan occur at the range of pH 27. As a result, when the pH increases above pH 7, the chitosan should start to dissociate from the alginate; this is presumably due to the chitosan macromolecules becoming less charged (pKa of chitosan is ∼6.5). At pHs below pH ∼2, the chitosan layer should start to dissociate because of the alginate becoming less ionic. The above effect of pH is relevant to the alginate-coated microcapsules because interactions between alginate and chitosan should take place during passage through the stomach (pH 25). However, as the pH increases in the intestine (>6), the two polymers should start to dissociate. Swelling and Dissolution in Simulated Gastro-Intestinal Solutions. In this experiment, the objective was to simulate the behavior of the microcapsules during their transit through the stomach and intestine. To explain the data, it can be assumed that if the microcapsules swell until a plateau is reached then the microcapsule will be subsequently stable and thus will retain the bacteria. If the degree of swelling then decreases, it suggests that the microcapsule is dissolving and will release its load. These experiments show a combination of both effects, with the swelling degree indicating a net weight change as a result of both the swelling and dissolution of the capsules. Besides dissolution, the other possible mechanism of release is by diffusion through the pores in the polymer network.22 However, this is highly unlikely to occur because the pores in the alginate gels are typically between 5 and 200 nm in width,22 whereas bifidobacteria are of micrometer size. The swelling behavior of alginate and alginatechitosan microcapsules in simulated gastric solution is shown in Figure 3. A range of pHs was used to show the effect of the changing pH within the stomach on the microcapsules. The data were measured over 3 h because although gastric half-emptying time is ∼80 min,23 any dissolution occurring in the early stages of swelling may be hidden if a shorter time was used. The data in Figure 3 show that at pH 2 the capsules swelled to a steady weight 2836

dx.doi.org/10.1021/bm200576h |Biomacromolecules 2011, 12, 2834–2840

Biomacromolecules

ARTICLE

Figure 2. Turbidity of alginate-chitosan solution as a function of pH. Each experiment was repeated in triplicate, and each data point is given as the mean ( standard deviation. Polynomial fitted to graph is a guide to the eye.

Figure 3. Swelling behavior of dried alginate (squares) and alginate-chitosan (circles) microcapsules at different simulated gastric pH values (n = 3, mean ( standard deviation).

after 60 min in both cases. This result implies that the capsules would remain intact inside the stomach, retaining B. breve. In the pH 3 experiment, the microcapsules continued to swell over the 3 h period. In this case, there appears to be little, if any dissolution of the capsules. When the capsules were exposed to simulated gastric solution at pH 4 and 5, the swelling degree increased over 120 min in all cases, with chitosan-coated alginate microcapsules having a maximum swelling degree ∼15 and 20 arbitrary units higher than that of alginate only, at pH 4 and 5, respectively. After 120 min, the swelling degree decreased, indicating partial dissolution of the microcapsule. The reason for this dissolution is most likely due to the monovalent salts in the gastric juice competing with calcium in binding with the carboxylic groups

within the gels. We believe that the reason that the gels do not dissolve at pH 2 and 3 is that alginate has an acid gel character below the pKa of the urunate residues (∼3.5). The microcapsules swelled to a lesser extent at pH 2 because the acid gel strength increased with increasing proton concentration until pH 2.5.24 The chitosan coating of the microcapsules appears to reduce the rate of dissolution of the capsules at higher pHs, having swelling degrees at 180 min 23 and 12 arbitrary units higher, at pH 4 and 5, respectively, than alginate alone. This indicates that the chitosan layer stabilizes the microcapsules against dissolution at pH 4 and 5. It is important for the capsules to be stable at pH 4 and 5 to ensure that they do not dissolve at the higher pHs of the stomach or in the duodenum, releasing the bacteria early. Chitosan-coated 2837

