Probiotics and immunosenescence: cheese as a carrier

Abstract

Oral intake of specific probiotics has been reported to enhance the immunity of the elderly. Earlier studies have used milk or yoghurt as a probiotic carrier. We chose a commercial probiotic cheese to evaluate its potential as a probiotic food. Thirty-one healthy elderly volunteers (21 female, 10 male) aged from 72 to 103 (median 86) consumed a commercial probiotic cheese containing approximately 10⁹ CFU day⁻¹ of Lactobacillus rhamnosus HN001 and Lactobacillus acidophilus NCFM. The 4-week probiotic intervention was preceded by a 2-week consumption of probiotic-free cheese (run-in) and followed by a 4-week wash-out period with the same control cheese. The cytotoxicity of peripheral blood mononuclear cells (PBMCs), the relative numbers of natural killer (NK) and NKT cells in the total PBMCs, and phagocytic activity were assessed. Consumption of the probiotic cheese significantly increased the cytotoxicity of NK cells. A significant increase in phagocytosis was observed for both the control and the probiotic cheese. Cheese was found to be an effective carrier for the study of probiotics, and daily consumption of the probiotic enhanced parameters of innate immunity in elderly volunteers. It remains to be determined whether this enhancement correlates with a beneficial effect on the health of the elderly population.

probiotics, immunity, cheese, elderly

Introduction

Immunosenescence refers to age-related changes in the immune system. The main alterations in the immune system with age include reduced humoral responses after vaccination or infection, decreases in dendritic cells efficiency to activate T and B cell populations, declines in the generation of new naive T and B cells, and in natural killer (NK) cells' ability to kill tumor cells (Aw et al., 2007; Hakim & Gress, 2007; Candore et al., 2008). Because of these changes, the frequency and severity of infectious disease, chronic inflammatory disorders, autoimmunity, and cancer incidence are hallmarks for immunosenescence (Gomez et al., 2008; Provinciali, 2009).

The increase in the proportion of aged individuals globally and especially in western countries (World Population Prospects, 2008) necessitates the search for innovative strategies to thwart the effects of immunosenescence. These strategies should be focused on preventing the deviations or restoring the function of the immune system in older individuals. Interventions such as specific vaccination against viruses, anti-inflammatory treatments, nutrition interventions, exercise, and pre- and probiotic have been suggested to restore the immune functions in the elderly (Candore et al., 2008; Mocchegiani et al., 2009).

The gastrointestinal tract is the main entry for bacterial cells through foods and drinks and is the site for presenting millions of antigens to the gut-associated-lymphoid tissues, which contains 70% of the immunoglobulin-producing cells. The intake of specific probiotic bacteria has been reported to enhance the immune response in a strain-specific manner (Nova et al., 2007; Borchers et al., 2009).

Earlier studies have reported that specific strains of lactic acid bacteria have immune-enhancing properties (Nova et al., 2007; Candore et al., 2008). However, probiotic bacteria have often been assessed in milk, fermented milks, or as dietary supplements. Therefore, we decided to investigate the effect of a commercial probiotic cheese containing Lactobacillus rhamnosus HN001 and Lactobacillus acidophilus NCFM on the nutritional modulation of immune parameters in older volunteers (70+). These would also serve as a model for immune compromised subjects. The aim of the current study was to determine whether specific probiotic bacteria in a cheese matrix would have immune-enhancing effects on healthy older individuals in a nursing home setting, similar to those reported earlier (Gill et al., 2000; Gill et al., 2001; Sanders & Klaenhammer, 2001).

Materials and methods

Study subjects

Elderly subjects without acute illness, 21 women and 10 men, age range from 72 to 103 (median, 86), residing in a nursing home for older individuals were recruited for this study (Table 1). The baseline data regarding the prevalence of disease and the use of medications are shown in Fig. 1. Dementia and cardiovascular disease were the most common conditions, while aspirin, diuretics, and calcium/vitamin D intake were the most frequent. As every subject was used as a self-control, this was not expected to affect the outcome. Either the volunteer or a relative gave their written informed consent, and the study was approved by the ethical committee of Hospital District of Southwest Finland. Exclusion criteria were the consumption of antibiotics in the last month and use of medication expected to either affect the immune function and/or the intestinal microbiota of the subject. Another exclusion criterion was the habitual use of pro- and/or prebiotic-containing products.

