Food Research International 40 (2007) 12611269 www.elsevier.

com/locate/foodres

Eect of homogenisation on bead size and survival of encapsulated probiotic bacteria

P. Capela
a

a,1

, T.K.C. Hay b, N.P. Shah

a,*

School of Molecular Sciences, Victoria University, Werribee Campus, P.O. Box 14428, Melbourne, Victoria 8001, Australia b DSTO-Scottsdale, Defence Nutrition, Scottsdale, Tasmania 7260, Australia Received 21 March 2007; accepted 14 August 2007

Abstract This study investigated homogenisation techniques for reducing the size of calcium alginate beads during the microencapsulation of Lactobacillus acidophilus, Lactobacillus casei, Lactobacillus rhamnosus and Bidobacterium longum. Bead sizes were reduced using three homogenisation techniques including Ultra-Turrax benchtop homogeniser, Avestin Inc. piston homogeniser and Silverson mixer for various durations, number of passes and pressures. The survival of probiotic organisms during homogenisation was also investigated. The smallest beads (39.2 lm) were created using the Ultra-Turrax benchtop homogeniser at 13500 rpm for 4 min. There was a signicant reduction in the survival of each organism (<5.5%) using the Silverson mixer. However, homogenisation using Ultra-Turrax and Avestin Inc. homogenisers resulted in better survival at 64.4% and 47.7%, respectively. Overall, homogenisation reduced size of beads containing viable probiotic organisms during microencapsulation. 2007 Elsevier Ltd. All rights reserved.
Keywords: Homogenisation; Bead size; Survival; Encapsulation; Probiotic bacteria

1. Introduction The consumption of certain species of probiotic microorganisms is benecial for reducing the duration and severity of diarrhoeal symptoms (Isolauri, Juntunen, Rautanen, Sillanaukee, & Koivula, 1991). Supplementing a diet with food containing benecial bacteria can be used as a strategy for preventing diarrhea (Oksanen et al., 1990; Shahani & Chandan, 1979; Siitonen et al., 1990). Probiotics are dened as living microorganisms, which upon ingestion in certain numbers, exert health benets beyond inherent basic nutrition (Guarner & Schaafsma, 1998). However, for obtaining health benets, a minimum of one million

Corresponding author. Tel.: +61 3 9216 8289; fax: +61 3 9216 8284. E-mail address: Nagendra.Shah@vu.edu.au (N.P. Shah). 1 Present address: DSTO-Scottsdale, Defence Nutrition, Scottsdale, Tasmania 7260, Tasmania, Australia. 0963-9969/$ - see front matter 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.foodres.2007.08.006

probiotic organisms per gram of a food is recommended (Kurman & Rasic, 1991). During processing and storage of foods, probiotic microorganisms may be exposed to high temperatures, low pH, high osmotic pressure and high levels of oxygen (Gardiner et al., 2000; Prasad, McJarrow, & Gopal, 2003; Talwalkar & Kailasapathy, 2003). These factors may have deleterious eects on probiotics. The survival of probiotic organisms is also aected by acid encountered in the stomach and the bile salts in the intestine tract (Chandramouli, Kailasapathy, Peiris, & Jones, 2004; Conway, Gorbach, & Goldin, 1987). Microencapsulation is the process of applying a shell to sensitive microorganisms to protect them from their external environment. Microencapsulation for food applications involves the use of a non-toxic, food grade coating material, such as sodium alginate. When sodium alginate is combined with calcium chloride, a gel matrix is formed creating a shell to protect sensitive probiotic organisms (Shah & Ravula,

