Importance of intact secondary protein structures of cell envelopes and glass transition temperature of the stabilization matrix on the storage stability of probiotics

Food Research International 123 (2019) 198–207

Contents lists available at ScienceDirect

Food Research International journal homepage: www.elsevier.com/locate/foodres

Importance of intact secondary protein structures of cell envelopes and glass transition temperature of the stabilization matrix on the storage stability of probiotics

T

⁎

Arup Naga, , Mark Waterlandb, Patrick Janssenc, Rachel Andersona,d, Harjinder Singha a

Riddet Institute, Massey University, Private Bag 11 222, Palmerston North 4442, New Zealand Institute of Fundamental Sciences, Massey University, Private Bag 11 222, Palmerston North 4442, New Zealand c Massey Institute of Food Science and Technology, Massey University, Private Bag 11 222, Palmerston North 4442, New Zealand d Food Nutrition & Health Team, AgResearch Grasslands, Private Bag 11 008, Palmerston North 4442, New Zealand b

A R T I C LE I N FO

A B S T R A C T

Keywords: Lactobacillus reuteri Probiotic Stabilization Ambient temperature storage FTIR study Glass transition Membrane structural integrity Scanning electron microscopy Shelf stability

Lactobacillus reuteri LR6 cells were stabilized using a novel combination of wet granulation and fluidized-beddrying techniques. The stabilized cells were stored at 37 °C and at two water activity (aw) levels (0.11 & 0.30). Superior storage stability was recorded in the lower aw environment, supported by a stronger glassy matrix when skim milk powder was used as the excipient. The initial viable cell populations of the samples stabilized in different matrices ranged from 8.3 to 9.1 log CFU/g. At the end of the storage period, the viable cell populations were reduced to 6.7 to 7.3 log CFU/g at aw 0.11 and to 6.1 to 6.6 CFU/g when the aw was maintained at 0.30. Fourier transform infrared spectroscopic examination of the cell envelopes revealed substantial dissimilarities between samples at the beginning and at the end of the storage period, which indicated alteration in the secondary protein structures of the cell envelope and also correlated well with the loss in cell viability. In milkpowder-based matrices, adjusting the aw to 0.30 resulted in a weaker or no glassy state whereas the same matrices had a high glass transition temperature at aw 0.11. This strong glassy matrix and low aw combination was found to enhance the bacterial stability at the storage temperature of 37 °C. Scanning electron microscopy revealed the formation of corrugated surfaces and blister-type deformations on the cell envelopes during the stabilization process.

1. Introduction The international scientific association for probiotics and prebiotics has recently published a consensus statement for redefining the term ‘probiotics’ which unlike the earlier definition (FAO/WHO, 2001) should ideally be more specific in terms of the strain validation for efficacy and the amount of live cultures present per serving of the product containing such probiotics (Hill et al., 2014). The refrigerated or frozen storage of probiotic bacteria has been found effective in maintaining their viability and functionality (Batista et al., 2015; Nandelman et al., 2017). However, these forms of storage and the subsequent handling of the bacteria involve high costs, together with an enhanced risk of intermittent thawing. A suitable stabilization technique using either a simple drying method (Miao et al., 2008) or complex microencapsulation (Dianawati & Shah, 2011) can improve the storage stability of probiotics in the temperature range from 20 to 25 °C. However, higher relative humidity, higher temperature and ⁎

longer storage periods were found deleterious to the survival of probiotics (Rodrigues et al., 2011). Fluidized bed drying is an established method for drying granulated solids and potentially could be used with a moderate and controlled drying temperature of between 40 and 70 °C to stabilize heat-sensitive probiotics. The use of this technique in the stabilization of probiotics has not been explored extensively and there are only a few studies in which a fluidized bed has been used as a spray coater (Stummer et al., 2012). The major reason for the reduction in cell viability in a fluidized bed drier (FBD) is the osmotic shock caused to the cells when excipients with low aw, such as milk powders, maltodextrin and starch, come into contact with the wet cells. However, the advantage of fluidized bed drying is the granulation achieved by the interaction between the cell bodies and the excipient, which creates a protective environment against oxidative stress during drying and storage (Santivarangkna, Kulozik, & Foerst, 2007). The superiority of fluidized bed drying over freeze drying in terms of storage stability was also validated by Poddar et al. (2014). In one of our previous studies

Corresponding author. E-mail address: [email protected] (A. Nag).

https://doi.org/10.1016/j.foodres.2019.04.058 Received 19 December 2018; Received in revised form 10 April 2019; Accepted 24 April 2019 Available online 27 April 2019 0963-9969/ © 2019 Elsevier Ltd. All rights reserved.

Food Research International 123 (2019) 198–207

A. Nag, et al.

(Nag & Das, 2013), we used the fluidized-bed-drying technique along with the wet granulation process, similar to the combination in the current study, to stabilize 3 lactobacilli and 2 bifidobacterium strains of commercial probiotics and observed very good storage stability at 25 °C but very poor stability at 37 °C. In the current study, we investigated the underlying mechanisms responsible for the loss in cell viability of a validated probiotic strain Lactobacillus reuteri LR6, at a storage temperature of 37 °C, which could help in the design of better stabilization processes in the future. Lactobacillus reuteri LR6 was originally isolated from the feces of the breast-fed human infants (less than 3 months). This particular strain survived in acid, bile, and simulated stomach–duodenum passage conditions, indicating its high tolerance to gastric juice, duodenal juice and bile environments. The strain LR6 did not show strong hydrophobic properties because the percentages of adhesion to the apolar solvents (Singh et al., 2012). An in vivo study focusing on the cholesterol-lowering action of L. reuteri LR6 was also conducted in which the values for total cholesterol, triglyceride and LDL were reduced significantly in the group fed with L. reuteri LR6 but for HDL this difference was not significant (Singh, Malik, Snehal, & Kaur, 2015). There is no general acceptance of any particular type of carrier agent that is essential for the stabilization of probiotic cells and that is solely responsible for offering superior storage stability at enhanced temperatures. Also. the importance of maintaining probiotic functionality along with the live quantity has been emphasized by Champagne, da Cruz, and Daga (2018). It is also not clear which compositional factors (proteins, lipids or carbohydrates), singly or in combination, are more important for this purpose. Therefore, based on several publications, a more direct correlation can be drawn between storage stability and low aw than between storage stability and the nature of the carrier agents (Heidebach, Forst, & Kulozik, 2010; Miao et al., 2008). Sugar alcohols, such as glycerol and mannitol, have been found to have a protective effect during the room temperature storage of probiotics (Savini et al., 2010). Similarly, sorbitol was found to be effective in maintaining the viability of Lactobacillus paracasei and L. plantarum cells even after 150 days of storage at 25 °C when packed in the absence of oxygen (Coulibaly et al., 2010). In our current study, various carrier agents for the stabilization of Lactobacillus reuteri LR6 cells were investigated. They included polysaccharides, milk proteins, a combination of lactose and milk protein and a combination of milk solids and sugar alcohols. The temperature of storage was 37 °C; as this temperature has not commonly been chosen by other researchers, little prior information on the effect of the mentioned carrier agents on the stability of probiotic bacteria at this storage temperature is available. To understand the underlying mechanism responsible for partially or solely causing bacterial death during storage, some previous studies (Chavez & Ledeboer, 2007; Dianawati, Mishra, & Shah, 2013) focused on examining the alterations in the secondary protein structures of the cell envelopes. Although the stabilization techniques and the carrier agents were different in each case, it was found that such alterations took place during room temperature storage even when there was no change in the glass transition temperature of the stabilization matrix. A similar approach was taken in our current study; we investigated such alterations in the secondary protein structures and observed how they correlated with the decay in cell viability over time. The glass transition temperatures of the samples at different aw levels after desiccation and during storage were also determined and the correlation with bacterial death was investigated. Two aw levels were chosen, keeping in mind the common delivery formats in the dry form, i.e. 0.11 for dietary supplements in capsules and 0.30 for formulated powdered foods such as infant formula.

