In
fluence of
L
-cysteine, oxygen and relative humidity upon survival throughout
storage of probiotic bacteria in whey protein-based microcapsules
D. Rodrigues
a, S. Sousa
b, T. Rocha-Santos
a,c, J.P. Silva
d, J.M. Sousa Lobo
d, P. Costa
d, M.H. Amaral
d,
M.M. Pintado
b, A.M. Gomes
b, F.X. Malcata
b,1,2, A.C. Freitas
a,*aISEIT/Viseu, Instituto Piaget, Estrada do Alto do Gaio, Galifonge, P-3515-776 Lordosa, Viseu, Portugal
bCBQF/Escola Superior de Biotecnologia, Universidade Católica Portuguesa, Rua Dr. António Bernardino de Almeida, P-4200-072 Porto, Portugal cCESAM and Department of Chemistry, University of Aveiro, 3810-193 Aveiro, Portugal
dFaculdade de Farmácia, Universidade do Porto, Rua Aníbal Cunha no 164, P-4050-123 Porto, Portugal
a b s t r a c t
The survival rates of Lactobacilus acidophilus Ki, Lactobacillus paracasei L26 and Bifidobacterium animalis BB-12 were studied after whey protein microencapsulation via spray-drying, with or withoutL
-cysteine-HCl, and storage up to 6 months at 5C and 22C, with variation in relative air humidity and oxygen levels. Lb. paracasei L26 was the least susceptible to storage conditions: above 106cfu g1were recorded by 180 d at 22C, irrespective of relative humidity, and the presence/absence of oxygen andL-cysteine.
Higher relative humidity, higher temperature and longer storage periods were deleterious to survival of both B. animalis BB-12 and Lb. acidophilus Ki; the effect ofL-cysteine-HCl was dependent on the probiotic
strain. The effect of overhead oxygen was not significant upon any probiotic strain studied. Whey protein microcapsules containing L-cysteine-HCl protected probiotic cultures from simulated gastrointestinal conditions.
1. Introduction
Probiotic microorganisms can bring about beneficial effects to the health of the human host, provided they are consumed at appropriate levels and as part of a balanced diet. These microor-ganisms hold great promise for success in the food marketplace; however, a critical issue concerning use of foods as vectors for probiotic bacteria is the fastidious nature of the organisms coupled with the stressful environment of food processing. Since large viable numbers of probiotic strains are required at the moment of consumption (above 107cfu g1) for a significant role to be even-tually played in the gastrointestinal tract (Adhikari, Mustapha, Grun, & Fernando, 2000; Doleyres & Lacroix, 2005), stability throughout processing and storage are critical issues to be addressed for large-scale production. Therefore, food manufac-turers aim to increase viable cell counts, not only during biomass
production and food formulation, but also throughout downstream processing and storage prior to ingestion.
Probiotic bacteria, often belonging to the Lactobacillus and Bi fi-dobacterium genera (Weinbreck, Bodnár, & Marco, 2010), are nor-mally added to fresh and cultured dairy foods. These products are characterized by high water activity and shelf-lives of the order of a few days, requiring refrigeration, as is the typical case of yoghurt (Capela, Hay, & Shah, 2006; Gomes & Malcata, 1999; Kailasapathy, 2006; Krasaekoopt, Bhandari, & Deeth, 2003). When drier prod-ucts containing probiotic strains are tested at room temperature, dramatic losses of their viability over time severely hamper shelf-life (Ross, Desmond, Fitzgerald, & Stanton, 2005; Ubbink & Krueger, 2006). High water activity and storage temperature, as well as the presence of oxygen, apparently compromise extended storage (Anal & Singh, 2007; Teixeira, Castro, Malcata, & Kirby, 1995), thus constitute an obvious limitation for diversification of probiotic foods.
Microencapsulation of microorganisms, via tailor-made carriers composed of non-toxic materials, has frequently been used to impart protection against stressful environmental factors. This has allowed several types of food products to become effective carriers of sensitive microorganisms. In particular, delivery of active pro-biotic cells in a microencapsulated form has received increasing attention in recent years as microencapsulation provides * Corresponding author. Tel.: þ351 232910117; fax: þ351 232910193.
E-mail address:afreitas@viseu.ipiaget.org(A.C. Freitas).
1 Current affiliation: ISMAI e Instituto Superior da Maia, Avenida Carlos Oliveira
Campos, P-4475-690 Avioso S. Pedro, Portugal.
2 Current affiliation: Instituto de Tecnologia Química e Biológica, Universidade
a particularly suitable micro-environment in which the bacteria can more easily survive during processing and storage and be released at the appropriate location(s) in the gastrointestinal tract (Weinbreck et al., 2010). Experimental evidence has been put forward that microencapsulation protects probiotic strains, in liquid dairy products, against the low pH prevailing in the stomach (Anal & Singh, 2007; Kailasapathy, 2006; Krasaekoopt et al., 2003; Pimentel-González, Campos-Montiel, Lobato-Calleros, Pedroza-Islas, & Vernon-Carter, 2009). However, the performance of pro-biotic strains in terms of survival in dry food products still remains to be comprehensively tested, and conveniently rationalized.
Spray-drying is one of the oldest and most commonly used methods of encapsulation in the food industry (Picot & Lacroix, 2004) based on dehydration, which allows maintaining viability of microbial biomass, including probiotic strains belonging to Lactobacillus and Bifidobacterium genera (Goderska & Czarnecki, 2008; Simpson, Stanton, Fitzgerald, & Ross, 2005). For certain strains, and under certain processing parameters, it is considered a technology that is able to produce high rates of dry and stable powders at relatively low costs (Silva, Carvalho, Teixeira, & Gibbs, 2002), which may be applied downstream as adjuncts in several foods (Gardiner et al., 2002). Microencapsulation is promoted by dissolution or suspension of an active material in a melt or polymer solution that then becomes trapped in the dried particle. The main advantage of spray-drying is the ability to handle labile materials because of the short contact time in the dryer; in addition, the operation is economical. Through understanding of the different phenomena that occur during drying, formulation and drying conditions can be designed for optimum coating and encapsulation of probiotic strains. Hence, whey protein microencapsulation was selected as a potential technology to protect Lactobacillus and Bifidobacterium strains in this study.