1262

P. Capela et al. / Food Research International 40 (2007) 12611269

2000). Other food-grade agents used to immobilize lactic acid bacteria include kappa-carrageenan, locust bean gum (Audet, Lacroix, & Paquin, 1992) and gellan-xanthan (Sun & Griths, 2000). The emulsion technique of microencapsulation involves combining a mixture of probiotic organisms and an encapsulating agent such as sodium alginate with vegetable oil (Capela, Hay, & Shah, 2006; Krasaekoopt, Bhandari, & Deeth, 2004). An alginate/water-in-oil emulsion is formed by agitating the mix, usually with a magnetic stirrer. The alginate beads are formed by slowly adding calcium chloride to the emulsion while stirring. Shah and Ravula (2000) found that calcium alginate beads formed using the emulsion technique improved the viability of probiotic organisms during processing and storage of frozen yoghurt. However, there are several disadvantages of using the emulsion technique including a wide distribution of bead size, diculty in automating the technique and exceedingly large beads with diameters ranging from 200 to 1000 lm (Poncelet et al., 1992). Studies by Audet, Paquin, and Lacroix (1998) and Arnaud, Lacroix, and Choplin (1992) found that encapsulation techniques such as the emulsion technique are conducive to forming large beads which can inuence the texture and mouthfeel of products such as yoghurt. Tyle (1993) found that particles that are generally soft, rounded, and up to 80 lm are not perceived as gritty, but beads exceeding this size may be detectable in the mouth. Techniques employing spray guns (Lee, Hwang, Park, & Park, 2003) and air atomizers (Kwok, Groves, & Burgess, 1991) with varying nozzle sizes have been used to reduce the bead size. The bead sizes could be reduced by using a homogeniser during the encapsulation process to form a ne water-in-oil emulsion containing small droplets. High speed mixers are commonly used in the food industry for homogenising oil and aqueous phases. During homogenisation, the interface between the oil and water is disrupted causing the liquids to blend together. Various mixing heads can be attached to high speed mixers to reduce droplet size by generating intense disruptive forces. High-pressure valve homogenisers (such as the Avestin Inc. and Manton-Gaulin APV) create small droplets by forcing liquid through a narrow valve under high-pressure. A mixture can be passed through a high-pressure valve homogeniser repeatedly to achieve the desired droplet size (McClements, 1999). Food emulsions containing a varied particle size distribution of droplets are referred to as polydisperse emulsions (McClements, 1999). Small particles in a microscopic polydisperse emulsion may not be visible to the naked eye. A particle size distribution diagram illustrating the size range and population of calcium alginate beads can be generated. Particle sizing equipment, such as MastersizerTM 2000 can be used to measure particles ranging between 0.02 lm and 2000 lm by passing the particles through a laser beam (http://www.malvern.co.uk). However, due to the sensitivity of a Mastersizer, probiotic organisms may be

recorded as particles which may inuence the particle size distribution. As the population of probiotic organisms may be aected during the microencapsulation process itself, it is essential to ensure that the selected microencapsulation technique is gentle to sensitive probiotic organisms. Some treatments may have deleterious eects on anaerobic probiotic organism due to the presence of oxygen, heat generated during the process or mechanical shear (Arnaud, Lacroix, Foussereau, & Choplin, 1993; Talwalkar & Kailasapathy, 2003). Individual species of probiotic organisms may vary in sensitivity to external stresses such as those encountered during homogenisation. Cell wall elasticity is thought to improve resistance to mechanical stress due to variations in cell morphology (Scha r-Zammaretti & Ubbink, 2003). The eect of homogenising sodium alginate (containing probiotic organisms) in oil using high speed mixers and high-pressure valve homogenisers in order to reduce the size of calcium alginate beads has not been thoroughly investigated. The aim of this study was to determine the eect of various homogenisers on the uniformity and diameter of calcium alginate beads and the survival of encapsulated Lactobacillus acidophilus 33200, Lactobacillus casei 279, Bidobacterium longum 536 and Lactobacillus rhamnosus GG. 2. Materials and methods 2.1. Preparation of probiotic bacteria B. longum 536 was obtained from the Victoria University Culture Collection (Werribee, Victoria, Australia). The organism was originally obtained from Morinaga Milk Industry Co. Ltd. (Tokyo, Japan). L. casei 279 was obtained from the Australian Starter Culture Collection (Werribee, Victoria, Australia). L. acidophilus 33200 was obtained from the American Type Culture Collection (Manassas, VA, USA), and L. rhamnosus GG was isolated from a commercial product. Each of the organisms was grown in sterile MRS (deMann, Rogosa, Sharp) broth (Oxoid Ltd., Hampshire, UK) using a 1% inoculum. B. longum 536 was supplemented with lter-sterilized 0.05% (w/v) L-cysteine hydrochloride (Sigma Chemical Co., Castle Hill, Sydney, Australia) to create an anaerobic environment. Each culture was grown and propagated three times successively at 37 C for 1215 h for activation. Probiotic organisms were harvested by centrifuging (Sorvall RT7 Newtown, CT, USA) at 1510g at 4 C for 15 min and the cell pellet was suspended in saline solution followed by microencapsulation. 2.2. Microencapsulation of probiotic organism using the emulsion technique The emulsion method of microencapsulation was used to encapsulate probiotics (Sheu & Marshall, 1993). Briey,