Table 1

Profiles of the 31 elderly volunteers

	Age (years)	Weight (kg)	Height (cm)	BMI
Mean	85	69	164	25
Median	86	68	164	25
Minimum	72	43	140	18
Maximum	103	101	186	35

BMI, body mass index.

Figure 1

The baseline data of the volunteers regarding the prevalence of disease and the use of medications.

Study design

The study protocol consisted of three consecutive phases. In phase 1, the subjects consumed a control cheese during breakfast for 2 weeks (run-in). In phase 2, the subjects consumed a probiotic cheese for 4 weeks (intervention). In phase 3, the subjects consumed the same control cheese again for 4 weeks (wash-out). The products were blinded to the volunteers and were identical in taste and appearance. The total duration of the study was 10 weeks, and during the time, the food at the elderly home remained stable. Heparinized peripheral blood (9 mL) was drawn by a venipuncture from each subject at baseline (T0), after run-in (T1), after intervention (T2), and after wash-out (T3) for immunological analysis. On the same occasion, a blood sample was collected for general health monitoring tests carried out at the University of Turku Hospital.

Test products

The probiotic and the control Gouda cheese were commercial products (Mills DA, Oslo, Norway). Identical slices of both types of cheese (15 g) were prepared and packed before the commencement of the study. The probiotic cheese slice contained approximately 10⁹ CFU of L. rhamnosus HN001 (AGAL NM97/09514) and L. acidophilus NCFM (ATCC 700396). The viability of the strains was assessed throughout the study and was observed to remain stable. Both probiotic and control cheese contained proprietary starter strains (Choozit 712™, Danisco, Paris). The volunteers consumed one slice of cheese per day during breakfast. The probiotic cheese had been on the Norwegian market for approximately 1 year. The probiotic strains have been in commercial use for approximately 7 years (L. rhamnosus HN001) and 30 years (L. acidophilus NCFM) and have substantial safety and efficacy data (Shu et al., 1999; Zhou et al., 2000; Gill & Rutherfurd, 2001; Sanders & Klaenhammer, 2001; Sheih et al., 2001). The same probiotic cheese was tested for bacterial survival using a human gastrointestinal tract-simulating model, and it was shown that the strains (L. acidophilus NCFM and L. rhamnosus HN001) survived the simulated upper gastrointestinal tract (Makelainen et al., 2009).

Immunological analyses

The cytotoxicity of the peripheral blood mononuclear cells (PBMCs), proportions of CD3−CD56+ cells (NK cells), CD3+CD56+ cells (NKT cells), CD3+CD56− cells, and CD3−CD56− cells in the total PBMCs, and phagocytic activity were assessed using flow cytometry (FACScan flow cytometer, BD biosciences). The data were analyzed using cellquest pro software.

The extraction of PBMCs from the whole blood was performed over Ficoll-paque gradient centrifugation (Ficoll-Paque™ PLUS, GE Healthcare). The cells were counted using the Trypan blue exclusion test and adjusted to 1 × 10⁶ cells mL⁻¹ in RPMI 1640 Complete (RPMI 1460+Glutamax™-I, 10% fetal calf serum, and 100 IU mL⁻¹ penicillin, and 100 µg mL⁻¹ streptomycin).

NK cell activity was assessed as described earlier (Johann et al., 1995). In brief, nonadherent K562 myeloid leukemia cells (NK-sensitive cell line, ECACC) were used as target cells (25 : 1 effector : target ratio). The K562 cells were incubated for 20 min at 37 °C (5% CO₂) with DiOC₁₈ (3), 3 mM in DMSO (Invitrogen), subsequently washed twice with phosphate-buffered saline (PBS), and suspended in RPMI 1640 Complete (4 × 10⁴ cells mL⁻¹). PBMCs (100 µL, 10⁶ cells mL⁻¹) were mixed with 100 µL DiOC₁₈ (3)-labelled K562 (4 × 10⁴ cells mL⁻¹) in 12 × 75 mm flow cytometer tubes. The samples were centrifuged at 200 g for 30 s and incubated for 4 h, at 37 °C (5% CO₂). Propidium iodide (50 µL, 100 µg mL⁻¹) was added to the samples before the flow cytometric analysis:

The proportions of different lymphocyte subsets in the total PBMCs were identified using specific fluorescein-conjugated monoclonal antibodies (Morimoto et al., 2005). PBMCs (100 µL, 10⁶ cells mL⁻¹) were mixed with 20 µL FITC-conjugated Mouse Anti-Human CD3 mAb and 20 µL PE-conjugated Mouse Anti-Human CD56 mAb (BD Pharmingen™) and incubated on ice for 30 min and washed twice with PBS (1 mL, 350 g, 5 min). The samples were suspended in 500 µL PBS and left in the dark on ice until FACS analyses, which were performed within two hours.

The phagocytosis activity was evaluated according to the protocols of the pHrodo™Escherichia coli BioParticles Phagocytosis kit for flow cytometry (Molecular Probes, Cat# A10025, Invitrogen).

To assess the health status of the patients during the course of the study, the following general health parameters were determined on the same sampling day for the immunological tests: white blood cell count, erythrocyte count, hemoglobin, hematocrit, average red blood cell size, hemoglobin amount per red blood cell, platelet count, total cholesterol, potassium, sodium, creatinine, albumin, high-density lipoprotein (HDL) cholesterol, C-reactive protein (CRP), and glycosylated hemoglobin.

Statistical analysis

Sample size estimation based on previous studies showed that 16 subjects are needed to achieve equal mean difference to that obtained in earlier studies with the same strains using supplemented milk.

The changes in the immune parameters over time were analyzed using mixed-model anova in the statistical analysis system (Proc Mixed, sas 9.1).

The test was carried out on the transformed variable (BoxCox transformation) to normalize the error part of the model. The Tukey–Kramer adjusted paired t-test was used for evaluating the differences between all sets of time points. The Pearson correlation coefficient was used to test for correlations. The percentages of phagocytotic granulocytes and monocytes were not amenable to transformation and Friedman anova for repeated measures, followed by the Wilcoxon signed ranks test for paired comparisons were used. P-values <0.05 were considered significant.

Results

The mean cytotoxicity of PBMCs increased significantly from 21.69% at the baseline to 29.96% by the end of the intervention (Fig. 2; P=0.014). The mean cytotoxicities after the run-in (24.17%) and wash-out (20.72%) were not significantly different from the baseline, but they were significantly different compared with the intervention (P=0.047 and <0.001, respectively). The control cheese, which also contains starter strains, did not have a significant effect on the cytotoxicity. There was a significant negative correlation between the magnitude of change in the cytotoxicity after the intervention and the baseline level (ρ=0.66, P<0.001).

Figure 2

Changes in the cytotoxicity of PBMCs of 31 elderly volunteers residing in the same nursing home during a 10-week intervention study. The values represents the mean cytotoxicity at the baseline (T0), after 2 weeks of consumption of a control cheese (T1), after 4 weeks of consumption of a probiotic cheese (T2), and after 4 weeks of consumption of the same control cheese (T3). The probiotic cheese was supplemented with Lactobacillus rhamnosus HN001 and Lactobacillus acidophilus NCFM. Error bars indicate the SE of the mean. The values assigned with different letters are significantly different at a 0.05 level.

The relative numbers of lymphocyte subsets appeared to be slightly modulated during the course of the study. A significant reduction in CD3−CD56− cells was observed after the run-in period compared with the baseline (P=0.008) and compared with the wash-out period (P=0.022). This reduction continued during the intervention and increased after the wash-out period to a level similar to that at the baseline (P=0.62). On the other hand, there was no significant modulation in the other types of lymphocyte subsets measured in this study (Fig. 3).