Microencapsulation appears to provide anoxic regions inside the capsules, thus providing an extra protection to oxygen-sensitive strains, especially when there are significant levels of dissolved oxygen in the product (Talwalkar & Kailasapathy, 2003). Therefore,
L-cysteine-HCl appears as a good candidate for an oxygen scavenger
role, thus reducing the oxidative stress upon microencapsulated probiotic strains; furthermore, it can act as a source of amino nitrogen for such fastidious microorganisms as bifidobacteria (Dave & Shah, 1997).
The main objective of this research was to study the survival rates of various probiotic bacteria (Lactobacilus acidophilus Ki, Lactobacillus paracasei L26 and Bifidobacterium animalis BB-12) upon containment as fresh cultures in microcapsules and ascer-tain their response to exposure to different levels of air humidity and oxygen during storage, for a period up to 6 months at 22C. The microcapsules were produced via spray-drying from whey protein concentrate, with or without L-cysteine-HCl. To the best of our knowledge, such a study on how probiotics may behave in dry formulations normally held at room temperature has not been attempted. In parallel, a similar assay to that carried out at 22C was conducted to assess survival of the microencapsulated pro-biotic bacteria at 5 C storage temperature, which was com-plemented with a preliminary in vitro test of their resistance to gastrointestinal conditions.
2. Materials and methods 2.1. Inoculum preparation
The probiotic cultures tested e B. animalis BB-12 (Nu-trishÒ, Chr.-Hansen, Hørsholm, Denmark) and Lb. paracasei LAFTIÒ L26 (DELVO-PROÒ, DSM, Sydney, NSW, Australia)e were obtained as freeze-dried cultures, whereas Lb. acidophilus Ki, previously
isolated from fermented milk, was obtained from CSK (Leeu-warden, The Netherlands) as a frozen concentrate. The aforemen-tioned microorganisms were reactivated as pre-cultures in de ManeRogosaeSharpe (MRS) broth (Biokar Diagnostics, Beauvais, France), which were incubated overnight at 37C. Except for Lb. paracasei LAFTIÒL26, MRS was supplemented withfilter-sterilized 0.5 g L1L-cysteine-HCl (Fluka, Buchs, Switzerland) to lower the redox potential, and incubated in a plastic anaerobic jar using an AnaeroGen sachet (Oxoid, Cambridge, UK).
The cultures were propagated by inoculating fresh MRS media at 10% (v/v), and then incubating under appropriate conditions. The resulting cultures were centrifuged at 1789 g for 20 min, at 4C.
The centrifuged growth medium supernatant was then discarded, and the pellet was re-suspended in one tenth of its original volume of aqueous 0.85% (w/v) NaCl (Panreac, Barcelona, Spain) as an isotonic solution to re-suspend cells; salt solution or Ringer solu-tion is commonly used by several authors (Picot & Lacroix, 2004) to prepare cultures.
2.2. Microorganism microencapsulation
Each probiotic suspension was added, at 10% (v/v), to an unsterilized 5% (w/v) whey protein concentrate containing 50% (w/w) protein, WPC50(Formulab, Maia, Portugal), and delivered by
a peristaltic pump to a rotary atomizer (GEA Niro, Søborg, Denmark) coupled to a drying chamber (Arsopi, Vale da Cambra, Portugal), and spray-dried using 160 and 75C as inlet and outlet-air temperatures, respectively, to generate the intended micro-capsules. In order to obtain microcapsules with L-cysteine-HCl,
a similar procedure was used for each probiotic suspension with WPC50, but supplemented with 0.5 g L1L-cysteine-HCl.
2.3. Storage survival
After microencapsulation, microcapsules (1e1.5 g) were stored (in duplicate) in perforated petri dishes (50 mm) at 22C up to 6 months, which in turn were maintained in glass sterilized sealed flasks (0.5 L) in the presence (or absence) of oxygen (StabilOxÒ
Oxygen Absorbers, Multisorb, Buffalo, NY, USA), under various relative humidities (12, 32 and 45%) brought about by two-way gels (Humidipack, Wayzata, MN, USA). Assessment of survival of free cells of each probiotic strain during storage under similar condi-tions had been carried out previously and revealed that viable cells of Lb. acidophilus Ki, B. animalis BB-12 and Lb. paracasei L26 could not survive longer than approximately 21, 60 and 90 d of storage, respectively (data not shown).
To serve as a comparison reference, WPC50microcapsules, as
well as free cells (suspended in an isotonic solution) of each pro-biotic strain were also stored at 5C during the same timeframe, only in the presence of oxygen and without relative humidity control (i.e., under conditions mimicking regular storage in a domestic refrigerator). In all situations, sampling was made at 0, 7, 14, 21, 30, 45, 60, 90, 120, 150 and 180 d.
2.4. Microbiological analyses
For each sample, 0.6 g of microcapsules was diluted in 6 mL of a sterile aqueous solution of 0.85% (w/v) NaCl at room temperature (22C), and homogenized in a Stomacher at 230 rpm for 7 min. Viable cell numbers were enumerated, following the procedure of Miles and Misra (1938), on MRS agar in the case of Lb. paracasei L26, and on MRS agar supplemented with 0.5 g L1L-cysteine-HCl for the other two strains. Plates were incubated for 48 h at 37C, after several decimal dilutions with sterile 0.1% (m/v) peptone water, in boxes maintained under anaerobiosis with AnaeroGen sachets
(Oxoid) in the case of Lb. acidophilus Ki and B. animalis BB-12, and under regular aerobic conditions in the case of Lb. paracasei L26. Plate count agar, incubated aerobically at 37C for 48 h, was used in parallel to monitor putative cross-contamination arising from laboratory manipulation, but the viable counts found were always negligible (data not shown).