P. Capela et al. / Food Research International 40 (2007) 12611269

1263

120 mL of sterile 3% (v/w) sodium alginate was mixed with 30 mL of suspension containing 9.010.0 log10cfu/g of combined L. acidophilus 33200, L. casei 279 B. longum 536 and L. rhamnosus GG. Fifty millilitres of the suspension of sodium alginate and probiotic organisms was gently dispensed using a pipette into a beaker containing 200 mL of sterile vegetable oil (Eta Blended Vegetable Oil, Goodman Fielder Pty. Ltd., Melbourne, Australia) and stirred at 200 rpm using a magnetic stirrer (IEC Industrial Equipment & Control Pty. Ltd., Melbourne, Australia). Calcium chloride (0.1 M) was gently added down the side of the beaker until the emulsion was broken. 2.3. Microencapsulation of L. acidophilus 33200, L. casei 279, B. longum 536 and L. rhamnosus GG using homogenisation (Ultra-Turrax benchtop, Avestin Inc. piston or Silverson mixer) The sodium alginate solution and probiotic organisms were combined with vegetable oil as described in Section 2.2. An emulsion was formed by homogenising the mixture at various speeds, pressures, passes and times using the Ultra-Turrax benchtop (Ika Laboratory and Analytical Equipment, Staufen, Germany), Avestin Inc. piston homogeniser (Avestin, Ottawa, Canada) or Silverson mixer (Silverson Machine Ltd., Waterside, Chesham Bucks, England) as described below. To prepare alginate beads using the Ultra-Turrax benchtop homogeniser, an emulsion was prepared by homogenising 200 mL mixtures (as described in Section 2.2) at 8000 rpm and 13,500 rpm for 2 or 4 min. To prepare alginate beads using the Avestin Inc. piston homogeniser an emulsion was formed by passing the mixture (as described in Section 2.2) through the homogeniser 2 or 3 times at 5 or 10 MPa of pressure. Calcium alginate beads were formed by homogenising 200 mL of the mixtures using the Silverson mixer (described in Section 2.2) for 2 or 4 min. The homogenisation temperature during each of the treatments was 21 C. After each of the homogenisation treatments, each emulsion was transferred into a beaker. Calcium chloride (0.1 M) was gently added to the side of the beaker to the emulsions while stirring using a magnetic stirrer at 200 rpm. The mixture was stirred for 10 min to ensure that the emulsion was completely broken. Small calcium alginate microspheres were formed and subsequently measured using a Mastersizer (Hydro-2000G Malvern Instruments Limited, Worcester, UK). 2.4. Measurement of particle size of calcium alginate beads Calcium chloride solution (0.1 M) containing 15 g of calcium alginate beads was passed through a Mastersizer Hydro-2000G to measure the size of the beads. Samples were added to circulating ltered water until laser obscuration exceeded 10%. The mean particle size was expressed as d32, which represented the area-volume mean diameter.

Acceptability of the bead size was also assessed by examining the d (0.9) values (the diameter of bead for an observation at 0.9 of the distribution). 2.5. Equations used The area-volume mean diameter was determined using the following equation, in which d represented the droplet diameter, n represented the number of droplets in a class. P ni d 3 d 32 Pi1 i2 i1 ni d i The following equation was used to determine the survival of probiotic organisms during homogenisation: y z Survival % 100 x where x is the initial population of probiotic organisms prior to homogenisation; y is the population of probiotic organisms encapsulated within calcium alginate beads; z is the population of probiotic organisms in the external uid surrounding the calcium alginate beads. The following equation was used to evaluate the eciency of the homogenisation treatments: y z Efficiency % 100 x where x is the initial population of probiotic organisms prior to homogenisation; y is the population of probiotic organisms encapsulated within calcium alginate beads; z is the population of probiotic organisms in the external uid surrounding the calcium alginate beads. 2.6. Statistical analysis Data analysis was carried out using SPSS Inc. software (14.0: SPSS Inc., Chicago, IL). The signicant dierence between means (level of signicance P = 0.05) was analysed using one-way ANOVA. Multiple comparisons between means were analysed using Tukeys test. The enumeration of probiotic population was independently replicated three times (n = 3), with two measurements per replicate. 2.7. Enumeration of L. acidophilus 33200, L. casei 279, B. longum 536 and L. rhamnosus GG within calcium alginate beads The entrapped probiotic organisms were released from the calcium alginate beads by sequestering calcium ions with phosphate buer at neutral pH. Once released, probiotic organisms were selectively enumerated using the techniques of Tharmaraj and Shah (2003). L. acidophilus was selectively enumerated on MRS-sorbitol agar using anaerobic incubation at 37 C for 72 h. L. casei was enumerated on LC agar using anaerobic incubation at 37 C for 72 h. B. longum was selectively enumerated on MRS-NNLP (nalidixic acid, neomycin sulphate, lithium chloride and

