Significant effect of Ca2+ on improving the heat resistance of lactic acid bacteria


  • Song Huang,

    1. Department of Chemical and Biochemical Engineering, College of Chemistry and Chemical Engineering, Xiamen University, Xiamen, China
    Search for more papers by this author
  • Xiao Dong Chen

    Corresponding author
    1. Department of Chemical Engineering, Monash University, Clayton, Vic., Australia
    • Department of Chemical and Biochemical Engineering, College of Chemistry and Chemical Engineering, Xiamen University, Xiamen, China
    Search for more papers by this author

Correspondence: Xiao Dong Chen, Department of Chemical and Biochemical Engineering, College of Chemistry and Chemical Engineering, Xiamen University, 422 Siming South Road, Xiamen 361005, China. Tel.: +86 592 2189073; fax: +86 592 2188855; e-mails:,


The heat resistance of lactic acid bacteria (LAB) has been extensively investigated due to its highly practical significance. Reconstituted skim milk (RSM) has been found to be one of the most effective protectant wall materials for microencapsulating microorganisms during convective drying, such as spray drying. In addition to proteins and carbohydrate, RSM is rich in calcium. It is not clear which component is critical in the RSM protection mechanism. This study investigated the independent effect of calcium. Ca2+ was added to lactose solution to examine its influence on the heat resistance of Lactobacillus rhamnosus ZY, Lactobacillus casei Zhang, Lactobacillus plantarum P8 and Streptococcus thermophilus ND03. The results showed that certain Ca2+ concentrations enhanced the heat resistance of the LAB strains to different extents, that is produced higher survival and shorter regrowth lag times of the bacterial cells. In some cases, the improvements were dramatic. More scientifically insightful and more intensive instrumental study of the Ca2+ behavior around and in the cells should be carried out in the near future. In the meantime, this work may lead to the development of more cost-effective wall materials with Ca2+ added as a prime factor.


Lactic acid bacteria (LAB) are among the most important groups of microorganisms used in the food industry. Many species of LAB have obtained ‘Generally Recognized as Safe' status (Sonomoto & Yokota, 2011). As starters for producing a variety of fermented products, LAB is also the most common microbe employed as ‘probiotics’ (FAO/WHO, 2002). Therefore, large-scale production of LAB products containing high levels of viable cells is necessary. It has been expected that enhancing the heat resistance of microbic cells would facilitate large-scale production of live microorganisms via effective and more economical convective thermal drying (Chen & Patel, 2007; Huang et al., 2009; Fu et al., 2011).

Reconstituted skim milk (RSM) has been proved to be an adequate wall material (probably the best illustrated by far) for maintaining the survival of microorganisms in thermal microencapsulating processes (Li et al., 2006; Fu & Chen, 2011). However, the mechanism by which it conveys this protection is unclear and is an attractive subject of scientific research, apart from its significant practical impact. It is worth noting that RSM has a large amount of calcium (≥ 30 mM, depending on different methods of production), although it is the proteins and lactose in RSM that have been speculated to be the major causes of the enhanced heat resistance of bacterial cells (Morgan et al., 2006; Ghandi et al., 2012).

Extensive investigations have shown that calcium influenced the heat stability of milk (Bansal & Chen, 2006; Chandrapala et al., 2010). Essentially, it has a considerable effect on the stability of milk proteins (Omoarukhe et al., 2010; Ramasubramanian et al., 2012). From a biological point of view, the importance of calcium as a signal messenger or cell regulator in eukaryotes has also been well established (Dolmetsch et al., 1998; Tidow et al., 2012). However, the role of calcium in prokaryotes is still elusive (Dominguez, 2004). It has been found that calcium salt is beneficial for the growth of LAB (Nakano et al., 2012). Moreover, Ca2+ improved the heat stability of the cell envelope proteinase of LAB (Exterkate & Alting, 1999; Espeche Turbay et al., 2009).

Based on the above studies on the effects of calcium, it could be reasoned that calcium may have a quantitative role in protecting LAB against heat and dehydration. In the current work, we have conducted experiments to establish the extent of the effect of calcium on preserving LAB during heat treatments.