2.5. Microscopical observations
To ascertain the morphological features of microcapsules, samples taken at various storage times were placed on a double sided adhesive carbon tape (NEM tape, from Nisshin, Tokyo, Japan), and observed with a low-vacuum scanning electron microscope (JSM-5600LV, from JEOL, Tokyo, Japan) operated under 10 Pa and at an accelerating voltage of 15e20 kV.
2.6. Gastrointestinal simulation
Free cells of each probiotic strain, as well as WPC50
microcap-sules containing either of the three strains, were exposed to simulated gastrointestinal conditions after 15, 60 and 120 d of storage at 5C, according to the procedure described byMadureira, Amorim, Pintado, Gomes, and Malcata (2011), with slight modi fi-cations. The typical conditions prevailing in the mouth, oesopha-gusestomach, duodenum and ileum were sequentially applied as follows: the mouth micro-environment (including mastication) was paralleled following the method ofHöld, de Boer, Zuidema, and Maes (2000)andChoi, Chung, Lee, Shin, and Sung (2007), using a synthetic saliva solution prepared with 100 U mL1
a
-amylase in 1 mM CaCl2, and pH was adjusted to 6 using 1M NaHCO3; thissimulated saliva was added at a rate of 0.6 mL min1. For the oesophagusestomach step, 25 mg mL1 pepsin was prepared in
0.1MHCl; this solution was added in equal-sized aliquots during
the gastric phase, at a rate of 0.05 mL per mL or g of sample and pH was gradually decreased to pH 2 using 1MHCl (Aura et al., 2005).
Duodenum conditions were simulated with 2 g L1pancreatin and 12 g L1 bile salts dissolved in 0.1MNaHCO3; this solution was
added at a rate of 0.25 mL per mL or g of sample (Laurent, Besançon, & Caporiccio, 2007). The increase of pH that takes place in the ileum was simulated by gradually adding 0.1MNaHCO3.
All enzyme solutions were freshly prepared for each experiment andfilter-sterilized through a 0.22
m
m-membranefilter (Millipore, Billerica, MA, USA). A rotary water bath at 37 C was used to simulate the temperature and peristaltic movements that prevail during human gastrointestinal transit.2.7. Statistical analyses
For each probiotic bacterium, a two-way analysis of variance (ANOVA) was carried out with SigmaStatÔ (Systat Software, Chi-cago, IL, USA), to assess whether each factor (viz. the presence or absence ofL-cysteine-HCl in the microcapsule composition, as well
as presence or absence of oxygen, and several relative humidities during storage) was a significant source of variation throughout storage time, at a significance level of P ¼ 0.05. Two-way ANOVA was also applied to yield data presented inTable 1. ANOVA is valid provided that the experimental errors are independently and nor-mally distributed, and possess a constant variance, hence, the original data had to be transformed via
l
-power forms, so as to achieve normality (Box, Hunter, & Hunter, 1978; Freitas & Malcata, 1996; Freitas, Sousa, & Malcata, 1995). The Holm-Sidak method was finally used for pairwise comparisons, also at the P ¼ 0.05 level of significance.3. Results
3.1. Storage at 22C
The microencapsulation yield (defined as the mass of micro-capsules formed normalized by the amount of protein prevailing right before microencapsulation) was dependent on the presence or absence ofL-cysteine-HCl (Table 1). However, yield values did not differ substantially for microcapsules containing L-cysteine-HCl
(P> 0.05). Values of 39e46% were recorded for the three probiotic strains under assessment, but such values dropped to 23e31% for B. animalis BB-12 or Lb. paracasei L26 microcapsules without
L-cysteine-HCl. Despite the differences observed, probiotic strains,
as well as the presence or absence ofL-cysteine-HCl, did not appear
as significant factors (P > 0.05) for microencapsulation yield. A higher density of viable cells was observed when microencapsu-lation included L-cysteine-HCl, especially in the case of Lb.
acid-ophilus Ki; however, such an effect was negligible for Lb. paracasei L26, since 109cfu g1was observed in both types of microcapsules. For Lb. acidophilus Ki and B. animalis BB-12, absence ofL
-cysteine-HCl led to substantial losses in viability in comparison to their initial viable numbers prior to microencapsulation (Table 1).
The evolution in time of viable cells of each probiotic strain, stored under various conditions, is displayed inFig. 1. Lb. paracasei L26 appeared to be the strain least susceptible to the parameters under scrutiny; highest numbers of viable cells (i.e., above 106cfu g1) were recorded by 180 d of storage at 22C, irrespective of relative humidity, the presence or absence of oxygen or
L-cysteine-HCL in the microcapsules. Besides storage time, only
relative humidity appeared to be significant (P < 0.05) towards Lb. paracasei L26 viability.
As expected, higher relative humidity and longer storage periods also led to lower numbers of viable cells; such decreases were more significant in the case of B. animalis BB-12 and Lb. acidophilus Ki, however, the variation in viability as storage time elapsed was not uniform among probiotic strains. In the case of Lb. paracasei L26, a slower overall decrease over time was observed (with a maximum reduction of 3 log cycles at 45% relative humidity, RH), but such decrease was faster in thefirst 60 d of storage fol-lowed by a much smoother pattern up to 180 d; Lb. acidophilus Ki
Table 1
Variation in the number of viable cells (average standard deviation) of Lactobacillus acidophilus Ki, Bifidobacterium animalis BB-12 and Lactobacillus paracasei L26 prior to, and after microencapsulation in WPC50by spray-drying, and the corresponding microcapsule (MC) yield.