1264

P. Capela et al. / Food Research International 40 (2007) 12611269

paromomycin sulphate) agar incubated at 37 C for 72 h and L. rhamnosus was enumerated on MRS-vancomycin agar using anaerobic incubation at 43 C for 72 h. 3. Results and discussion 3.1. Eect of homogenisation on bead size The eect of homogenisation on the size of calcium alginate beads is shown in Table 1. Each homogenisation treatment aided in the formation of calcium alginate beads with area-volume mean diameters (AVMD) that were signicantly smaller (P < 0.05) than the control (381.2 lm). Large beads in the control samples were due to the emulsion being created by agitation using a magnetic stirrer. Ideally, particle size should be below 80 lm, otherwise the mouth feel of products such as yoghurt, as well as overall acceptability of the product is aected (Tyle, 1993). There was no signicant reduction (P < 0.05) in bead size when the homogenisation speed was increased from 8000 rpm to 13,500 rpm. Nor was there a signicant reduction in particle size when the duration of homogenisation
Table 1 Eect of homogenisation on the size (lm) of calcium alginate beads using Ultra-Turrax T 25, Avestin Inc. piston or Silverson mixer Treatment d32 (lm) areavolume mean diameterd 381.2 29.5 118.6ab 37.2
ce

was increased from 2 to 4 min (when using the Ultra Turrax). The AVMD of beads created using the Silverson mixer (41.6 lm after 2 min and 42.9 lm after 4 min) were signicantly smaller than those created using the Avestin Inc. piston homogeniser (191.6 lm after three passes at 5 MPa). Acceptability of calcium alginate beads were assessed by examining d (0.9) values (which represented the diameter of the bead at the 90th percentile). The values of all beads formed by instrumental homogenisation (209.1519.8 lm) were signicantly smaller than the control (1312.2 lm) at d (0.9) as shown in Table 1. However, despite these observations being signicantly smaller (P < 0.05) than the control, a proportion of these beads is still unacceptable for incorporation into a commercial product as many of these particles exceeded the 80 lm sensory perception threshold proposed by Tyle (1993). The large dierences between AVMD of the beads and d (0.9) values indicate an unacceptably wide distribution in diameters of the beads. Particle size distributions are illustrated in Figs. 16. Beads which were formed using a conventional emulsion
12 11 10 9 8 Volume (%) 7 6 5 4 3 2 1

d (0.9) (lm)

Control Ultra-Turrax benchtop homogeniser 8000 rpm for 2 min Ultra-Turrax benchtop homogeniser 8000 rpm for 4 min Ultra-Turrax benchtop homogeniser 13500 rpm for 2 min Ultra-Turrax benchtop homogeniser 13500 rpm for 4 min Silverson Mixer for 2 min Silverson Mixer for 4 min Avestin Inc. piston Homogeniser 5 MPa 2 passes Avestin Inc. piston Homogeniser 5 MPa 3 passes Avestin Inc. piston Homogeniser 10 MPa 2 passes Avestin Inc. piston Homogeniser 10 MPa 3 passes
a,b,c

1312.2 51.7 248.0a 154.3

be

82.1ab 15.6

506.6a 173.5

0.01

0.1

10 Particle size (m)

100

1000 3000

52.6ab 7.7

360.7a 80.9

Fig. 1. Particle size distribution of calcium alginate beads prepared using the emulsion microencapsulating technique; Control a, - - - Control b, Control c, . . . . Control d.