2. Materials and methods 2.1. Preparation of bacterial cell pellets After two growth cycles (up to the stationary phase), L. reuteri LR6 cells (collected from the National Dairy Research Centre, Karnal, India) were inoculated at the 1.0% level into MRS broth (Oxoid, Hampshire, UK), incubated at 37 °C and harvested at the early stationary phase with centrifugation (11,750 ×g for 10 min). After discarding the supernatant, the cell pellets were washed once with 0.25% (w/v) peptone water and the pellets were collected. The average pellet or wet biomass weight obtained from 500 mL of growth medium was 4.0 ± 0.11 g. 2.2. Stabilization of the LR6 cells using wet granulation and an FBD For the wet granulation, a freshly harvested cell pellet (obtained from 1000 mL of growth medium) was slowly mixed under agitation into a bed of 200 g of skim milk powder (SMP) (Fonterra, Auckland, New Zealand) and blended in a dough mixer (Make, Kenwood; Model, Chef). In another two sets of samples, D-sorbitol (Sigma-Aldrich, Auckland, New Zealand) or xylitol (Danisco, Auckland, New Zealand) in powder form (5.0% by weight of the total sample weight, i.e. 10 g of sorbitol/xylitol added into the quantity of cell mass for 200 g of SMP) was directly added to the cell pellets and allowed to dissolve fully before being added to the SMP. The chosen excipients for another three sets of samples were microcrystalline cellulose (MCC; Sigma-Aldrich), corn starch (National Starch, Bridgewater, NJ, USA) and milk protein concentrate (MPC-80; Fonterra). The rationale behind choosing these carrier excipients was the selection of a relatively inert but food grade substance (MCC), a pure carbohydrate but with longer chain lengths of the polysaccharides than lactose (corn starch) and a high concentration of milk protein in the dried form (MPC). The granulation process was carried out for 15 min, followed by drying of the mix in the FBD (Glatt, Binzen, Germany). The inlet temperature was maintained at 75 ± 3 °C. Although there was no provision to control the outlet temperature in the FBD, a consistent difference of 20–25 °C between the inlet and outlet temperatures was observed. The whole experiment was replicated thrice. 2.3. Storage stability of the stabilized LR6 cells After the fluidized bed drying, the stabilized LR6 cells, supported by the selected carrier agents, were adjusted for aw equilibrium. In desiccators, a saturated solution of lithium chloride was used to lower the aw to 0.10–0.11 and a saturated solution of magnesium chloride was used to adjust the aw to 0.30–0.32. The desiccators were placed in a room that was maintained constantly at 37 °C and periodic examination and necessary adjustment to the concentration of the salt solutions were performed. After equilibrium had been reached (within 7–14 days), duplicate samples from each batch were packed into double-layered polypropylene sachets, each containing approximately 5 g of powder, and were placed inside heat-sealed aluminium foil pouches. The packages were stored at 37 °C for 6 months. The samples were analysed every 30 days for viable lactobacilli counts on MRS agar plates (Oxoid) using the pour plating technique. 2.4. Fourier transform infrared (FTIR) spectroscopy 2.4.1. Preparation of samples for FTIR spectroscopy The cells that had been dried in the FBD along with the supporting excipients were first isolated from the dried powder matrix and the subsequent processing to prepare the samples for spectroscopic observations were made following the method suggested by Santivarangkna, Wenning, Foerst, and Kulozik (2007). The individual powder samples, containing the embedded cells on the particle surfaces or in an agglomerated form, were mixed thoroughly with Ringer's 199

Food Research International 123 (2019) 198–207

A. Nag, et al.

techniques used to analyse dairy products (Farah, Silva, Cruz, & Calado, 2018). DSC (TA Instruments, New Castle, DE, USA) was used to determine the glass transition temperatures of the stabilized LR6 cells carried by the various excipient materials. The samples were weighed (about 10 mg) in the stainless steel DSC pans and were then hermetically sealed by putting on the lids and using pressure. The first run of heating was designed to heat the samples to 70 °C at a constant heating rate of 10 °C/min, followed by cooling to 0 °C at the same rate. The second heating round comprised heating the samples from 20 to 120 °C at a uniform 10 °C/min. The mid-point glass transition temperature (Tg) was determined by spotting the change in the heat flow curve using the Universal Analysis 2000 software (TA Instruments).

solution (Sigma-Aldrich). Approximately 10 g of powder was made up to 50 mL in a sterile centrifuge tube. To break down the granules, the samples were vortexed and the tubes were then kept stationary for 1 h to allow the insoluble solids to settle. The MCC and starch powders, both insoluble in cold water, settled at the bottom of the tubes and the cell suspensions at the top were decanted carefully into other tubes. This process was repeated once again to remove any heavier powder particles from the decanted cell suspension and the cells were finally harvested by centrifugation at 4400 ×g for 5 min. The insoluble components in the MPC- and SMP-based samples, thought to be composed principally of coagulated proteins, were allowed to settle. The dissolved portions at the top of the tubes were decanted into fresh tubes and the same process was repeated for the final cell harvesting, but with an additional two washing cycles for these samples. The pellets at the bottom of the tubes were re-suspended in fresh Ringer's solution. The cell suspension (approximate volume, 10 μL) was then spread as a thin film on to a 1-mm-thick CaF2 window (Crystran, Poole, UK) and the moisture was removed by holding the plate at 42 °C for 60 min. Triplicate samples from each batch were spread on to three such plates and were used for the direct IR measurement. The whole experiment was repeated twice.