Figure 3

Changes in the relative proportions of lymphocyte subsets in the PBMCs of 31 elderly volunteers residing in the same nursing home during a 10-week intervention study. The values represents the mean cytotoxicity at the baseline (T0), after 2 weeks of consumption of a control cheese (T1), after 4 weeks of consumption of a probiotic cheese (T2), and after 4 weeks of consumption of the same control cheese (T3). The probiotic cheese was supplemented with Lactobacillus rhamnosus HN001 and Lactobacillus acidophilus NCFM. Error bars indicate the SE of the mean. The values assigned with different letters are significantly different at a 0.05 level.

There was no significant correlation between the cytotoxicity after the intervention and any of the lymphocyte subsets. However, when the data were analyzed as a whole, significant correlations, although weak, were found between the cytotoxicity values and the relative numbers of CD3−CD56+ cells (ρ=0.28, P=0.002), CD3+CD56+ cells (ρ=0.18, P=0.044), CD3+CD56− cells (ρ=0.28, P=0.001), and CD3−CD56− cells (ρ=−0.32, P<0.001).

The granulocyte and monocyte phagocytic activity were separately identified using forward and side scatters in a FACScan flow cytometer. Phagocytosis activity was expressed as the mean fluorescence intensity (Table 2). From these results, it is shown that there is a significant increase in both granulocyte and monocytes phagocytic activity after the consumption of control cheese compared with the baseline (P<0.001 for each). In addition, there was a significant increase in granulocyte and monocyte phagocytic activity upon consumption of probiotic cheese compared with the run-in (P<0.01 for each) and compared with the wash-out period (P <0.01 for each). Furthermore, the percentages of phagocytotic cells were also enhanced in a similar manner as the phagocytic activity (Table 2). The percent of phagocytic cells was significantly correlated with the phagocytic activity (ρ=0.37, P=0.040; ρ=0.78, P<0.001 for granulocytes and monocytes, respectively).

Table 2

Effect of probiotic cheese consumption on the phagocytosis activity of elderly populations

	Phagocytosis activity		Percentage of phagocytotic cells
	Granulocytes	Monocytes	Granulocytes	Monocytes
T0	273a (21.9)	202a (14.5)	28a (5.17)	14a (3.0)
T1	551b (32.6)	402b (28.2)	62b (6.53)	44b (5.9)
T2	685c (23.2)	964c (39.5)	91c (1.72)	93c (0.9)
T3	573b (25.9)	813d (43.6)	76b (4.95)	79d (5.1)

Phagocytosis activity was expressed as the mean fluorescence intensity. The values assigned with different letters are significantly different at a 0.05 level.

The values represent the mean and SE of the phagocytosis activity and percentage of phagocytotic cells in the peripheral blood polymorphonuclear cells at the baseline (T0), after 2 weeks of consumption of a control cheese (T1), after 4 weeks of consumption of a probiotic cheese (T2), and after 4 weeks of consumption of the same control cheese (T3).

The general health parameters were within the physiological ranges during the course of the study and no significant changes were observed (results not shown). However, the mean CRP values after the run-in period were significantly higher than that at the baseline level, but there were no significant differences among all other multiple comparisons (the median values were 2.5, 3.6, 2.4, and 2.9 for T0, T1, T2, and T3, respectively).

Discussion

In this study, the volunteers were all selected to be above 70 years of age as a model of immune-compromised subjects. Furthermore, all volunteers were living in the same elderly home. This was expected to reduce differences in the diet and environmental conditions, leading to reduced inter-individual variability during the study.

As shown in this study, the probiotic combination tested showed a significant improvement in NK cell ability to kill target tumor cells and the phagocytosis activity of granulocytes and monocytes. This may be of practical benefit to the health of the elderly population. A previous study reported an enhancement of immune parameters to be more pronounced in volunteers aged 70 years or more (Gill et al., 2001). Our results support the earlier studies demonstrating an enhancement of natural and acquired immunity indices in mice and in elderly populations (Gill et al., 2000; Gill et al., 2001; Sanders & Klaenhammer, 2001). In addition, this study verified that the reported enhancement of immune indices could also be achieved when the probiotic bacteria are embedded in a cheese matrix, while earlier studies used reconstituted fat-free milk as a carrier (Gill et al., 2000; Gill et al., 2001; Sheih et al., 2001).