39.2a 4.6

209.1a 4.9
8

41.6a 4.2 42.9a 4.9 93.1ab 4.9

224.2a 50.3 290.9a 0.6 330.6a 45.0

7 6

Volume (%)

191.6b 51.9

519.8a 132.3

5 4 3 2 1

121.6

ab

4.5

444.4 82.8

102.4ab 11.2

296.9a 13.4

Means in the same column with dierent superscripts are signicantly dierent (P < 0.05). d Results are expressed as means standard error of two repeated experiments. e Data point is expressed as means standard error of four repeated experiments.

0.01

0.1

10 Particle size (m)

100

1000 3000

Fig. 2. Particle size of calcium alginate beads prepared using the UltraTurrax T 25 benchtop homogeniser at 8000 for 2 or 4 min; 8000 rpm for 2 min, - - - - 8000 rpm for 4 min.

P. Capela et al. / Food Research International 40 (2007) 12611269

11 7 6 5 Volume (%) 4 3 2 1 Volume (%) 10 9 8 7 6 5 4 3 2 1 0.01 0.1 1 10 Particle size (m) 100 1000 3000 0.01 0.1 1 10 Particle size (m) 100

1265

1000 3000

Fig. 3. Particle size of calcium alginate beads prepared using the UltraTurrax T 25 benchtop homogeniser at 13,500 rpm for 2 or 4 min; 13,500 rpm for 2 min, - - - - 13,500 rpm for 4 min.

Fig. 6. Particle size of calcium alginate beads prepared using an Avestin Inc. piston homogeniser at 10 MPa after two and three passes; two passes, - - - - three passes.

7 6 5 Volume (%) 4 3 2 1

technique (Fig. 1) displayed the largest span of particle size. The wide peaks in this distribution highlighted the lack of uniformity in the diameters of calcium alginate beads. The narrowest spans were recorded for beads produced using the Ultra-Turrax or Silverson mixer (Figs. 24). Beads formed using the Ultra-Turrax and Silverson mixer were smaller than the control due to mechanical disruption of particles encountered during the emulsion stage of microencapsulation (McClements, 1999). Small particle size was also obtained using the Avestin Inc. piston homogeniser (Figs. 5 and 6). This was possibly due to the eect of high-pressure on the size of particles formed during the emulsifying stage (as described in Section 3.2).
0.1 1 10 Particle size (m) 100 1000 3000

0.01

3.2. Eect of homogenisation on the survival of encapsulated probiotic organisms The eect of homogenisation using Ultra-Turrax benchtop, Avestin Inc. piston or Silverson mixer on survival of L. rhamnosus GG, L. casei 279, L. acidophilus 33200, and B. longum 536 is shown in Figs. 710. L. rhamnosus proved to be the most robust organisms as it displayed the highest percentage survival (64.4%) when homogenised for 2 min at 8000 rpm using the Ultra-Turrax (Fig. 7). However, there was a signicant reduction (P < 0.05) in the survival of L. rhamnosus when homogenisation was increased from 8000 rpm (64.4%) to 13,500 rpm (25.5%). A similar reduction in the survival of L. casei was identied when the homogenisation speed was increased (Fig. 8). These results suggest a mechanical shear eect on the survival of certain probiotic organisms using an Ultra-Turrax homogeniser. The survival of L. rhamnosus (Fig. 7) was signicantly lower than the control after homogenisation using the Silverson mixer. This treatment was the most severe as counts of all probiotic organisms were below 5.5%. The reduction in viability of probiotic organisms was possibly due to heat and mechanical shear generated during homogenisation (Arnaud et al., 1993). Mechanical energy in the system

Fig. 4. Particle size of calcium alginate beads prepared using a Silverson Mixer for 2 or 4 min, 2 min; - - - - 4 min.

12 11 10 9 8 Volume (%) 7 6 5 4 3 2 1 0.01 0.1 1 10 Particle size (m) 100 1000 3000

Fig. 5. Particle size of calcium alginate beads prepared using an Avestin Inc. piston homogeniser at 5 MPa after two and three passes; two passes, - - - - three passes.