dx.doi.org/10.1021/bm200576h |Biomacromolecules 2011, 12, 2834–2840

Biomacromolecules microcapsules exhibit similar swelling profiles at pH 4 and 5, whereas the alginate-only microcapsules’ behavior differs between the pH values. The reason for this is unclear, as alginate should not exhibit any acid gel character in this pH range. Further experimentation may reveal additional trends in this pH range, but this is beyond the scope of this publication. Figure 4 shows the swelling of the dried microcapsules in simulated intestinal solution. Because of the very high swelling degrees and therefore high susceptibility to mechanical damage, large errors are associated with each measurement (which are higher in chitosan-coated microcapsules). This confirms that the microcapsules should start to dissolve upon entry to the small intestine. These data indicate that coating with chitosan increases the time needed for complete dissolution of the capsules by ∼10 min, which should extend the release of the bacteria further down the GI tract. Determination of Coat Thickness by Confocal Microscopy. The interactions between oppositely charged hydrogels and linear polymers have been reported in the literature;25,26 however, there has been no quantitative information on how linear polymers penetrate into the gel network. In the present work, wet microcapsules were imaged successfully by a novel confocal laser-scanning microscopy-based method, and the thickness of the visible chitosan layer was measured (Figure 5). This experiment showed that after initial binding to the alginate

Figure 4. Swelling behavior of dried alginate (squares) and alginatechitosan (circles) microcapsules in simulated intestinal fluid. Data given as mean ( standard deviation (n = 3).

ARTICLE

microcapsule through electrostatic interaction the penetration of FITC-chitosan into the gel was extremely slow, the difference in coat thickness being unnoticeable over 30 min. It has been reported27 that the penetration of a linear polymer (PAA) into a hydrogel (2-[(methacryloyloxy)ethyl]trimethylammonium chloride/N-isopropylacrylamide copolymer) occurs quickly, judging by the rate of syneresis and color change of the gel. However, it is likely that in the case of chitosan penetration into alginate, the stiffened chain conformation of the chitosan is caused by intramolecular bonding,28 and the increased size of the polysaccharide monomers (relative to acrylic acid) leads to the decreased rate of penetration. A marked increase at 1080þ min was seen in confocal images, indicating that penetration is indeed occurring, albeit slowly, consistent with the previous hypothesis. From the data, it was decided that probiotic-loaded microcapsules would be coated with chitosan for 10 min because they should show similar results as those coated for short times. This finding could be of particular interest because alginate microcapsules are typically coated for significantly longer periods than this, not uncommonly up to 40 min;29 therefore, these findings may reduce the time needed for microcapsule production by a significant factor. Viability of Free and Encapsulated B. breve. B. breve was shown to have no viability after short exposure to simulated gastric juice (pH 2) (Figure 6), as it decreased from 9.5 log cfu/mL to no viable cells. This suggests that B. breve should not be viable after transit through the stomach (half gastric emptying time is ∼80.5 min).23 The encapsulation process itself resulted in a small loss of viability (∼0.4 log), although this could be due to the process of dissolving the microcapsules rather than the act of encapsulation itself, which uses very gentle conditions. Crucially, there was not a loss in viability associated with chitosan coating. Although chitosan is a known antimicrobial,11 its macromolecules do not penetrate deep into the microcapsules, as demonstrated in Figure 5. There was a 1.1 to 1.5 log decrease in viable cells associated with the drying process, which could possibly be improved if lyoprotectants were used.30 After 60 min of exposure to simulated gastric solution, all encapsulated bacteria demonstrated significantly higher viability relative to the free cells.

Figure 5. Coat thickness as a function of chitosan exposure time. Each experiment was repeated in triplicate, and each data point is given as the mean ( standard error. Insert: Confocal microscopy image showing layer of chitosan on the surface of the microcapsule (green). 2838

dx.doi.org/10.1021/bm200576h |Biomacromolecules 2011, 12, 2834–2840

Biomacromolecules

ARTICLE

Figure 6. Viability of free and encapsulated B. breve in simulated gastric solution (n = 5). Data shown as mean ( standard deviation. p values denoted by * (p < 0.05) and *** (p < 0.001), calculated using ANOVA and Bonferroni posthoc tests. “ns” signifies no statistical difference.