In the present study, there was no significant association between the probiotic-induced enhancement of cytotoxicity and any of the lymphocyte subsets. This is in accordance with the observations by Gill and colleagues (Gill et al., 2001; Morimoto et al., 2005; Takeda & Okumura, 2007) with elderly volunteers. On the other hand, no significant correlation was found between the increase of NK cytotoxicity after the intervention and age in contrast to that observed by the authors (Gill et al., 2001). The significant negative association between the cytotoxicity values after the intervention and that at the baseline indicates that the increase of cytotoxicity is higher for volunteers with lower baseline cytotoxicity. This suggests that the consumption of these probiotics may benefit mostly those with reduced immune functions.

Because the significant reduction in the relative proportion of the CD3−CD56− level after the run-in and the intervention was not accompanied by a significant increase in at least some of the other cell types (CD3−CD56+, CD3+CD56+, and CD3+CD56−), this shows that the expected increase was distributed between those three types of cells. The weak, but significant, association between the cytotoxicity vs. NK, NKT, and CD3+CD56− cells indicates that these cells may be the main contributors to the cytotoxicity observed. On the other hand, because no substantial changes in the relative numbers of lymphocytes subsets have been observed, cytokine-activated NK cells' activity may be a more probable pathway. Significant production of interleukin-12 in the human PBMCs was observed after oral administration of Lactobacillus casei spp. casei and L. casei Shirota (Ogawa et al., 2006).

The augmentation of phagocytosis activity and the percentage of phagocytotic cells after the probiotic intake compared with the other time points demonstrated efficient enhancement of innate immunity in an elderly population after 4 weeks of probiotic cheese consumption. Additionally, the increase in phagocytosis activity related to the consumption of control cheese indicates that the starter strains also have immune stimulation properties at least for the phagocytosis. The increase in phagocytosis activity might play a role in the observed enhancement of NK cytotoxicity as it has been reported that the phagocytosis of bacteria by monocytes provides an additional signal on accessory cells inducing NK cell activation (Haller et al., 2002). NK cells' activity is known to be important for immune surveillance against cancer cells and pathogenic infection. The incidence of cancer and the rate of mortality were reported to be higher in populations with a low NK activity compared with those with higher NK activities (Morales & Ottenhof, 1983; Imai et al., 2000; Ogata et al., 2001). Moreover, phagocytosis measurement is a useful tool in the assessment of macrophage function in immunotoxicological and immunopharmacological evaluations (Musclow et al., 1991). However, with the present findings, further studies are needed to investigate whether there is an association of this size effect of immune modulation with clinical benefits.

The general health parameters for the subjects were within the physiological ranges throughout the course of the study. Although the mean values for erythrocytes, hemoglobin, hematocrit, and % HDL cholesterol were slightly lower after the probiotic intake, they were all within the normal ranges and were not significantly different from the baseline or the wash-out values. The two individuals with high CRP values (43.2 and 50.9 mg L⁻¹) were suffering from urinary tract and respiratory infection, respectively. The values influenced the mean after the consumption of the control cheese so that a significant difference was observed between the baseline and the run-in. A recent study (Hostmark et al., 2009) reported that cheese intake was negatively associated with triacylglycerols and HDL cholesterol. The amount of cheese consumed in this study was constant throughout the study and no correlation could be found between the amount of cheese consumption and the serum lipids. Considering that there were no significant changes in the total cholesterol or the HDL cholesterol level during the study, and the values were in the normal ranges, there seems to be no risk associated with the amount of cheese consumed. However, these values are worthwhile monitoring in future studies when cheese is used as a probiotic carrier.

In conclusion, the present study demonstrates that cheese with L. rhamnosus HN001 and L. acidophilus NCFM may be beneficial in improving the immune response of healthy elderly subjects. This may have application in the modulation of the diet of elderly individuals to improve their immune response against harmful external challenges. However, further studies are needed to investigate whether this immune stimulation is associated with a significant effect on the health of the elderly population.

Author notes

Editor: Patrick Brennan