1266
100 90

P. Capela et al. / Food Research International 40 (2007) 12611269

L. rhamnosus

Survival of L. rhamnosus (%)

80 70 60 50 40 30 20 10 0 P 5 MPa - 2 passes P 5 MPa - 3 passes P 10 MPa - 2 passes P 10 MPa - 3 passes SM 2 min SM 4 min UT 8000rpm 2 min UT 8000rpm 4 min UT 13500rpm 2 min UT 13500rpm 4 min Control- Not homogenised
Control- Not homogenised

Treatment
Fig. 7. Eect of homogenisation with Ultra-Turrax T 25 (UT), Avestin Inc. piston (P) and Silverson mixer (SM) on the counts of L. rhamnosus GG within a mixture of sodium alginate and vegetable oil during microencapsulation.

100 90 80

L. casei

Survival of L. casei (%)

70 60 50 40 30 20 10 0
P 10 MPa - 2 passes UT 13500rpm 2 min UT 13500rpm 4 min P 10 MPa - 3 passes P 5 MPa - 2 passes UT 8000rpm 2 min UT 8000rpm 4 min P 5 MPa - 3 passes SM 2 min SM 4 min

Treatment

Fig. 8. Eect of homogenisation with Ultra-Turrax T 25 (UT), Avestin Inc. piston (P) and Silverson mixer (SM) on the counts of L. casei within a mixture of sodium alginate and vegetable oil during microencapsulation.

was most likely responsible for the heat energy due to the viscosity of the sodium alginate solution (McClements, 1999). Homogenisation using the Silverson mixer may have

also incorporated air into the mixture, thus aecting oxygen sensitive probiotic organisms such as B. longum 536 (Talwalkar & Kailasapathy, 2003).

P. Capela et al. / Food Research International 40 (2007) 12611269

100 90 80

1267

L. acidophilus

Survival of L. acidophilus (%)

70 60 50 40 30 20 10 0
UT 13500rpm 2 min UT 13500rpm 4 min P 5 MPa - 2 passes UT 8000rpm 2 min UT 8000rpm 4 min P 5 MPa - 3 passes P 10 MPa - 2 passes P 10 MPa - 3 passes SM 2 min SM 4 min Control- Not homogenised

Treatment
Fig. 9. Eect of homogenisation with Ultra-Turrax T 25 (UT), Avestin Inc. piston (P) and Silverson mixer (SM) on the counts of L. acidophilus within a mixture of sodium alginate and vegetable oil during microencapsulation.

100 90 80

B. longum

Survival of B. longum(%)

70 60 50 40 30 20 10 0 P 5 MPa - 2 passes P 5 MPa - 3 passes P 10 MPa - 2 passes UT 13500rpm 2 min UT 13500rpm 4 min P 10 MPa - 3 passes UT 8000rpm 2 min UT 8000rpm 4 min Control- Not homogenised SM 2 min SM 4 min

Treatment
Fig. 10. Eect of homogenisation with Ultra-Turrax T 25 (UT), Avestin Inc. piston (P) and Silverson mixer (SM) on the counts of B. longum within a mixture of sodium alginate and vegetable oil during microencapsulation.

Little heat was generated during homogenisation using the Avestin Inc. piston homogeniser. During this form of homogenisation, probiotic organisms were subjected to pressures of 5 and 10 MPa. There was a signicant reduc-

tion (P < 0.05) in the population of L. acidophilus when the pressure was increased from 5 to 10 MPa (Fig. 9). High-pressure has been found to have a deleterious eect on the survival of probiotics organisms (Ulmer, Ga nzle, &

1268

P. Capela et al. / Food Research International 40 (2007) 12611269

All organisms
100 95 90

Efficiency (%)

85 80 75 70 65 60

P 5 MPa - 2 passes

P 5 MPa - 3 passes

P 10 MPa - 2 passes

P 10 MPa - 3 passes

UT 8000rpm 2 min

UT 8000rpm 4 min

UT 13500rpm 2 min

UT 13500rpm 4 min

Treatment
Fig. 11. Eciency of homogenisation process using Ultra-Turrax T 25 (UT), Avestin Inc. piston (P) and Silverson mixer (SM). (Eciency is expressed as the percentage of probiotic organisms retained in calcium alginate matrix and not lost to surrounding uid).