2.6. Scanning electron microscopy (SEM) Freshly harvested L. reuteri LR6 cells and the cells isolated from the stabilization matrix were examined using SEM. The samples were observed under an FEI Quanta 200 scanning electron microscope (Eindhoven, The Netherlands). They were first mounted on a standard SEM sample stud made of aluminium, using double-sided sticky tape. The samples were then sputter coated with gold using a BAL-TEC SCD 050 sputter coater under a vacuum of 5 × 10−2 millibar. The images were recorded in TIF format using an accelerated voltage of 20 kV and magnifications from 250 to 20,000 times.

2.4.2. Determination of the state of the cell envelopes and the protein secondary structures of LR6 cells Mid-IR spectra were recorded using an FTIR spectrometer (Nicolet 5700, Thermo Fisher Scientific, Waltham, MA, USA) fitted with a KBr beam splitter. The parameters for spectral measurement were: resolution 4/cm; encoding interval 1/cm; scanning speed 0.2/cm/s. For each sample, the mid-IR range of 4000–500/cm was recorded after averaging 256 scans. Before each operation, a background spectrum of air was measured and subtracted using the instrument itself. The mean spectra were the average of three samples from two independent and replicated experiments. The optical chamber was purged continuously with carbon dioxide and water-free air that was generated by the FTIR purge gas generator. The spectra obtained were analysed using Omnic software (Version 7.1, Thermo Fisher Scientific). Resolution of complex bands into their underlying components was achieved by calculating the first derivative using the Savitzky–Golay algorithm with nine-point smoothing, followed by calculating the deconvoluted second-derivative spectra using the same algorithm and smoothing technique (Dziuba, Babuchowski, Nalecz, & Niklewicz, 2007).

3. Results and discussion 3.1. Storage stability of LR6 cells maintained at two different aws For the samples maintained at aw 0.11 and stored at 37 °C, on average, the initial viable cell populations of all samples ranged from 8.34 to 9.14 log CFU/g, with the lowest population recorded in milk potein concentrate (MPC) samples and the highest in corn starch samples (Fig. 1a). The loss in viability was slow during the first 12 weeks but accelerated thereafter. After 24 weeks, the minimum loss was recorded for sorbitol- and xylitol-coated cells that were supported by skim milk powder (SMP) (1.4 log CFU/g). The maximum reduction in cell viability was observed in the corn starch (CS) and SMP samples (1.8 and 1.9 log CFU/g respectively). The net reductions in the viable cell populations for the microcrystalline cellulose (MCC) and MPC samples were 1.6 and 1.7 log CFU/g respectively after 24 weeks. Fig. 1(b) shows that reductions of greater than 2.0 log CFU/g were recorded for all samples stored at 37 °C and maintained at aw 0.30. In contrast to the differential rate of decline observed for storage at aw 0.11, the rate of decline was more uniform throughout the storage period. The minimum reduction in viability was recorded for the MPC samples and sorbitol- or xylitol-coated cells carried by SMP (2.2 log CFU/g). SMP samples without any coating of polyol compound showed a slightly greater reduction of 2.35 log CFU/g at the end of 24 weeks. The maximum loss was observed for the MCC and CS samples (2.5 and 2.6 log CFU/g). In both aw environments, although coating the cells with sorbitol or xylitol proved to be beneficial, the effect was not significant. A linear regression model was used to determine the decay constant for the loss in cell viability over the storage period. The decay constant is represented by the slope of the regression lines shown in Fig. 1(a) and (b). The coefficient of determination (R2) for all the samples stored at aw 0.11 and 0.30 ranged between 80.9% and 96.9%, indicating satisfactory fit of the data to the regression model. The good fit was also confirmed by the ANOVA of regression performed on the data set. Considerable variations in the stability of desiccated probiotic cells at non-refrigerated storage temperatures have been observed by many researchers using different carrier agents. Improved storage stability at non-refrigerated temperatures was reported when fermented soy milk solids (Wang, Yu, & Chou, 2004), skim milk powder (Ananta, Volkert, & Knorr, 2005) or a combination with 10% gum Arabic were used as excipients (Desmond, Ross, O'Callaghan, Fitzgerald, & Stanton, 2002).

2.4.3. Anaconda 3 environment for principal components analysis (PCA) and hierarchical clustering Chemometrics is applied when there is a large and complex dataset, in terms of sample numbers, types, and responses (Granato et al., 2018). To detect variations among individual sample spectra, the data in the comma-separated values format were subjected to PCA, which is a dimension reduction technique that is applied to simplify data and to visualize the most important information in a data set. Hierarchical clustering is another means of studying the similarities between individuals with respect to all variables. Code for the PCA and the hierarchical cluster analysis was written using Python3.6 (in Jupyter Notebook format) via the Anaconda 3 open source distribution of Python. To reduce the noise from the second-derivative data and to increase the resolution, a smoothing filter was applied as per the protocol described in Dziuba et al. (2007). The Savitzky–Golay filter is a low-pass filter that is well adapted to smoothing noisy data. After the baseline correction, first derivatives and second derivatives were calculated (Savitzky–Golay algorithm with nine-point smoothing) using the Omnic software (Version 7.1, Thermo Fisher Scientific). 2.5. Differential scanning calorimetry (DSC) Differential scanning calorimetry (DSC) is a simple thermal analysis methodology recognized as one of the rapid, non-invasive and precise 200

Food Research International 123 (2019) 198–207

A. Nag, et al.

10 9.5

Corn Starch (CS)

9 8.5

Microcrystalline Cellulose (MCC)

Log cfu/g

8

Milk Protein Concentrate (MPC)

7.5 7

Skim Milk Powder (SMP)

6.5

Skim Milk Powder + Sorbitol (SMPs)

6

Skim Milk Powder + Xylitol (SMPx)

5.5 5 0

(a)

4

8

12

Weeks

16

20

24

10 9.5

Corn Starch (CS)

9

Log cfu/g

8.5

Microcrystalline Cellulose (MCC)

8

Milk Protein Concentrate (MPC)