Probiotic strain WPC50microcapsules Viable cells prior to
microencapsulation (cfu mL1)
Viable cells after
microencapsulation (cfu g1)
MC yield (%, w/w)a
Lb. acidophilus Ki WithL-cysteine-HCl (0.5 g L1) (8.25 0.35) 108 (5.28 0.35) 108 41.7
WithoutL-cysteine-HCl (6.50 0.71) 108 (6.50 1.41) 107 44.0
B. animalis BB-12Ò WithL-cysteine-HCl (0.5 g L1) (9.75 0.35) 108 (1.48 0.03) 108 39.0
WithoutL-cysteine-HCl (1.83 0.18) 109 (1.28 0.03) 108 22.7
Lb. paracasei L26 WithL-cysteine-HCl (0.5 g L1) (4.13 0.38) 109 (5.88 0.63) 109 45.7
WithoutL-cysteine-HCl (6.38 1.49) 109 (7.50 2.80) 109 30.7 aBased on the theoretical maximum weight of microcapsules produced from 6 L of 5% (w/v) WPC
and B. animalis BB-12 underwent decreases in their viable cells, with the former exhibiting a higher loss of viability rate, revealing it to be the most susceptible, with reductions of 4e5 log cycles by 180 d of storage at 45% RH.Simpson et al. (2005)reported similar magnitude of viable cell reduction for B. animalis BB-12 in skimmed milk powders obtained by spray-drying, but after 90 days of storage at 25C.
The presence of oxygen throughout storage did not have significant effect (P > 0.05) for any of the probiotic strains studied; only a slight negative effect was observed for Lb. acidophilus Ki microencapsulated with L-cysteine-HCl by 120 d of storage at
22 C. Conversely, the presence or absence of L-cysteine-HCl proved to be a significant effect (P < 0.05) for survival of micro-encapsulated Lb. acidophilus Ki and B. animalis BB-12. Surprisingly, inclusion ofL-cysteine-HCl in microcapsule formulation negatively
affected (P < 0.05) the viability of B. animalis BB-12 throughout time, for all relative humidities tested. On the other hand, the presence ofL-cysteine-HCl turned to be, a protection factor upon
L. acidophilus Ki, as made apparent by the lower decreases in its viable cells.
To assess the morphological features of the microcapsules, scanning electron microphotographs are displayed in Fig. 2. No effects of either relative humidity or storage time upon micro-capsule structure and dimension can be detected. In fact, the microcapsules did not lose their initial morphology, and in particular maintained their original shape and no swelling was observed at higher humidities. It should be emphasized that the microcapsules were characterized by irregular geometrical
shapes similar to spheres from which air had been removed, with variable dimensions (5e50
m
m), owing to the nature of the spray-drying process. Similar shapes were reported byLian, Hsiao, and Chou (2002)for spray-dried starch microcapsules. The presence or absence of L-cysteine-HCl or oxygen was not reflected uponvisual differences in morphology (data not shown) of the microcapsules.
3.2. Storage at 5C
Microcapsules containing each probiotic strain, with or without
L-cysteine-HCl, were also stored for up 6 months at 5C, but only in
the presence of oxygen and without relative humidity control in an attempt to assess survival under refrigeration conditions; the resulting data are depicted inFig. 3. As expected, this low storage temperature proved beneficial to all microencapsulated strains, in particular, to Lactobacillus strains (Fig. 3a,c). The lowest decrease in microencapsulated viable cell numbers was observed for Lb. para-casei L26, with some degree of variability especially within thefirst 60 d, but the numbers of viable cells were always above 109cfu g1. As happened with storage of Lb. paracasei L26 at 22C,L-cysteine did not play a role in the survival of viable cells in the microcap-sules. The protective effect by whey protein microencapsulation, in itself, is particularly evident by 60 d of storage at 5C; the number of viable cells in free form decreased continuously throughout storage time, with a sharper decrease in the last 30 d against a relatively stable protective effect for microencapsulated cells. Fig. 1. Variation in log number of viable cells (average standard deviation) of: a) Lactobacillus acidophilus Ki, b) Bifidobacterium animalis BB-12 and c) Lactobacillus paracasei L26, microencapsulated in 50% whey protein concentrate, WPC50, with storage time at 22C under relative humidities of: (i) 12%; (ii) 32%; and (iii) 45%, withL-cysteine-HCl in the absence (B) or presence (C) of oxygen, and withoutL-cysteine-HCl in the absence (,) or presence (-) of oxygen.
Survival rates in microencapsulated form of both Lb. acidophilus Ki and B. animalis BB-12 were also higher at 5C than at 22C as storage time elapsed; by 180 d, viable numbers above 106cfu g1 were observed for B. animalis BB-12, in the presence or absence of
L-cysteine. Once again, the effect of L-cysteine was significant
(P< 0.05), and responsible for higher rates of survival of Lb. acid-ophilus Ki: by 180 d, 107cfu g1was recorded forL
-cysteine-con-taining microcapsules, whereas absence of L-cysteine-HCl
promoted a sharper decrease of viable cell numbers, especially after 60 d to ca. 105 cfu g1. The protective effect promoted by whey protein microencapsulation is remarkable in the case of Lb. acid-ophilus Ki, since those cells could not survive longer than 60 d when in free form; for B. animalis BB-12 the onset of this protective effect is less visible. Lower values were reported by Goderska and Czarnecki (2008)for Lb. acidophilus and Bifidobacterium bifidum after 120 d of storage at 4C in starch capsules obtained by spray-drying.