Materials and methods


Among numerous LAB, Lactobacillus and Streptococcus thermophilus have attracted a great deal of attention in recent years due to their long historical applications and multitudinous beneficial influences on fermented food and human (Iyer et al., 2010; Sonomoto & Yokota, 2011). Three Lactobacillus strains and one Streptococcus thermophiles were used in this study. Lactobacillus rhamnosus ZY (LZY) was isolated from a commercial probiotic capsule from Taiwan and identified as L. rhamnosus by alignment of the 16S rRNA gene sequence (GenBank accession no. KC012630) through the basic local alignment search tool (ncbi-blast). Lactobacillus casei Zhang (LCZ), Lactobacillus plantarum P8 (LP) and S. thermophilus ND03 (ST) were provided by the Key Laboratory of Dairy Biotechnology and Engineering, Ministry of Education, China (Sun et al., 2011; Bao et al., 2012; Wang et al., 2013).


MRS broth (MRS), MRS agar (MRSa) and maximum recovery diluent (MRD) were purchased from Oxoid (Hampshire, UK). All other reagents were purchased from Sigma (Shanghai, China). The water was obtained from a Milli-Q system with resistivity > 18 MΩ cm−1. All glass, plasticware and water were sterilized at 121 °C for 20 min. MRS, MRSa and MRD were prepared according to the manufacturer's instructions and sterilized at 121 °C for 15 min.

Bacterial strains and culture conditions

All the strains were maintained as a stock culture frozen at −80 °C in MRS with 15% (v/v) glycerol. For broth-grown cultures, the stock culture of each strain was first propagated by streaking a loop-full of frozen culture on MRSa and incubating at 37 °C for 48 h. A single colony of each strain was then inoculated in 10 mL MRS broth and statically incubated 12 h at 37 °C for activation. Subsequently, a 1% inoculation of each strain was made for sub-culturing in a 37 °C stationary incubator for 24 h prior to heat treatments.

Preparation of bacterial suspension

To create a simple scenario, calcium chloride (CaCl2) was added to lactose solution (with no protein present) to formulate the protective medium. Cell suspensions of 1 and 10 mL of each strain were prepared for two heating regimes, respectively. Each 1- or 10-mL cultures was transferred into a 1.5- or 50-mL centrifugal tube. The cells were harvested using centrifugation at 8000 g, 4 °C for 15 min and washed with sterile water in the same centrifugal conditions. The cell pellets of each strain were then re-suspended in 1 or 10 mL 10% (w/v) lactose solutions containing 1, 5, 10, 100 and 1000 mM CaCl2, respectively. The control samples were made by re-suspending the above harvested cells in the solution with no CaCl2 addition.

To confirm the independent effect of Ca2+, 1-mL cell suspensions of LZY in maltose solutions with 0, 1, 5, 10, 100 and 1000 mM CaCl2 and lactose solutions with 0, 5, 10 and 50 mM calcium acetate (CaAc) were also prepared as described above. The pH of each solution containing CaAc was adjusted previously to the value of 10% lactose solution (c. 6.0) with 9.98% (v/v) acetic acid using a pH meter (Mettler Toledo, Columbus, OH). The maltose or lactose solution without Ca2+ addition was also used as control. All cell suspensions were then placed at 25 °C for 30 min before heat treatment experiment.

Heat treatment through a temperature elevation scheme (heat treatment I)

From the 1-mL sample in a 1.5-mL tube, a 500-μL cell suspension was transferred into 4.5 mL MRD for 10 times dilution as the initial sample before heat treatments. The remaining 500-μL samples in 1.5-mL centrifugal tubes were heated simultaneously, supported by a floating board at 90 °C for 45 s using a thermostat water bath (Jing Hong, Shanghai, China). The temperature within the tubes was monitored using a micro-thermocouple (Pico Technology, Garden City, KS) inserted through the hole on its cap (a typical temperature trace is shown in Fig. 1). Following the heat treatment, the sample tubes were immediately immersed in an ice-water bath for 2 min and diluted 10 times with MRD. All the samples in MRD were stored at 4 °C prior to further use.

Figure 1.

The temperature/time curves in two heat treatment regimes. The intensity of heat stress in the rising temperature period is expressed as the area below the temperature/time curve, the area value was calculated with the trapezoidal rule.