7.5 7

Skim Milk Powder (SMP)

6.5

Skim Milk Powder + Sorbitol (SMPs)

6 5.5

Skim Milk Powder + Xylitol (SMPx)

5 0

(b)

Fig. 1. Changes in viable cell populations of L. reuteri LR6 stabilized in difference matrices during storage at 37 °C and aw 0.11 (a) at aw 0.30 (b) for 24 weeks. The linear regression model estimating the rate of decay in cell viability and being represented by the following equations. For (a): Corn starch y = −0.2689x + 9.6086; R2 = 0.8086 Microcrystalline cellulose y = −0.3129x + 9.2229; R2 = 0.9003 Milk protein concentrate y = −0.0765x + 8.5154; R2 = 0.9557 Skim milk powder y = −0.3782x + 9.6771; R2 = 0.8948 Skim milk powder + Sorbitol y = −0.265x + 9.06; R2 = 0.9353 Skim milk powder + Xylitol y = −0.062x + 8.435; R2 = 0.9582 For (b): Corn starch y = −0.0994x + 9.0211; R2 = 0.9652 Microcrystalline cellulose y = −0.0933x + 8.4954; R2 = 0.9033 Milk protein concentrate y = −0.0876x + 8.5139; R2 = 0.9441 Skim milk powder y = −0.1006x + 8.6546; R2 = 0.8283 Skim milk powder + Sorbitol y = −0.0802x + 8.1621; R2 = 0.8203 Skim milk powder + Xylitol y = −0.0764x + 8.0029; R2 = 0.8292

4

8

12

16

20

24

Weeks

agents, temperature and amount of free moisture available) throughout the storage period play an equally important role. Literatures suggest that the task of maintaining superior stability and functionalities is achievable if the probiotics were carried by a matrix maintained at a very low aw level around 0.10. However, the real challenge is to maintain this superior stability after fortifying such ingredient into a powdered food formulation, mostly done through dry blending, where the aw of the bulk of the materials is kept around 0.20–0.30. Once the probiotic ingredient is mixed into this bulk and stored for some period, the equilibrium water activity is bound to go up and is thought to be the primary reason of deterioration of losing the cell viability over long term storage. Therefore, a thorough understanding of the factors responsible for the loss in probiotic stability at high aw range will help in developing better probiotics products intended for ambient storage.

Similarly, poorer stability was recorded with starch as the excipient (Lian, Hsiao, & Chou, 2002; O'Riordan, Andrews, Buckle, & Conway, 2001) Ananta et al. (2005) concluded that poor protection at ambient temperature may be expected if the carrier polymers do not interact in some way with the bacterial cell membrane during the desiccation process. The results shown in Fig. 1(a) and (b) are generally consistent with the above observation. According to the water replacement theory proposed by Crowe (2002), the corn starch and MCC samples did not contain any disaccharide molecule that could replace the evaporating water molecules during desiccation to form hydrogen bonds with the cell membrane proteins, which, in turn, could prevent their denaturation. Crowe, Oliver, Hoekstra, and Crowe (1997) showed that the vitrification effect of a combination of hydroxyethyl startch and glucose was necessary in reducing the membrane phase-transition temperature and forming a strong glass while successfully stabilizing biological membranes. The absence of such combinations in the starch or cellulose samples in our study could possible explain the enhanced damage to the LR6 cell membranes which effect resulted into poorer storage stability. Any denaturation of membrane proteins could also result in their secondary structural changes affecting bacterial viability. However, it is interesting to observe that, in a low aw storage environment, the negative effects of corn starch and MCC were not very pronounced. The differences in the viable cell populations in these samples compared with those in the best-performing samples (sorbitol- and xylitol-coated cells carried by SMP) were only a maximum 0.5 log CFU/g after 24 weeks of storage at 37 °C. Therefore, it is clear that damage to the bacterial cells during the desiccation process is not the only factor that is decisive for their storage stability. The interactions between the bacterial cells and the surrounding environment (nature of the carrier

3.2. PCA of the FTIR spectra of the LR6 cells isolated at the beginning and the end of the storage period The Fourier transform infrared technique (FTIR) has been used extensively to investigate the conformational changes in the secondary protein structures (Santivarangkna, Naumann, Kulozik, & Foerst, 2010). The FTIR technique is able to assign infrared absorbance bands in the mid-IR region (wavenumber 4000–500/cm) to the cell components, out of which the proteins and polypeptides comprised of the amide 1 and amide 2 bands. The polypeptide and protein repeat units represent nine infra-red absorption bands which are named as amide A, B, and 1–7 (Kong & Yu, 2007). The most prominent vibrational bands among these representing the protein backbone are amide 1 and 2. Approximately 80% of the C]O stretching vibrations of the peptide 201

Food Research International 123 (2019) 198–207

A. Nag, et al.

Fig. 3. Hierarchical cluster map showing heterogeneity of the amide zone spectra of the cells stored at aw 0.11 (a) and at aw 0.30 (b). SMP = Skim milk powder, SMP(S) = SMP + sorbitol, SMP(X) = SMP + xylitol, Starch = Corn starch, MCC = Microcrystalline cellulose, MPC = Milk protein concentrate. Fig. 2. Factor map showing PCA of the amide zone spectra of the cells stored at aw 0.11 (a) and at aw 0.30 (b). SMP = Skim milk powder, SMP (S) = SMP + sorbitol, SMP(X) = SMP + xylitol, Starch = Corn starch, MCC = Microcrystalline cellulose, MPC = Milk protein concentrate.

substantial alteration in the secondary structures of the cell envelope proteins during the storage period. The hierarchical clustering map in Fig. 3(a) shows the heterogeneity among individual samples within a cluster, which was useful in explaining the differences in viable cell counts in different samples and whether these differences were directly correlated with protein structural alterations. The broad cluster (lower half) containing the stored samples included two small subclusters each with two samples (Fig. 3a). The sorbitol- and xylitol-coated cells (SMP(S)_Day180 and SMP(X)_Day180) were grouped together, indicating maximum similarity between them, and the cells carried by CS and SMP formed another subcluster. Cells stabilized with MPC or MCC did not group with any other sample, indicating maximum dissimilarity in their secondary protein structures to those of the other group samples. These results correlate well with the residual viable cell counts in these samples after 180 days of storage at 37 °C and aw 0.11 (Fig. 1a). The LR6 cells were best protected in the sorbitol- and xylitol-coated samples and the subclustering of these two sample spectra indicated the similarity in their secondary protein structures at the end of storage. The maximum reduction in cell viability was observed in the CS and SMP samples, which also formed a cluster. The MCC and MPC samples were moderately affected by the long duration, non-refrigerated storage and were not part of any cluster. Storage at aw 0.30 resulted in a further loss in cell viability in all samples, compared with storage at aw 0.11 (Fig. 1b). This was reflected in the factor map in Fig. 2(b), in which all the Day 0 samples were clustered (within the circle) on the left-hand side and showed considerable distance in the horizontal dimension (representing approximately 62% of the total variance) from the other two clusters (within the ellipse and the rectangle). The rectangle cluster was scattered,