3.3. Gastrointestinal transit
Free cells and WPC50microcapsules, containing either of the
three strains and previously stored at typical refrigeration condi-tions for 15, 60 and 120 d, were exposed to simulated gastroin-testinal tract. (Due to constraints pertaining to the amounts of sample available, it was not possible to carry out assays of micro-capsules withoutL-cysteine-HCl in the case of Lb. acidophilus Ki and
B. animalis BB-12 stored at 5C as well on microcapsules stored at 22 C). The resulting data are shown inTable 2. Higher rates of survival were exhibited by the three probiotic strains, when microencapsulated in WPC50 containing L-cysteine-HCl, than by their free cell counterparts after exposure to the simulated gastrointestinal conditions. Specifically, Lb. acidophilus Ki could not survive in free cell form upon passage through simulated duodenum and ileum, but its resistance improved dramatically when microencapsulated; storage at 5 C did not decrease Lb. Fig. 2. Microphotographs of microcapsules containing B. animalis BB-12 before storage (a) and by 90 d of storage under relative humidities of (b) 12%, (c) 32%, and (d) 45%.
Fig. 3. Variation in log number of viable cells (average standard deviation) of: a) Lactobacillus acidophilus Ki, b) Bifidobacterium animalis BB-12 and c) Lactobacillus paracasei L26, in free cell form (:) or microencapsulated in 50% whey protein concentrate, WPC50, with (C) or without (-)L-cysteine-HCl, and storage time at 5C.
acidophilus Ki resistance to low pH and bile salt, irrespective of storage time, except when storage took place for 120 d, which led to a faster decline under ileum conditions.
In the case of B. animalis BB-12 and Lb. paracasei L26, microen-capsulation did improve their resistance to the simulated gastro-intestinal tract. Decreases of 2 log cycles and above were indeed recorded for free cells, whereas a maximum of only 1 log cycle occurred for the microencapsulated counterparts. The assay carried out with Lb. paracasei L26 microencapsulated with or without
L-cysteine-HCl, and previously stored for 120 d, did not show any significant protective effect (P > 0.05) by L-cysteine-HCl during
exposure to the gastrointestinal environment, consistent with the survival results obtained for storage at 5 or 22C.
The effect of storage conditions on the survival of probiotic strains under gastrointestinal conditions caused slight reductions of viable cells in the microcapsules as storage time elapsed in some cases, but mostly were not significant. A more pronounced effect was observed for Lb. acidophilus Ki and B. animalis BB-12. 4. Discussion
Probiotic microorganisms can bring about health benefits to the host only upon consumption to adequate viable levels in order to influence the profile of the intestinal microflora in a sustained manner (Lourens-Hattingh & Viljoen, 2001); usually, the recom-mended number of viable cells by the time of consumption is at least 107cfu mL1(or cfu g1; equivalent to 109cfu per 100 g or
100 mL portions), hence, the spray-drying microencapsulation process itself, and the subsequent storage at 22 C appeared feasible in attempts tofind a dry vector to deliver probiotic bacteria at room temperature throughout extended storage periods. The actual performance in terms of viability appears to be strain-dependent, as well as greatly influenced by the presence of
L-cysteine-HCl, and independent of the presence of oxygen, but
affected by the relative humidity prevailing during storage. Commercial powders of freeze-dried probiotics have well above 1010cfu g1, therefore the viable counts in this study appear low and its commercial interest is debatable. There are two reasons for the low numbers. First, no attempt was made to have the highest density possible in the cell suspension before spray-drying. A ten-fold improvement could be done at that level. Secondly, viability losses during spray-drying led to only a partial recovery of the powder, resulting in a yield of viable biomass lower than 10%. Improvements to the spray-drying process may allow technology to be applied to all strains besides those that better survive the high temperatures used in the drying process.
The effect of overhead oxygen was not significant (P > 0.05) for any probiotic strain studied; these results seem to contradict the usual heuristic rule, that the presence of oxygen compromises cell survival during extended storage periods (Anal & Singh, 2007; Teixeira et al., 1995) probably due to detrimental oxidative changes of the cell membrane (Hsiao, Lian, & Chou, 2004). Although oxygen tolerance varies among Bifidobacterium species, it is still unclear how oxygen sensitivity plays a role in determining survival following spray-drying (Simpson et al., 2005; Talwalkar, Kailasapathy, Peiris, & Arumugaswamy, 2001). It is known that proteins are good barriers to oxygen and moisture (Gennadios, 2002); this may account for the protection provided by microcap-sules made from WPC50 that contain 50% protein. The lactose
content of WPC50can also limit the diffusion of substances through
the whey protein wall and thus lead to high microencapsulation efficiency values.