Heat treatment with a constant temperature regime (heat treatment II)

To ensure that the findings were not dependent on the heating regime, a constant temperature treatment was performed with LZY and ST. The 10-mL sample within a 50-mL tube was heated according to a previous study (Desmond et al., 2001). The temperature was increased from 25 to 60 °C by placing the culture-containing tube in a 70 °C water bath. The tube was then transferred quickly into a 60 °C water bath and held for 10 min. A 500-μL sample was transferred from each tube to the 4.5-mL MRD for 10 times dilution at 0, 5 and 10 min after holding at 60 °C. The temperature within the tube was monitored with a duplicate sample using the micro-thermocouple throughout the experiment (Fig. 1). All the samples in MRD were then stored at 4 °C for further use.

Measurement of bacterial growth curves and survivals

The Bioscreen C system (Labsystems, Vantaa, Finland) was used to measure the growth curve of LAB. A 1% inoculation was done by adding a 35-μL sample-containing MRD to the well of the honeycomb plate (Labsystems) containing 315-μL MRS. All runs were performed at stationary states at 37 °C for 36 h and optical density (OD)600 was measured at 30-min intervals with moderate shaking before measurement. Meanwhile, the remaining sample-containing MRD were serially diluted by MRD and spread on MRSa plates. The plates were incubated at 37 °C for 48 h.

Determination of bacterial viability

The survival data obtained using the plate method was expressed as log N/N0. N is the viable cell count following heat treatments and N0 represents the viable cell count before heat treatments. The regrowth lag time (λ) and LAB survival were tested. The λ was calculated from the growth curve and defined as the point of intersection of the tangent at the steepest slope of the growth curve and the x-axis (Pirt, 1975; Smelt et al., 2002). Since the growth curves measured by Bioscreen have shown clear exponential phases, the five data points on each growth curve which were the closest to (ODmax − ODmin)/2 were used to obtain the tangents (see Fig. 2).

Figure 2.

An example to demonstrate the determination of regrowth lag time (λ) in a growth curve.

Statistical analysis

All data were presented as mean values ± SD. The mean values and SD were calculated from three independent assays. Statistical analysis was performed using anova followed by Dunnett's test with spss v19.0.

Results and discussion

Before heat treatment

The cells were exposed to the different concentrations of Ca2+ for 30 min at 25 °C before heat treatments as described above. It was necessary first to determine the effect of Ca2+ on the viability of unheated cells (i.e. N0 and λ0). The N0 and λ0 of four strains are shown in Table 1. The cells in the solutions containing 0–100 mM Ca2+ had almost the same level of N0 and λ0. However, the cells suspended in 1000 mM CaCl2 showed a visible reduction of N0 and extension of λ0. This may be due to the high osmotic pressure produced by 1000 mM CaCl2. Even so, the viabilities of all samples prior to heat treatments were still maintained at acceptably high levels and they could therefore be used for further experiments.

Table 1. Viable cell counts N0 (CFU mL−1) and regrowth lag time λ0 (h) of the samples before heat treatments
 Concentrations of Ca2+ (mM)
  1. Data without * and means the cells in lactose solutions with CaCl2.

  2. Superscript represents a significant difference from the control values (0 mM Ca2+) on the same lines (a: < 0.1; b: < 0.05; c: < 0.01).

  3. a

    Cells in maltose solutions with CaCl2.

  4. b

    Cells in lactose solutions with CaAc.