Alginate encapsulation alone increased the viability from 0 to 5.2 ( 0.8 log for the wet and 5.6 ( 1.0 log for the dry microcapsules. Coating these microcapsules with chitosan further improved the viability to 7.3 ( 0.2 log for the wet microcapsules and 6.6 ( 0.5 log for the dry ones. This shows that chitosan coating improved the protective effect of alginate microcapsules (which between wet alginate and wet chitosan capsules is statistically significant, p < 0.05). After 120 min, the difference between alginate and alginate-chitosan microcapsules was starker. Whereas the bacteria in the alginate-only microcapsules lost viability completely, the chitosan-coated alginate microcapsules still contained 6.6 ( 0.6 and 4.9 ( 0.9 log viable cells for wet and dry microcapsules, respectively. Another important observation is that the drying process did not seem to deplete the protective effect of the microcapsules. The protective effect could be due to the polysaccharides acting as a buffer, reducing the activity of the acid. This is quite likely because chitosan, a basic polysaccharide, improved the survival of B. breve considerably. Cell Release from Microcapsules. During passage through the GI tract, the capsules should be exposed first to low pH in the stomach and then to higher pH values as they move into the intestine. The aim of this part of the work was to study how the bacteria were expelled from the capsules to understand at which regions of the gut the bacteria are likely to be deposited. The release profile of B. breve in simulated gastric and intestinal fluids was first determined by viable and total cell counts (Figure 7). According to the viable cell counts (Figure 7A), the chitosancoated microcapsules retained the bacteria during gastric passage and gave continued release of viable B. breve cells under intestinal conditions over 2 h, whereas the alginate microcapsules gave a sudden release of viable cells in the gastric environment before a relatively quick release in intestinal solution, which leveled off after 120 min. From Figure 6, a recovery of 6.6 ( 0.5 log viable cells for alginate-chitosan and 5.6 ( 1.0 log viable cells for alginate would be expected. The viable cells released from the alginate-chitosan microcapsules (6.8 ( 0.02 log, 180 min, Figure 7A) were within this expected range; however, those released from the alginate microcapsules were significantly less than expected (2.7 ( 0.2 log at 120 min). The reason for this is unknown but could be partially due to osmotic damage to the bacteria by the simulated intestinal fluid. To understand the release better, Figure 7B shows the total counts obtained from

Figure 7. Release of cells from alginate and alginate-chitosan microcapsules during sequential exposure to simulated gastric solution (SGS, 1 h) and simulated intestinal solution (SIS, 2 h), by (A) viable counts and (B) total cell counts (n = 3). Blue dotted line shows the transference from gastric to intestinal fluid.

the release experiments; these data show that the chitosan-coated microcapsules produced a slightly more controlled release of the cells during exposure to intestinal solution compared with the alginate microcapsules. This is most likely related to the decreased dissolution time seen for the alginate microcapsules. As in Figure 7A, alginate-chitosan microcapsules retained B. breve while in simulated gastric solution, but alginate microencapsulation alone did not.

’ CONCLUSIONS Dry, probiotic-containing, alginate, and alginate-chitosan microcapsules were produced in a matter that is scalable and produces a uniform, granular product by a drying method not previously used for this system type. Swelling studies indicated that swelling and dissolution were pH-dependent, with dissolution occurring at pH values above the pKa of alginate. Alginate microcapsules, both wet and dry, were shown to improve the survival of B. breve during exposure to simulated gastric juice. Coating with chitosan improved the survival further. When the release profile was measured, coating with chitosan gave no release of the B. breve cells into simulated gastric fluid and a reduction in the rate of bacterial release during exposure to simulated intestinal fluid. The coating process was examined using confocal laser scanning microscopy and showed that the penetration of chitosan into the microcapsule was very slow, and that over a reasonable time scale (130 min) it made negligible difference to the thickness of the chitosan layer. This method was shown to be effective at measuring coat thickness and may be utilized for other polymer-based systems. ’ ASSOCIATED CONTENT

bS

Supporting Information. SEM images showing microcapsules produced by freeze-drying and air-drying techniques.

2839

dx.doi.org/10.1021/bm200576h |Biomacromolecules 2011, 12, 2834–2840

Biomacromolecules This material is available free of charge via the Internet at http:// pubs.acs.org.

’ AUTHOR INFORMATION Corresponding Author

*E-mail: [email protected]; d.charalampopoulos@ reading.ac.uk.

ARTICLE

(27) Nurkeeva, Z. S.; Mun, G. A.; Sergaziev, A. D.; Fefelova, N. A.; Sarsenbaeva, A. S.; Khutoryanskiy, V. V. J. Polym. Sci., Polym. Phys. 2006, 44, 845–853. (28) Nishimura, S. I.; Kohgo, O.; Kurita, K.; Kuzuhara, H. Macromolecules 1991, 24, 4745–4748. (29) Zhou, Y.; Martins, E.; Groboillot, A.; Champagne, C. P.; Neufeld, P. C. J. Appl. Microbiol. 1998, 84, 342–348. (30) Capela, P.; Hay, T. K. C.; Shah, N. P. Food Res. Int. 2006, 39, 203–211.