As expected, higher relative humidities showed an unfavourable effect upon the three probiotic bacteria; similar results were recently reported by Heidebach, Forst, and Kulozik (2010) for
T able 2 Variation in the log number of viable cells aof Lactob acillus acidophilus Ki, Bi fi dobacterium animalis BB-1 2 and Lact obacillus par acasei L26, fr ee or micro encapsulate d in 50% whe y pro tein concentra te, WPC 50 , with L -cyst eine, and st ore d for v arious periods at 5 C, under simulat ed gastr oint estinal conditions. Gastrointestinal compartment b Samplin g (min) Lb. acidophilus Ki (log cfu g 1) B. animalis BB-12 (log cfu g 1) Lb. paracasei L26 (log cfu g 1) Free cells Microencapsulated cells Free cells Microencapsulated cells Free cells Microencapsulated cells 15 d 60 d 1 20 d 15 d 60 d 120 d 15 d 60 d 120 d 120 d c Mouth (pH 6, 37 C, 200 rpm) 0 9.7 0.4a 8.12 0.03ab 8.04 0.17b 7 .96 0 .04b 8.9 0.2ab 8.24 0.01a 7.96 0.03b 7.59 0.04b 9.01 0.05b 10.3 0.1a 9.75 0.04b 9.40 0.09b 9.40 0.05 2 9.4 0.1a 8.02 0.06ab 8.5 0.1b 7 .88 0 .03b 8.9 0.2ab 8.2 0.4a 7.95 0.01b 7.67 0.03b 9.14 0.05b 10.2 0.1a 9.77 0.08b 9.35 0.07b 9.5 0.1 Oesophagus e stomach ( D pH 6 to 2, 37 C, 130 rpm) 10 9.4 0.2a 8.07 0.01ab 8.53 0.01bc 7 .73 0 .14b 8.8 0.2b 8.3 0.4a 7.9 0.1b 7.66 0.03b 9.17 0.09b 10.2 0.2a 9.7 0.2b 9.46 0.05b 9.34 0.05 20 9.4 0.2a 8.0 0.2ab 8.62 0.03bc 7 .85 0 .03b 8.8 0.2b 8.3 0.3a 8.0 0.2b 7.58 0.02b 9.17 0.06b 10.1 0.2a 9.85 0.01ab 9.44 0.02b 9.42 0.02 30 9.4 0.2a 7.9 0.2ab 8.26 0.01b 7 .79 0 .01b 8.7 0.4b 8.34 0.08a 8.0 0.1b 7.45 0.01b 9.24 0.08b 10.0 0.2a 9.69 0.02ab 9.4 0.1b 9.42 0.07 40 8.9 0.5a 7.9 0.1ab 8.55 0.02bc 7 .64 0 .08b 8.7 0.2ab 8.2 0.4a 7.7 0.2b 7.4 0.1b 8.9 0.1b 10.1 0.2a 9.5 0.1b 9.3 0.1b 9.17 0.04 50 8 2a 7.86 0.02ab 8.6 0.1ab 7 .72 0 .06b 8.7 0.3ab 8.2 0.3a 8.05 0.01b 7.4 0.4b 8.53 0.05b 10.0 0.4a 9.49 0.08b 9.42 0.06b 9.32 0.03 60 6 2a 7.87 0.02ab 8.58 0.02b 7 .64 0 .02b 8.7 0.2ab 8.2 0.2a 7.8 0.2b 7.4 0.1b 8.53 0.05b 10.1 0.2a 9.5 0.2b 9.4 0.2b 9.33 0.03 Duodenum ( D pH 2 to 5, 37 C, 45 rpm) 90 ND 7.0 0.3a 7.6 0.1a 7.1 0 .3a 8.2 0.2a 7.2 0.1a 6.7 0.2a 6.43 0.05a 6.53 0.01a 9.2 0.1a 8.33 0.04b 8.7 0.2a 8.8 0.1 Ileum ( D pH 5 to 6.5, 37 C, 45 rpm) 1 30 ND 7.1 0.1a 7.52 0.09a 6.1 0 .4a 6.6 0.5a 7.59 0.08a 6.6 0.2a 6.7 0.2a 6.46 0.03a 8.98 0.02a 8.9 0.2a 8.68 0.03a 8.82 0.03 a Values are the average standard deviation. Different letters in the same row and for each probiotic strain indicate signi fi cant differences (P < 0.05) between free cells and microencapsulated cells. ND, not detected. b Typical conditions prevailing in each gastrointestinal compartment: pH or variation of pH ( D pH), temperature and agitation values (rpm) used in each compartment are indicated. c WPC 50 microcapsules without L -cysteine.
storage of Bifidobacterium BB-12 at 25C in freeze-dried
casein-based microcapsules. In the present study, storage at 22C and under 12% RH promoted the highest survival rates throughout storage: i) 180 d for Lb. paracasei L26, with viable numbers above 107cfu g1, irrespective of presence ofL-cysteine-HCl; ii) 150 d for
B. animalis BB-12, with viable numbers also above 107 cfu g1 withoutL-cysteine-HCl, or above 106cfu g1withL-cysteine-HCl; and iii) 120e150 d for Lb. acidophilus Ki, with viable numbers above 106cfu g1regardless of presence or absence ofL-cysteine-HCl. All
these values demonstrate a significant protective effect by whey protein microencapsulation relative to non-encapsulated cells, since free cells of Lb. acidophilus Ki, B. animalis BB-12 and Lb. par-acasei L26 could survive for no longer than 21, 60 and 90 d of storage, respectively.
Storage at high relative humidities increases the water content of the microcapsules, which may be detrimental to bacterial survival due to acceleration of oxidation processes (Teixeira et al., 1995); 7e11% RH has indeed been indicated as optimal values for bacteria survival (Castro, Teixeira, & Kirby, 1995). However, it is interesting to note that cell inactivation in the microencapsulated form does not follow simplefirst order kinetics (as reported byHigl et al., 2007, for glassy sugar matrices as microencapsulation material), but rather a two-stage behaviour (seeFig. 1), especially at higher humidities. A higher inactivation rate seems to prevail in the first 50 d and a lower inactivation rate afterwards, for both Lb. paracasei L26 and B. animalis BB-12. The existence of a critical range of water content where the survival rate decreases drastically may partially account for this observation.
The effect ofL-cysteine-HCl was dependent on the probiotic
strain but consistent at both storage temperatures for a given strain. No effect was actually detected for Lb. paracasei L26 survival, probably because lactobacilli are often facultative anaerobes (Talwalkar & Kailasapathy, 2003); a protective effect was observed for Lb. acidophilus Ki survival, whereas a negative effect was recorded in the case of B. animalis BB-12, especially at 22 C. The results pertaining to this bifidobacteria strain were somewhat unexpected since several authors have reported that milk supplemented with amino acids (e.g., cysteine) improved survival of bifidobacteria (Dave & Shah, 1997, 1998; Gomes, Vieira, & Malcata, 1998; Ravula & Shah, 1998); according toGüler-Akin and Akin (2007), specific addition of cysteine to bio-yogurts led to higher counts of B. bifidum, Lb. acidophilus Ki and Lb. paracasei L26 throughout storage for 14 d. Maybe an unfavourable inter-action of that amino acid with the strain in microencapsulated form, or with the microencapsulation process itself, played a role since the data available in the literature pertain exclusively to free cells.