Vogel, 2000). Interestingly, pre-treating a material containing L. rhamnosus to mild pressure (100 MPa for 10 min) prior to exposure to an elevated temperature (60 C) has been found to improve the survival of probiotic organisms (http://www.tu-berlin.de/~foodtech/ResearchFINAL). Homogenisation time was set at 2 or 4 min using the Ultra-Turrax and Silverson mixer to investigate whether there was a time eect on the survival of the probiotic organisms (Figs. 710). There was no signicant dierence in populations of probiotic organisms for any of the organisms homogenised at 2 or 4 min, with the exception of L. acidophilus using the Ultra-Turrax, as there was no reduction after initial cell death. Therefore, a time eect was not observed under the conditions of this study. Overall, the Ultra-Turrax and the Avestin Inc. piston homogeniser had the least impact upon the counts of viable probiotic organisms. The survival of probiotic organisms was expected to be highest for each of the controls (Figs. 7 10) as the solutions were not homogenised during these treatments. However, unexpectedly, the survival of L. casei, L. acidophilus and B. longum in control samples was lower than some Ultra-Turrax and Avestin Inc. piston homogenisation treatments. 3.3. Eciency of homogenisation technique Producing calcium alginate beads using homogenisation was not highly ecient as probiotic organisms were lost to

the uid surrounding the beads. The comparative eciencies of the homogenisation techniques are shown in Fig. 11. The Ultra-Turrax and Piston homogenisers were not signicantly dierent (P < 0.05) to the control. However, the Silverson mixer was the least ecient process as the eciencies after treatments for 2 or 4 min of homogenisation were 79.08% and 88.88%, respectively.

4. Conclusion It was possible to reduce the bead size below 100 lm using the Ultra-Turrax homogeniser. The Ultra-Turrax homogeniser was the most suitable for preparing calcium alginate beads with a narrow range of diameters. The population of L. acidophilus 33200, L. casei 279, B. longum 536 and L. rhamnosus GG was not adversely aected during homogenisation using Ultra-Turrax benchtop and Avestin Inc. piston homogenisers. Homogenisation using a Silverson mixer had a negative impact upon probiotic populations and would not be recommended for the preparation of calcium alginate beads. However, microencapsulating probiotic organisms using these techniques was laborious, time consuming, and produce a large amount of waste oil. The waste oil also contained small sodium alginate/ probiotic droplets that were dicult to remove by centrifugation. Therefore, further research is required to develop an automated technique for the large scale production of

Control- Not homogenised

SM 2 min

SM 4 min

P. Capela et al. / Food Research International 40 (2007) 12611269

1269

encapsulated probiotics in calcium alginate beads (in a continuous large scale operation). References
Arnaud, J. P., Lacroix, C., & Choplin, L. (1992). Eect of agitation rate on cell release rate and metabolism during continuous fermentation with entrapped growing. Biotechnology Techniques, 6(3), 265270. Arnaud, J. P., Lacroix, C., Foussereau, C., & Choplin, L. (1993). Shear stress eects on growth and activity of Lactobacillus delbrueckii subsp. bulgaricus. Journal of Biotechnology, 29(1-2), 157175. Audet, P., Lacroix, C., & Paquin, C. (1992). Continuous fermentation of a supplemented whey permeate medium with immobilized Streptococcus salivarius ssp. Thermophilus. International Dairy Journal, 2(1), 115. Audet, P., Paquin, C., & Lacroix, C. (1998). Immobilized growing lacticacid bacteria with k-carrageenan Locust bean gum gel. Applied Microbiology and Biotechnology, 29(1), 1118. Capela, P., Hay, T. K. C., & Shah, N. P. (2006). Eect of cryoprotectants, prebiotics and microencapsulation on survival of probiotic organisms in yoghurt and freeze-dried yoghurt. Food Research International, 39(3), 203211. Chandramouli, V., Kailasapathy, K., Peiris, P., & Jones, M. (2004). An improved method of microencapsulation and its evaluation to protect Lactobacillus spp. in simulated gastric conditions. Journal of Microbiological Methods, 56(1), 2735. Conway, P. L., Gorbach, S. L., & Goldin, B. R. (1987). Survival of lactic acid bacteria in the human stomach and adhesion to intestinal cells. Journal of Dairy Science, 70(1), 112. Gardiner, G. E., OSullivan, E., Kelly, J., Auty, M. A. E., Fitzgerald, G. F., Collins, J. K., et al. (2000). Comparative survival rates of humanderived probiotic Lactobacillus paracasei and L. salivarius strains during heat treatment and spray drying. Applied and Environmental Microbiology, 66(6), 26052612. Guarner, F., & Schaafsma, G. J. (1998). Probiotics. International Journal of Food Microbiology, 39, 237238. Isolauri, E., Juntunen, M., Rautanen, T., Sillanaukee, P., & Koivula, T. (1991). A human Lactobacillus strain (Lactobacillus casei sp. strain GG) promotes recovery from acute diarrhea in children. Pediatrics, 88, 9097. Krasaekoopt, W., Bhandari, B., & Deeth, H. (2004). The inuence of coating materials on some properties of alginate beads and survivability of microencapsulated probiotic bacteria. International Dairy Journal, 14, 737743. Kurman, J. A., & Rasic, J. L. (1991). The health potential of products containing bidobacteria. In R. K. Robinson (Ed.), Therapeutic properties of fermented milks (pp. 117158). London: Elsevier Applied Food Sciences. Kwok, K. K., Groves, M. J., & Burgess, D. J. (1991). Production of 515 microns diameter alginate-polylysine microcapsules by an air-atomization technique. Pharmaceutical Research, 8(3), 341344.