linkages are assigned to the amide 1 band (1700–1600/cm) of the protein secondary structures. The frequencies of the amide 1 band have been found to be highly correlated to each of the secondary structural elements of the proteins. The amide 2 band is relatively less correlated to the protein conformation compared to the amide 1 band and is derived from the in-pane NH bending (40–60%) and the CN stretching vibration (18–40%) (Krimm & Bandekar, 1986). As an example and to explain the methodology used, the total primary spectra of the cells isolated from the six types of powdered samples are shown in Fig. S1 (Supplementary material). Fig. S2 (Supplementary material) is an example of the second-derivative spectra of the amide zones (approximately 1500–1700 cm) for the stabilized LR6 cells and the freshly harvested cells. PCA of the spectral data shows that 85.6% of the total variations among the samples stored at aw 0.11 were contained within the first two principal components, which can be considered to be satisfactory. The MPC samples, both before and after storage, were substantially different from the rest of the group. This is reflected in the factor map (Fig. 2a), in which both samples (codes MPC_Day0 and MPC_Day180) were found to be outliers, the former being placed separately within the circled cluster (all Day 0 samples) and the latter not being a part of any cluster. The other samples were broadly grouped (within the rectangle and the ellipse) according to the storage time, i.e. the spectra of all samples at the beginning of storage were dissimilar to the spectra obtained for the same samples at the end of storage. This indicated a 202

Food Research International 123 (2019) 198–207

A. Nag, et al.

was found to be an effective plasticizer that, when gradually adsorbed by the matrix constituents at high temperature, helped in lowering the Tg of the particles (Pehkonen, Roos, Miao, Ross, & Stanton, 2008). This observation was found to be in agreement with the results shown in Fig. 4 and Table 1, in which the decreases in Tg were recorded with increasing aw of the matrix and during the storage period at 37 °C. The FTIR analysis detected deformations taking place in the secondary protein structures of the LR6 cell envelopes, and these deformations were found to be well correlated with the cell viability at the beginning and the end of the storage period. It could not be revealed from this analysis if such deformations resulted in thermodynamic destabilization and thereby unfolding of the proteins. When the same samples were adjusted to two different aw levels and were stored under similar conditions, a major difference in the physical properties of the samples was the glassy state of the carrier agents to which the bacterial cells adhered. It was earlier hypothesized that a living organism, when embedded in a glassy matrix during dehydration, supported better storage stability (Bruni & Leopold, 1991). To confirm this, Oldenhof, Wolkers, Fonseca, Passot, and Marin (2005) spray dried lactobacillus cells and found improved survival during ambient storage when carried by a 50:50 mixture of maltodextrin and sucrose. It was confirmed that maltodextrin caused strengthening of the glassy matrix, which correlated well with the survival during storage. Such a glassy structure in the external environment is thought to offer an effective barrier against oxygen permeability and to restrict molecular mobility (Ananta et al., 2005). The role of a suitable stabilizing agent is extremely important for protecting the cell envelope proteins during dehydration and also for providing a strong glassy matrix for long term storage stability (Crowe et al., 1988). A popular choice for such purposes has been the common disaccharides such as sucrose and trehalose because these compounds, when in the amorphous state, are able to form hydrogen bonds with the proteins by replacing the water molecules, which become excluded from the system during dehydration (Allison, Chang, Randolph, & Carpenter, 1999). Although the formation of a glassy matrix by these sugars is an important requirement for protecting the proteins during the dehydration step, the glass transition must take place at a temperature that is much higher than the storage temperature of the proteins to ensure long term storage stability (Garzon-Rodriguez et al., 2004). In Fig. 4, an upshift of the Tg for all samples (irrespective of the sorbitol or xylitol coating over the cells) from approximately 48 to approximately 73 °C was observed when the aw was lowered from about 0.23 to about 0.11. The high differential between the storage temperature (37 °C) and the Tg was probably an important factor that was responsible for the superior stability of the LR6 cells in these samples. In contrast, the Tg was depressed to about 35 °C (lower than the storage temperature) when the same samples were adjusted to a higher aw of 0.30, which indicates that the LR6 cells in these samples were not entrapped in a glassy matrix. The rubbery state of the stabilizing agent may have facilitated easy moisture migration to and from the protein molecules (Ananta et al., 2005), resulting in the deformation of their secondary structures, as evident from the FTIR analysis reported in the previous section. Therefore, it can be seen that a high Tg of the stabilizing matrix was favourable for retaining bacterial viability and that the lowering of the Tg during storage in both aw environments was well correlated with the loss in viability.