’ ACKNOWLEDGMENT We wish to thank the Centre for Advanced Microscopy (CfAM) and Dr. Peter Harris for assistance with the SEM images. Mr. Stephen Pountney is also acknowledged for help with confocal laser scanning microscopy. The Central Analytical Facility (University of Reading) is also acknowledged for providing access to the NMR equipment as well as the University of Reading RETF scheme for funding. ’ REFERENCES (1) Kuo, C. K.; Ma, P. X. Biomaterials 2001, 22, 511–521. (2) George, M.; Abraham, T. E. J. Controlled Release 2006, 114, 1–14. (3) Smidsrod, O.; Skjakbraek, G. Trends Biotechnol. 1990, 8, 71–78. (4) Chandramouli, V.; Kailasapathy, K.; Peiris, P.; Jones, M. J. Microbiol. Methods 2004, 56, 27–35. (5) Zhou, S. B.; Deng, X. M.; Li, X. H. J. Controlled Release 2001, 75, 27–36. (6) Gibson, G. R.; Beatty, E. R.; Wang, X.; Cummings, J. H. Gastroenterology 1995, 108, 975–982. (7) Kumar, M. React. Funct. Polym. 2000, 46, 1–27. (8) Shahidi, F.; Arachchi, J. K. V.; Jeon, Y. J. Trends Food Sci. Technol. 1999, 10, 37–51. (9) Sorlier, P.; Denuziere, A.; Viton, C.; Domard, A. Biomacromolecules 2001, 2, 765–772. (10) Helander, I. M.; Nurmiaho-Lassila, E. L.; Ahvenainen, R.; Rhoades, J.; Roller, S. Int. J. Food Microbiol. 2001, 71, 235–244. (11) Rabea, E. I.; Badawy, M. E. T.; Stevens, C. V.; Smagghe, G.; Steurbaut, W. Biomacromolecules 2003, 4, 1457–1465. (12) Singla, A. K.; Chawla, M. J. Pharm. Pharmacol. 2001, 53, 1047–1067. (13) Fordtran, J. S.; Walsh, J. H. J. Clin. Invest. 1973, 52, 645–657. (14) Sogias, I. A.; Williams, A. C.; Khutoryanskiy, V. V. Biomacromolecules 2008, 9, 1837–1842. (15) Bernkop-Schnurch, A.; Kast, C. E.; Richter, M. F. J. Controlled Release 2001, 71, 277–285. (16) Elzatahry, A. A.; Eldin, M. S. M.; Soliman, E. A.; Hassan, E. A. J. Appl. Polym. Sci. 2009, 111, 2452–2459. (17) Krasaekoopt, W.; Bhandari, B.; Deeth, H. Int. Dairy J. 2004, 14, 737–743. (18) Rathke, T. D.; Hudson, S. M. J. Polym. Sci., Polym. Chem. 1993, 31, 749–753. (19) Huang, M.; Ma, Z. S.; Khor, E.; Lim, L. Y. Pharm. Res. 2002, 19, 1488–1494. (20) Boucard, N.; Viton, C.; Domard, A. Biomacromolecules 2005, 6, 3227–3237. (21) Yuan, X. P.; Wang, J.; Yao, H. Y. Anaerobe 2005, 11, 225–229. (22) Gombotz, W. R.; Wee, S. F. Adv. Drug Delivery Rev. 1998, 31, 267–285. (23) Hellmig, S.; Von Schoning, F.; Gadow, C.; Katsoulis, S.; Hedderich, J.; Folsch, U. R.; Stuber, E. J. Gastroenterol. Hepatol. 2006, 21, 1832–1838. (24) Draget, K. I.; Braek, G. S.; Smidsrod, O. Carbohydr. Polym. 1994, 25, 31–38. (25) Gaserod, O.; Smidsrod, O.; Skjak-Braek, G. Biomaterials 1998, 19, 1815–1825. (26) Bartkowiak, A.; Hunkeler, D. Chem. Mater. 2000, 12, 206–212. 2840

dx.doi.org/10.1021/bm200576h |Biomacromolecules 2011, 12, 2834–2840