N0 (×109)8.6 ± 1.48.9 ± 1.110.1 ± 1.78.2 ± 0.9 9.5 ± 1.46.0 ± 1.9
8.0 ± 1.1a8.8 ± 1.6a7.9 ± 1.1a8.1 ± 0.3a 8.1 ± 2.0a6.8 ± 2.1a
 8.4 ± 1.6b8.0 ± 0.4b9.3 ± 1.6b9.0 ± 1.9b  
λ01.86 ± 0.041.70 ± 0.031.76 ± 0.041.77 ± 0.07 1.75 ± 0.112.46 ± 0.10c
2.13 ± 0.16a2.29 ± 0.14a2.17 ± 0.13a2.43 ± 0.03a 2.41 ± 0.18a2.81 ± 0.20ab
 2.09 ± 0.24b2.07 ± 0.19b2.09 ± 0.10b2.09 ± 0.19b  
N0 (× 109)5.7 ± 1.44.8 ± 1.85.6 ± 0.86.2 ± 0.6 5.1 ± 1.03.9 ± 1.5
λ01.49 ± 0.031.17 ± 0.071.24 ± 0.121.45 ± 0.06 1.20 ± 0.152.44 ± 0.31c
N0 (× 109)6.3 ± 2.16.9 ± 1.45.7 ± 0.96.1 ± 0.7 5.7 ± 1.14.8 ± 1.6
λ03.90 ± 0.064.13 ± 0.134.04 ± 0.013.88 ± 0.09 4.11 ± 0.174.65 ± 0.24c
N0 (× 109)7.6 ± 1.28.1 ± 0.86.2 ± 2.07.7 ± 1.5 6.0 ± 2.04.2 ± 1.7a
λ03.54 ± 0.213.66 ± 0.253.49 ± 0.173.61 ± 0.05 3.63 ± 0.074.01 ± 0.23b

Heat treatment I

As can be seen in Fig. 3a, the samples that were heated all showed a reduction of viable cells, that is log (N/N0) < 0. The addition of CaCl2 had an obvious effect on the viability of the LAB in heat treatment I. Although the optimal concentration for enhancing the heat resistance of each strain was strain-specific, 5 and 10 mM CaCl2 showed general protective effects for all strains in this work (P < 0.1 for LCA and < 0.05 for other strains). In contrast, all samples with 1000 mM CaCl2 displayed significant damage to all LAB strains with this heat treatment. However, 100 mM CaCl2 showed positive effects on improving the heat resistances of LCZ and ST, but very negative effects on that of LZY and LP.

Figure 3.

(a) Survival (log N/N0) of four LAB strains after heat treatment I. (b) Enlargement of regrowth lag time (λ–λ0) of four strains after heat treatment I. (c) Viability of LZY in different protective agents after heat treatment I. Solid symbols refer to the left log (N/N0) and hollow symbols refer to the right λ–λ0. The dotted lines indicate the λ which is longer than the tested time (36 h).

Meanwhile, after heating, all the samples showed an enlarged λ compared with the corresponding samples before heating (i.e. λ > λ0, see Fig. 3b). It is well known that λ is a significant characteristic for quantifying the viability of microorganisms (Florent & Marie-Laure, 2004). A large λ is undesirable in LAB fermentation and consumption. The increased λ in this work could be the result of the loss of viable cells and cell injury caused by heat stress (Busta, 1976; Smelt et al., 2002). When comparing λ–λ0 values, it was evident that 5 and 10 mM CaCl2 enhanced the heat resistance of the four LAB strains. In addition, the other concentrations of CaCl2 also displayed the coincident result of cell viabilities with that of in Fig. 3a.

The LZY in both maltose solutions with CaCl2 and lactose solutions with CaAc showed trends similar to those in lactose solutions with CaCl2 (Fig. 3c). The comparison of maltose and lactose solutions reveals that the effect of CaCl2 is not connected with the presence of lactose, which is commonly used as a nutrient of LAB. The experiments conducted for both CaCl2 and CaAc suggest that the Ca2+ may act independently on the cells and is arguably the main factor that influences the heat resistance of the LAB strains in this work.

Heat treatment II

Rising temperature period

The constant temperature heat treatment began at the point when the temperature of sample reached 60 °C. Before this point, the samples experienced a period of rising temperature (see Fig. 1). The viable cell counts of both LZY and ST in the solutions with different concentrations of Ca2+ were reduced to various extents when the temperature reached 60 °C (Fig. 4a). Compared with the results in heat treatment I (Fig. 3a), the samples with 5 mM CaCl2 still had the general protective effects in the two strains (< 0.05 for LZY and = 0.21 for ST). The 1000 mM CaCl2 also showed a significant reduction in LAB survival compared with controls. It is interesting to note that 10 and 100 mM Ca2+, which displayed positive effects on the heat resistances of LZY and ST, respectively, in heat treatment I, showed non-significant and negative effects on the heat resistances of corresponding strains in heat treatment II. The results were also well supported by the extension of regrowth lag time (i.e. λ–λ0 values, see Fig. 4b).

Figure 4.