which represented the moderately affected cells in the four samples, as shown in Fig. 1b. The cluster within the ellipse contained the maximumly affected cells carried by MCC and CS. This cluster was displaced by a small distance vertically from the rectangle cluster, indicating minor differences from other stored samples. As the vertical scale represented only approximately 27% of the total variance, it can be considered to be minor. The distances between samples within the circled cluster (Day 0 samples) and those within the ellipse (Day 180 samples) were substantial in both vertical and horizontal directions, indicating significant alterations of the secondary protein structures in the cells carried by MCC and corn starch during the storage period. The cluster map for the samples stored at aw 0.30 in Fig. 3(b) shows three subclusters within the stored samples. These subclusters consisted of: (1) the worst affected group, containing MCC and CS samples; (2) cells carried by MPC and sorbitol-coated cells carried by SMP; (3) cells carried by SMP only and the xylitol-coated cells carried by SMP. As the last four samples [(2) and (3)] contained almost identical levels of residual viable cells (range 2.2–2.35 log CFU/g), these two subclusters did not reveal any additional information in explaining the role of these carrier excipients as the protectant during storage. All the Day 0 samples were grouped together at the bottom part of the cluster map. This suggests that other factors might need to be included to correctly predict cell viability in a high aw storage environment. 3.3. DSC analysis of the stabilization matrix of the LR6 cells during storage Each DSC thermogram displayed a single observable glass transition (Tg) phase as detected by the change in slope when the samples were heated from 20 to 120 °C at a constant rate (Fig. 4). Immediately after the drying operation, when the aw values of the samples were closer to 0.20–0.23, the Tg was found to be 49.6 °C for the SMP samples, but was slightly reduced to 47.3 and 48.0 °C for the sorbitol- and xylitol-containing samples (carried by SMP) respectively. When the aw values of the samples were adjusted to 0.30, the Tg for all samples was depressed to around 35 °C, which was lower than the storage temperature during the stability trial. This indicates that these samples were not in a glassy state during the course of the storage trial. In contrast, when the aw was adjusted to 0.11, the Tg for all samples was shifted upwards to 72–75 °C, indicating that the samples were in a strong glassy physical state during the storage period. The DSC thermograms shown in Fig. 5 represent the stabilized LR6 cells carried with CS, MCC and MPC. There was no detectable change in the slope in the measured temperature range in any of the thermograms. Therefore, it is difficult to speculate whether the LR6 cells were at all trapped into any kind of strong or weak glassy matrix in these samples during the storage period. From the stability results shown in Fig. 1(a) and (b), it can be seen that the ability of these carriers, in terms of storage stability in low as well as high aw environments, was poorer compared with the abilities of the SMP group samples. This could be correlated with the absence of any strong glassy state in these samples, as found by the DSC analysis. Table 1 lists the changes in Tg of the same samples after storage at 37 °C for 121 days. For each sample, the measured Tg was slightly lower than that recorded at the beginning of storage. The depression in Tg for the samples maintained at aw 0.11 ranged from 3.3 to 7.1 °C and that for the samples maintained at aw 0.30 ranged from 5.8 to 8.5 °C. Lapsiri, Bhandari, and Wanchaitanawong (2013) found a direct relationship between the stability of spray-dried L. plantarum cells and both the storage temperature and the relative humidity, which in effect controls the aw of the stabilization matrix. They observed diminishing stability with both increased storage temperature and increased relative humidity and concluded that these two factors influence the Tgs of the matrix and consequently the degree of protection available to the viable cells. Increasing the storage temperature at a constant relative humidity of 33% was found to aid in lowering the Tg, probably because of the ease of water mobility within the matrix (Bhandari & Adhikari, 2009). Water

3.4. SEM of the LR6 cells at the beginning and the end of the storage period Electron microscopy has been found to be a useful tool in revealing the direct damage to the bacterial envelope, such as breakdown of the osmoregulatory capacity and changes in the intracellular DNA region caused by external physical stress or chemical action (Hartmann et al., 2010). Freshly harvested cells were found to have a smooth surface and an undamaged membrane of about 1.5 μm in length [Fig. 6(a)]. Desiccated 203

Food Research International 123 (2019) 198–207

A. Nag, et al.

Fig. 4. Representative DSC thermograms showing Tgs at the beginning of the storage trial for the LR6 cells stabilized with (a) only SMP, (b) SMP with sorbitol and (c) SMP with xylitol. Each figure displays three thermograms for the same sample after FBD drying and after adjustments of the aw to 0.11 and 0.30.

dried matrix did not improve the surface wrinkles, indicating that the damage was permanent in nature. Such corrugated surfaces with dimples and blisters, as seen in the desiccated samples, have been reported previously by Hartmann et al. (2010) when Staphylococcus aureus cells were treated with antimicrobial peptides. They suggested that, if the bacterial cell membrane is destabilized by some external means, in this case heating because of

cells along with MCC showed a corrugated surface with a few dimpletype structures [Fig. 6(b)]. About 20% of the cells showed a smooth surface, similar to that of the fresh cells. Cells coated with xylitol and dried along with SMP showed similar deformations on the surface but to a lesser extent [Fig. 6(c)]. About 35% were found to be undamaged. There was no visible reduction in the cell size for either of the dried samples. The rehydration step in the cell isolation process from the 204

Food Research International 123 (2019) 198–207

A. Nag, et al.

Fig. 5. DSC thermograms of the LR6 cells stabilized with corn starch (CS), Microcrystalline cellulose (MCC) and Milk protein concentrate (MPC). Each figure displays three thermograms for the same sample after FBD drying and after adjustments of the aw to 0.11 and 0.30.

could have been due to aberrations in the membrane lipid composition, altered membrane fluidity and/or membrane integrity, resulting in cell wall lysis and loss of intracellular dense material.

Table 1 Changes in glass transition temperatures (Tg) of the samples shown in Fig. 5 after storage at 37 °C for 121 days. Glass transition temperature (Tg)

SMP aw 0.11 aw 0.30 SMP + Sorbitol aw 0.11 aw 0.30 SMP + Xylitol aw 0.11 aw 0.30

After desiccation

After 121 days

°C

°C

72.1 34.7

68.9 28.9

73.2 34.6

68.7 26.1

74.9 33.5

67.8 27.7

4. Conclusions This study confirmed the importance of maintaining a lower aw of the stabilization mix when enhanced temperature storage is the objective. However, the stability in a high aw environment was not very encouraging. Therefore, the challenge to the food industry in delivering stabilized probiotics at around aw 0.30 remains and more research in finding a solution is recommended. The useful correlation that was found between the alteration in secondary protein structures (as obtained from the FTIR analysis) and the loss in cell viability is somewhat novel. This is one of the few studies that has investigated the storage of probiotics at 37 °C, which is very often the ambient temperature in countries with tropical weather; hence, the importance of shelf-stable probiotic products at this temperature is paramount.

the desiccation process, the local disruption of the inner membrane would lead to the cytoplasmic material filling the periplasmic space, causing the blisters or corrugated-type deformations without completely rupturing the outer membrane. They also found a direct correlation of such deformations with cell death, confirming our study in which maximum and minimum reductions in viability of the LR6 cells immediately after desiccation were recorded in the MCC samples and the xylitol-coated cells carried by SMP respectively. Santivarangkna, Kulozik, and Foerst (2006) used atomic force microscopy to examine the cell envelopes of vacuum-dried Lactobacillus helveticus cells. A similar type of wrinkled surface with some cracks was reported and was found to be well correlated with lysis of the cells, loss of cell integrity, metabolic activities and residual viability. Gilbert, Pemberton, and Wilkinson (1990) observed that minor damage to the cell membranes occurred during the harvesting process through high-speed centrifugation, which causes shear stresses. According to Wyber, Andrews, and Gilbert (1994), such sublethally injured cells can become more susceptible to loss of viability during secondary stresses such as drying. After the storage period of 180 days, when the cells were isolated from the same samples, a greater degree of damage, in terms of the number of affected cells and the visual quality of the wrinkles, was noticed. Almost all the cells dried with MCC developed further membrane deformations during the storage period [Fig. 6(d)], whereas a few cells were found to retain smooth surfaces similar to those of fresh cells when coated with xylitol and dried with SMP [Fig. 6(e)]. According to Sikkema, Debont, and Poolman (1994), these morphological alterations