Storage at 5 C produced a lower decrease in viable cell numbers throughout the 180 d-period of all probiotic strains but with a decreasing pattern somewhat resembling that obtained at 22 C. Higher rates of survival of the three microencapsulated probiotic strains in the presence ofL-cysteine-HCl were recorded,
as compared with the free cells, after passage through the simu-lated gastrointestinal tract; this clearly demonstrated its protective effect, chiefly against acids and bile salts. In general, the storage conditions did not significantly affect the stability throughout the simulated gastrointestinal tract, since the proportion of reduction from mouth to ileum was almost the same for the three storage times considered. Due to their buffering capacity, proteinaceous matrices have been claimed to convey a protective effect, espe-cially under low pHegastric conditions (Heidebach, Forst, & Kulozik, 2009); a few studies have further corroborated the potential of microencapsulation to improve probiotic survival during gastrointestinal transit (Anal & Singh, 2007; Champagne & Fustier, 2007).
5. Conclusions
Microencapsulation allowed dry microcapsules to be employed as vectors for probiotic strains that are intended for storage under regular room conditions. The exact benefits of WPC50-based
microencapsulation depend on the specific strain and the envi-ronmental conditions provided. Lb. paracasei L26 was the least susceptible to storage conditions whereas higher relative humidity, temperature and longer storage periods were harmful to B. animalis BB-12 and Lb. acidophilus Ki. The impact of L-cysteine-HCl was a strain-specific trait, yet overhead oxygen did not significantly affect any probiotic strain studied. Besides permitting expansion of the range of probiotic foods to those characterized by lower water activity, the need for a cold storage network may essentially be waived with obvious benefits for economic feasibility of the overall process.
Acknowledgements
The authors gratefully acknowledge Formulab, Multisorb, DSM and Chr. Hansen for providing the whey protein concentrate, oxygen absorbers and probiotic strains, respectively. This work was funded by FEDER under the Operational Program for Competi-tiveness Factorse COMPETE and by National funds via FCT - Fun-dação para a Ciência e a Tecnologia within the framework of project PROBIOCAPSe references PTDC/AGR-ALI/71051/2006 and FCOMP-01-0124-FEDER-008792.
References
Adhikari, K., Mustapha, A., Grun, I. U., & Fernando, L. (2000). Viability of micro-encapsulated bifidobacteria in set yogurt during refrigerated storage. Journal of Dairy Science, 83, 1946e1951.
Anal, A. K., & Singh, H. (2007). Recent advances in microencapsulation of probiotics for industrial applications and targeted delivery. Trends in Food Science and Technology, 18, 240e251.
Aura, A. M., Martin-Lopez, P., O0Leary, K. A., Williamson, G., Oksman-Caldentey, K. M.,
Poutanen, K., et al. (2005). In vitro metabolism of anthocyanins by human gut microflora. European Journal of Nutrition, 44, 133e142.
Box, G. E. P., Hunter, W. G., & Hunter, J. S. (1978). Statistics for experimenters. New York, NY, USA: Wiley.
Capela, P., Hay, T. K. C., & Shah, N. P. (2006). Effect of cryoprotectants, prebiotics and microencapsulation on survival of probiotic organisms in yoghurt and freeze-dried yoghurt. Food Research International, 39, 203e211.
Castro, H. P., Teixeira, P. M., & Kirby, R. (1995). Storage of lyophilized cultures of Lactobacillus bulgaricus under different relative humidities and atmospheres. Applied Microbiology and Biotechnology, 44, 172e176.
Champagne, C. P., & Fustier, P. (2007). Microencapsulation for the improved delivery of bioactive compounds into foods. Current Opinion in Biotechnology, 18, 184e190.
Choi, S. Y., Chung, M. J., Lee, S., Shin, J. H., & Sung, N. J. (2007). N-Nitrosamine inhibition by strawberry, garlic, kale, and the effects of nitrite-scavenging and N-nitrosamine formation by functional compounds in strawberry and garlic. Food Control, 18, 485e491.
Dave, R. I., & Shah, N. P. (1997). Effect of cysteine on the viability of yoghurt and probiotic bacteria in yoghurts made with commercial starter cultures. Inter-national Dairy Journal, 7, 537e545.
Dave, R. I., & Shah, N. P. (1998). Ingredient supplementation effects on viability of probiotic bacteria in yogurt. Journal of Dairy Science, 81, 804e2816.
Doleyres, Y., & Lacroix, C. (2005). Technologies with free and immobilised cells for probiotic bifidobacteria production and protection. International Dairy Journal, 15, 973e988.
Freitas, A. C., & Malcata, F. X. (1996). Influence of milk type, coagulant, salting procedure and ripening time on thefinal characteristics of Picante cheese. International Dairy Journal, 6, 1099e1116.
Freitas, A. C., Sousa, M. J., & Malcata, F. X. (1995). Effect of ripening time and the combination of ewe and goat milk on the microflora of Picante cheese. Italian Journal of Food Science, 7, 361e377.
Gardiner, G. E., Bouchier, P., O’Sullivan, E., Kelly, J., Collins, J. K., Fitzgerald, G., et al. (2002). A spray-dried culture for probiotic Cheddar cheese manufacture. International Dairy Journal, 12, 749e756.
Gennadios, A. (2002). Protein-basedfilms and coatings. Boca Raton, FL, USA: CRC Press.
Goderska, K., & Czarnecki, Z. (2008). Influence of microencapsulation and spray drying on the viability of Lactobacillus and Bifidobacterium strains. Polish Journal of Microbiology, 57, 135e140.