Lee, D. W., Hwang, S. J., Park, J. B., & Park, H. J. (2003). Preparation and release characteristics of polymer-coated and blended alginate microspheres. Microencapsulation, 20(2), 179192. McClements, D. J. (1999). Food emulsions: Principles, practice, and techniques (2nd ed.). Florida: CRC Press. Oksanen, P. J., Salminen, S., Saxelin, M., Hamalainen, P., IhantolaVormisto, A., Muurasniemi-Isoviita, L., et al. (1990). Prevention of travellers diarrhea by Lactobacillus GG. Annals of Internal Medicine, 22, 5356. Poncelet, D., Lencki, R., Beaulieu, C., Halle, J. P., Neufeld, R. J., & Fournier, A. (1992). Production of alginate beads by emulsication/ internal gelation. Applied Microbiology and Biotechnology, 38(1), 3945. Prasad, J., McJarrow, P., & Gopal, P. (2003). Heat and osmotic stress responses of probiotic Lactobacillus rhamnosus HN001 (DR20) in relation to viability after drying. Applied and Environmental Microbiology, 69(2), 917925. Scha r-Zammaretti, P., & Ubbink, J. (2003). The cell wall of lactic acid bacteria: Surface constituents and macromolecular conformations. Biophysical Journal, 85(12), 40764092. Shahani, K. M., & Chandan, R. C. (1979). Nutritional and healthful aspects of cultured and culture-containing dairy foods. Journal of Dairy Science, 62, 16851694. Shah, N. P., & Ravula, R. R. (2000). Microencapsulation of probiotic bacteria and their survival in frozen fermented dairy desserts. Australian Journal of Dairy Technology, 55(3), 139144. Sheu, T. Y., & Marshall, R. T. (1993). Microencapsulation of lactobacilli in calcium alginate gels. Journal of Food Science, 54(3), 557561. Siitonen, S., Vapaatalo, H., Salminen, S., Gordin, A., Saxelin, M., Wikberg, R., et al. (1990). Eect of Lactobacillus GG yogurt in prevention of antibiotic associated diarrhea. Annals of Medicine, 22, 5759. Sun, W. R., & Griths, M. W. (2000). Survival of bidobacteria in yogurt and simulated gastric juice following immobilization in gellan-xanthan beads. International Journal of Food Microbiology, 61(1), 1725. Talwalkar, A., & Kailasapathy, K. (2003). Eect of microencapsulation on oxygen toxicity in probiotic bacteria. The Australian Journal of Dairy Technology, 58(1), 3639. Tharmaraj, N., & Shah, N. P. (2003). Selective enumeration of Lactobacillus delbrueckii ssp. bulgaricus, Streptococcus thermophilus, Lactobacillus acidophilus, Bidobacteria, Lactobacillus casei, Lactobacillus rhamnosus and Propionibacteria. Journal of Dairy Science, 86, 22882296. Tyle, P. (1993). Eect of size, shape and hardness of particles in suspension on oral texture and palatability. Acta Psychologica, 83, 111118. Ulmer, H. M., Ga nzle, M. G., & Vogel, R. F. (2000). Eects of high pressure on survival and metabolic activity of Lactobacillus plantarum TMW1.460. Applied and Environmental Microbiology, 66(9), 39663973.