(a) Survival (log N/N0) of LZY and ST after rising temperature period of heat treatment II. (b) Enlargement of regrowth lag time (λ–λ0) of LZY and ST after rising temperature period of heat treatment II.

Thus, according to the different temperature/time regimes used in the treatments, the extents of the Ca2+ effect may vary. It could be speculated that the Ca2+ effect may be related to the contact time of cells and Ca2+. The cells in heat treatment II were exposed to heat stress for a longer time and at a stronger intensity than cells exposed to heat treatment I (Fig. 1). This may exceed the threshold at which 10 mM or 100 mM CaCl2 still has a positive effect on LZY or ST. In addition, it may also be associated with the signal role of Ca2+ (Dominguez, 2004). It has been found that environmental stresses (include heat stress) can lead to intracytoplasmic Ca2+ oscillation (Holland et al., 1999; Torrecilla et al., 2000). This oscillation could also occur when the extracellular Ca2+ is changed directly (Watkins et al., 1995; Jones et al., 1999). The different concentrations of extracellular Ca2+ could produce different oscillatory statuses (Herbaud et al., 1998; Allen et al., 2001). It may be speculated that the different status of cytoplasmic Ca2+ oscillation would also act as the cell signal to produce the different cell responses or cell-guarded status in LAB.

Constant temperature period

The survival and enlargement of regrowth lag time of LZY and ST during the constant temperature period in the heat treatment II are shown in Fig. 5a and b, respectively. The survival of all the samples was reduced as the heating time increased. For LZY, the sample with 5 mM CaCl2 maintained the highest N/N0 of all the samples throughout the constant temperature period. The sample with 1 mM CaCl2 showed the next highest N/N0. For ST, 1 mM and 5 mM CaCl2 both showed significant improvements in cell heat resistance. In contrast, the samples containing up to 10 mM CaCl2 had log (N/N0) values lower than the control. The results of the λ–λ0 values after heat treatment II also followed the trend of the viability which is displayed by the cell survival (log N/N0).

Figure 5.

Viability of (a) LZY and (b) ST in constant temperature period of heat treatment II. The solid symbols refer to the left log (N/N0) and the hollow symbols to the right λ–λ0, the dotted lines indicate the λ which is longer than the tested time (36 h).

The λ–λ0 value of all the strains corresponded well to survival in this study. This might suggest that the ‘apparent λ’ may be related to the inoculum size or, more accurately, to the inoculum size of the viable cells (Augustin et al., 2000). In addition, 100- and 1000-mM CaCl2 samples both showed the obvious negative effects on LZY and ST in heat treatment II. This result differs from a previous study which found that the heat resistance of LAB cells could be enhanced by suspending the cells in a 300-mM NaCl solution before a heat treatment II regime (Desmond et al., 2001). Possibly, the Ca2+ effect may differ from the cross-protection induced by the salt stress.

In addition, 5 mM Ca2+ showed a generally increased heat resistance of LAB in this work. It is interesting that the milk contains approximate 2 mM free Ca2+ (Holt et al., 1981; Holt & Jenness, 1984). The RSM, which conveyed the most effective protection for microorganisms in the previous studies, were commonly prepared to solid concentrations of 11–20% (w/v), which is generally more concentrated than the original milk (Fu & Chen, 2011). In addition, according to the product specifications, the RSM obtained from commercial operations often have higher levels of calcium (Reddy et al., 2009). Therefore, the concentration of free Ca2+ in the RSM may be reasonably close to the 5 mM found in this study. It follows that the interactions between Ca2+ and milk proteins in RSM during heating should be studied for their individual and combined roles in the heat resistance of bacterial cells.

In conclusion, the effect of Ca2+ on the heat resistance of LAB has been found to be very significant and could be one of the primary reasons for the adequate protective capability displayed by RSM. This positive result has opened doors for further investigations on potential wall materials with the addition of Ca2+ which may be more cost-effective. The significant effects found in the current work also warrant more scientific and instrumentally intensive investigation at below the cell level.


The authors sincerely thank the undergraduate students Miss Xiaofen Zhou, Yi Yang and Qing Qin, Xiamen University, for their technical assistances. Thanks to Prof. Heping Zhang, Inner Mongolia Agricultural University, and Dr I-Son Ng, Xiamen University, for providing the LAB strains.