Gomes, A. M. P., & Malcata, F. X. (1999). Bifidobacterium spp. and Lactobacillus acidophilus: biological, biochemical, technological and therapeutical properties relevant for use as probiotics. Trends in Food Science and Technology, 10, 139e157.
Gomes, A. M. P., Vieira, M. M., & Malcata, F. X. (1998). Survival of probiotic microbial strains in a cheese matrix during ripening: simulation of rates of salt diffusion and microorganism survival. Journal of Food Engineering, 36, 281e301. Güler-Akin, M. B., & Akin, M. S. (2007). Effects of cysteine and different incubation
temperatures on the microflora, chemical composition and sensory character-istics of bio-yogurt made from goat’s milk. Food Chemistry, 100, 788e793. Heidebach, T., Forst, P., & Kulozik, U. (2009). Microencapsulation of probiotic cells by
means of rennet-gelation of milk proteins. Food Hydrocolloids, 23, 1670e1677. Heidebach, T., Forst, P., & Kulozik, U. (2010). Influence of casein-based
microen-capsulation on freeze-drying and storage of probiotic cells. Journal of Food Engineering, 98, 309e316.
Higl, B., Kurtmann, L., Carlsen, Ch. U., Rathjen, J., Först, P., Skibsted, L. H., et al. (2007). Impact of water activity, temperature and physical state on the storage stability of Lactobacillus paracasei ssp. paracasei freeze-dried in a lactose matrix. Biotechnology Progress, 23, 794e800.
Höld, K. M., de Boer, D., Zuidema, J., & Maes, R. A. A. (2000). Saliva as an analytical tool in toxicology. International Journal of Drug Testing, 1, 1e33.
Hsiao, H. C., Lian, W. C., & Chou, C. C. (2004). Effect of packaging conditions and temperature on viability of microencapsulated Bifidobacteria during storage. Journal of the Science of Food and Agriculture, 84, 134e139.
Kailasapathy, K. (2006). Survival of free and encapsulated probiotic bacteria and their effect on the sensory properties of yoghurt. LWT-Food Science and Tech-nology, 39, 1221e1227.
Krasaekoopt, W., Bhandari, B., & Deeth, H. (2003). Evaluation of encapsulation techniques of probiotics for yoghurt. International Dairy Journal, 13, 3e13. Laurent, C., Besançon, P., & Caporiccio, B. (2007). Flavonoids from a grape seed
extract interact with digestive secretions and intestinal cells as assessed in an in vitro digestion/Caco-2 cell culture model. Food Chemistry, 100, 1704e1712.
Lian, W. C., Hsiao, H. C., & Chou, C. C. (2002). Survival of bifidobacteria after spray-drying. International Journal of Food Microbiology, 74, 79e86.
Lourens-Hattingh, A., & Viljoen, B. (2001). Yogurt as probiotic carrier food. Inter-national Dairy Journal, 11, 1e17.
Madureira, A. R., Amorim, M., Pintado, M. E., Gomes, A. M. P., & Malcata, F. X. (2011). Protective effect of whey cheese upon probiotic strains exposed to simulated gastrointestinal conditions. Food Research International, 44, 465e470. Miles, O., & Misra, S. (1938). The estimation of the bactericidal power of the blood.
Journal of Hygiene, 38, 732e749.
Picot, A., & Lacroix, C. (2004). Encapsulation of bifidobacteria in whey protein-based microcapsules and survival in simulated gastrointestinal conditions and in yoghurt. International Dairy Journal, 14, 501e515.
Pimentel-González, D. J., Campos-Montiel, R. G., Lobato-Calleros, C., Pedroza-Islas, R., & Vernon-Carter, E. J. (2009). Encapsulation of Lactobacillus rhamnosus in double emulsions formulated with sweet whey as emulsifier and survival in simulated gastrointestinal conditions. Food Research International, 42, 292e297. Ravula, R. R., & Shah, N. P. (1998). Effect of acid casein hydrolysate and cysteine on the viability of yoghurt and probiotic bacteria in fermented frozen dairy desserts. Australian Journal of Dairy Technology, 53, 175e179.
Ross, R. P., Desmond, C., Fitzgerald, G. F., & Stanton, C. (2005). Overcoming the technological hurdles in the development of probiotic foods. Journal of Applied Microbiology, 98, 1410e1417.
Silva, J., Carvalho, A. S., Teixeira, P., & Gibbs, P. A. (2002). Bacteriocin production by spray-dried lactic acid bacteria. Letters in Applied Microbiology, 34, 77e81. Simpson, P. J., Stanton, C., Fitzgerald, G. F., & Ross, R. P. (2005). Intrinsic tolerance of
Bifidobacterium species to heat and oxygen and survival following spray drying and storage. Journal of Applied Microbiology, 99, 493e501.
Talwalkar, A., & Kailasapathy, K. (2003). Effect of microencapsulation on oxygen toxicity in probiotic bacteria. Australian Journal of Dairy Technology, 58, 36e39. Talwalkar, A., Kailasapathy, K., Peiris, P., & Arumugaswamy, R. (2001). Application of RBGRe a simple way for screening of oxygen tolerance in probiotic bacteria. International Journal of Food Microbiology, 71, 245e248.
Teixeira, P. C., Castro, M. H., Malcata, F. X., & Kirby, R. M. (1995). Survival of Lacto-bacillus delbrueckii ssp. bulgaricus following spray-drying. Journal of Dairy Science, 78, 1025e1031.
Ubbink, J., & Krueger, J. (2006). Physical approaches for the delivery of active ingredients in foods. Trends in Food Science & Technology, 17, 244e254. Weinbreck, F., Bodnár, I., & Marco, M. L. (2010). Can encapsulation lengthen the
shelf-life of probiotic bacteria in dry products? International Journal of Food Microbiology, 136, 364e367.