Declarations of interest None. Author contributions Arup Nag conducted the research work, analysed the data and prepared the manuscript. Associate prof. Mark Waterland guided on how to use the FTIR spectrometer and the Anaconda software to do the statistical analysis and generate the figures. Dr. Patrick Janssen and Dr. Rachel Anderson helped in supervising the routine progresses of the work and preparing the manuscript. Distinguished prof. Harjinder Singh conceived the project, advised on the planning, approved the budget and actively supervised the progress. All authors have approved the final version of the article. Acknowledgements This work was supported by the Riddet Institute, a ‘Centre of Research Excellence’ hosted by Massey University, and the Institute of Fundamental Sciences, Massey University, New Zealand. 205

Food Research International 123 (2019) 198–207

A. Nag, et al.

(a)

(b)

(c)

(d)

(e)

Fig. 6. SEM images of (a) freshly harvested LR6 cells, (b) cells carried by Microcrystalline cellulose and (c) cells coated with xylitol and carried by Skim milk powder after stabilization. 7(d) and 7(e) are images of the same samples respectively after 121 days of storage at 37 °C.

Appendix A. Supplementary data

Cruz, A. G. (2015). Quality parameters of probiotic yogurt added to glucose oxidase compared to commercial products through microbiological, physical-chemical and metabolic activity analyses. Food Research International, 77, 627–635. Bhandari, B., & Adhikari, B. (2009). Glass-transition based approach in drying of foods. In C. Ratti (Ed.). Advances in food dehydration (pp. 38–62). London: CRC Press. Bruni, F., & Leopold, A. C. (1991). Glass transitions in soybean seed – Relevance to anhydrous biology. Plant Physiology, 96, 660–663. Champagne, C. P., da Cruz, A. G., & Daga, M. (2018). Strategies to improve the functionality of probiotics in supplements and foods. Current Opinion in Food Science, 22, 160–166. Chavez, B. E., & Ledeboer, A. M. (2007). Drying of probiotics: Optimization of formulation and process to enhance storage survival. Drying Technology, 25, 1193–1201. Coulibaly, I., Dubois-Dauphin, R., Destain, J., Fauconnier, M.-L., Lognay, G., & Thonart, P. (2010). The resistance to freeze-drying and to storage was determined as the cellular ability to recover its survival rate and acidification activity. International Journal of

Supplementary data to this article can be found online at https:// doi.org/10.1016/j.foodres.2019.04.058. References Allison, S. D., Chang, B., Randolph, T. W., & Carpenter, J. F. (1999). Hydrogen bonding between sugar and protein is responsible for inhibition of dehydration-induced protein unfolding. Archives of Biochemistry and Biophysics, 365, 289–298. Ananta, E., Volkert, M., & Knorr, D. (2005). Cellular injuries and storage stability of spray-dried Lactobacillus rhamnosus GG. International Dairy Journal, 15, 399–409. Batista, A. L. D., Silva, R., Cappato, L. P., Almada, C. N., Garcia, R. K. A., Silva, M. C., ...

206

Food Research International 123 (2019) 198–207

A. Nag, et al.

Miao, S., Mills, S., Stanton, C., Fitzgerald, G. F., Roos, Y., & Ross, R. P. (2008). Effect of disaccharides on survival during storage of freeze dried probiotics. Dairy Science & Technology, 88, 19–30. Nag, A., & Das, S. (2013). Improving ambient temperature stability of probiotics with stress adaptation and fluidized bed drying. Journal of Functional Foods, 5(1), 170–177. Nadelman, P., Frazao, J. V., Vieira, T. I., Balthazar, C. F., Andrade, M. M., Alexandria, A. K., & Maia, L. C. (2017). The performance of probiotic fermented sheep milk and ice cream sheep milk in inhibiting enamel mineral loss. Food Res. Int. 97, 184–190. Oldenhof, H., Wolkers, W. F., Fonseca, F., Passot, S. P., & Marin, M. (2005). Effect of sucrose and maltodextrin on the physical properties and survival of air-dried Lactobacillus bulgaricus: An in situ Fourier transform infrared spectroscopy study. Biotechnology Progress, 21, 885–892. O'Riordan, K., Andrews, D., Buckle, K., & Conway, P. (2001). Evaluation of microencapsulation of a Bifidobacterium strain with starch as an approach to prolonging viability during storage. Journal of Applied Microbiology, 91, 1059–1066. Pehkonen, K. S., Roos, Y. H., Miao, S., Ross, R. P., & Stanton, C. (2008). State transitions and physicochemical aspects of cryoprotection and stabilization in freeze-drying of Lactobacillus rhamnosus GG (LGG). Journal of Applied Microbiology, 104(6), 1732–1743. Poddar, D., Das, S., Jones, G., Palmer, J., Jameson, G. B., Haverkamp, R. G., & Singh, H. (2014). Stability of probiotic Lactobacillus paracasei during storage as affected by the drying method. International Dairy Journal, 39, 1–7. Rodrigues, D., Sousa, S., Rocha-Santos, T., Silva, J. P., Lobo, J. M. S., Costa, P., ... Freitas, A. C. (2011). Influence of L-cysteine, oxygen and relative humidity upon survival throughout storage of probiotic bacteria in whey protein-based microcapsules. International Dairy Journal, 21(11), 869–876. Santivarangkna, C., Kulozik, U., & Foerst, P. (2006). Effect of carbohydrates on the survival of Lactobacillus helveticus during vacuum drying. Letters in Applied Microbiology, 42, 271–276. Santivarangkna, C., Kulozik, U., & Foerst, P. (2007). Alternative drying processes for the industrial preservation of lactic acid starter cultures. Biotechnology Progress, 23, 302–315. Santivarangkna, C., Naumann, B., Kulozik, U., & Foerst, P. (2010). Protective effects of sorbitol during the vacuum drying of Lactobacillus helveticus: An FT-IR study. Annales de Microbiologie, 60, 235–242. Santivarangkna, C., Wenning, M., Foerst, P., & Kulozik, U. (2007). Damage of cell envelope of Lactobacillus helveticus during vacuum drying. Journal of Applied Microbiology, 102(3), 748–756. Savini, M., Cecchini, C., Verdenelli, M. C., Silvi, S., Orpianesi, C., & Cresci, A. (2010). Pilot-scale production and viability analysis of freeze-dried probiotic bacteria using different protective agents. Nutrients, 2, 330–339. Sikkema, J., Debont, J. A. M., & Poolman, B. (1994). Interactions of cyclic hydrocarbons with biological-membranes. Journal of Biological Chemistry, 269(11), 8022–8028. Singh, T. P., Kaur, G., Malik, R. K., Schillinger, U., Claudia, G., & Kapila, S. (2012). Characterization of intestinal Lactobacillus reuteri strains as potential probiotics. Probiotics and Antimicrobial Proteins, 4, 47–58. Singh, T. P., Malik, R. K., Snehal, K. G., & Kaur, G. (2015). Hypocholesterolemic effects of Lactobacillus reuteri LR6 in rats fed on high-cholesterol diet. International Journal of Food Sciences and Nutrition, 66(1), 71–75. Stummer, S., Toegel, S., Rabenreither, M.-C., Unger, F. M., Wirth, M., Viernstein, H., & Salar-Behzadi, S. (2012). Fluidized-bed drying as a feasible method for dehydration of Enterococcus faecium M74. Journal of Food Engineering, 111, 156–165. Wang, Y. C., Yu, R. C., & Chou, C. C. (2004). Viability of lactic acid bacteria and bifidobacteria in fermented soymilk after drying, subsequent rehydration and storage. International Journal of Food Microbiology, 93, 209–217. Wyber, J. A., Andrews, J., & Gilbert, P. (1994). Loss of salt-tolerance and transformation efficiency in Escherichia coli associated with sublethal injury by centrifugation. Letters in Applied Microbiology, 19, 312–316.

Microbiology, 2010, 625239. Crowe, J. H., Crowe, L. M., Carpenter, J. F., Rudolph, A. S., Wistrom, C. A., Spargo, B. J., & Anchordoguy, T. J. (1988). Interactions of sugars with membranes. Biochimica et Biophysica Acta, 947, 367–384. Crowe, J. H., Oliver, A. E., Hoekstra, F. A., & Crowe, L. M. (1997). Stabilization of dry membranes by mixtures of hydroxyethyl starch and glucose: The role of vitrification. Cryobiology, 35(1), 20–30. Crowe, L. M. (2002). Lessons from nature: The role of sugars in anhydrobiosis. Comparative Biochemistry and Physiology Part A: Molecular and Integrative Physiology, 131, 505–513. Desmond, C., Ross, R. P., O'Callaghan, E., Fitzgerald, G., & Stanton, C. (2002). Improved survival of Lactobacillus paracasei NFBC 338 in spray-dried powders containing gum acacia. Journal of Applied Microbiology, 93, 1003–1011. Dianawati, D., Mishra, V., & Shah, N. P. (2013). Stability of microencapsulated Lactobacillus acidophilus and Lactococcus lactis ssp. cremoris during storage at room temperature at low aw. Food Research International, 50, 259–265. Dianawati, D., & Shah, N. P. (2011). Survival, acid and bile tolerance, and surface hydrophobicity of microencapsulated B. animalis ssp. lactis Bb12 during storage at room temperature. Journal of Food Science, 76, M592–M599. Dziuba, B., Babuchowski, A., Nalecz, D., & Niklewicz, M. (2007). Identification of lactic acid bacteria using FTIR spectroscopy and cluster analysis. International Dairy Journal, 7, 183–189. FAO/WHO (2001). Health and nutritional properties of probiotics in food including powder milk with live lactic acid bacteria. Report of a joint FAO/WHO expert consultation http://www.who.int/foodsafety/publications/fs_management/en/probiotics.pdf. Farah, J. S., Silva, M. C., Cruz, A. G., & Calado, V. (2018). Differential calorimetry scanning: Current background and application in authenticity of dairy products. Current Opinion in Food Science, 22, 88–94. Garzon-Rodriguez, W., Koval, R. L., Chongprasert, S., Krishnan, S., Randolph, T. W., Warne, N. W., & Carpenter, J. F. (2004). Optimizing storage stability of lyophilized recombinant human interleukin-11 with disaccharide/hydroxyethyl starch mixtures. Journal of Pharmaceutical Sciences, 93, 684–696. Gilbert, P., Pemberton, D., & Wilkinson, D. E. (1990). Synergism within polyhexamethylene biguanide biocide formulations. Journal of Applied Bacteriology, 69, 593–598. Granato, D., Putnik, P., Kovacevic, D. B., Santos, J. S., Calado, V., Rocha, R. S., ... Pomerantsev, A. (2018). Trends in chemometrics: Food authentication, microbiology, and effects of processing. Comprehensive Reviews in Food Science and Food Safety, 17(3), 663–677. Hartmann, M., Berditsch, M., Hawecker, J., Ardakani, M. F., Gerthsen, D., & Ulrich, A. S. (2010). Damage of the bacterial cell envelope by antimicrobial peptides gramicidin S and PGLa as revealed by transmission and scanning electron microscopy. Antimicrobial Agents and Chemotherapy, 54, 3132–3142. Heidebach, T., Forst, P., & Kulozik, U. (2010). Influence of casein-based microencapsulation on freeze-drying and storage of probiotic cells. Journal of Food Engineering, 98, 309–316. Hill, C., Guarner, F., Reid, G., Gibson, G. R., Merenstein, D. J., & Pot, B. (2014). The International Scientific Association for Probiotics and Prebiotics consensus statement on the scope and appropriate use of the term probiotic. Nat. Rev. Gastroenterol. Hepatol. 11(8), 506–514. Kong, J., & Yu, S. (2007). Fourier transform infrared spectroscopic analysis of protein secondary structures. Acta Biochimica et Biophysica Sinica, 39, 549–559. Krimm, S., & Bandekar, J. (1986). Vibrational spectroscopy and conformation of peptides, polypeptides, and proteins. Advances in Protein Chemistry, 38, 181–364. Lapsiri, W., Bhandari, B., & Wanchaitanawong, P. (2013). Stability and probiotic properties of Lactobacillus plantarum spray-dried with protein and other protectants. Drying Technology, 31(13–14), 1723–1733. Lian, W. C., Hsiao, H. C., & Chou, C. C. (2002). Survival of bifidobacteria after spraydrying. International Journal of Food Microbiology, 74, 79–86.

207