Food &

Function
View Article Online
PAPER View Journal | View Issue

The effects of OLL1073R-1 yogurt intake on

Published on 24 October 2019. Downloaded by Vanderbilt University Library on 1/2/2020 9:02:19 PM.

Cite this: Food Funct., 2019, 10, 8129

influenza incidence and immunological markers
among women healthcare workers: a randomized
controlled trial†
Tetsu Kinoshita, *a,b Koutatsu Maruyama, c Keiko Suyama,a Mariko Nishijima,a
Kimiko Akamatsu,d Akiko Jogamoto,d Kikumi Katakamid and Isao Saito e

Probiotics have been expected to enhance human immune function. The aim of this study was to evalu-
ate the effects of dietary intake of yogurt fermented with Lactobacillus delbrueckii ssp. bulgaricus
OLL1073R-1 (OLL1073R-1) on the prevention of influenza during winter and on the activation of immuno-
logical markers among women healthcare workers. 961 women aged 20–71 years were randomly
assigned to either the yogurt group (n = 479) or the control group (n = 482). Participants in the yogurt
group consumed a 112 mL yogurt drink fermented with OLL1073R-1 every day for 16 weeks, whereas
those in the control group consumed no yogurt during this period. All participants were instructed not to
consume any other kinds of yogurt or fermented dairy products throughout this trial. The cumulative inci-
dence rate of influenza was measured, and immunological markers were examined at the baseline and
after 16 weeks. No significant difference in the incidence rate of influenza was found between the two
groups (cumulative incidence rates of flu: yogurt 7.5% and control 7.7%). Natural killer (NK) cell activity did
not show a significant intervention effect ( p = 0.11), whereas the intervention effect on serum interferon
Received 12th September 2019, gamma (IFN-γ) production was significant ( p = 0.03). Other immunological markers did not show signifi-
Accepted 22nd October 2019
cant intervention effects. Consumption of OLL1073R-1 yogurt did not show a significant preventive effect
DOI: 10.1039/c9fo02128k against influenza or a significant enhancement in NK cell activity. However, intake of this yogurt showed
rsc.li/food-function an increase in IFN-γ production.

1. Introduction via lactic acid-producing bacteria or substances produced from

the bacteria.3,4 However, robust evidence that dietary intake of
Yogurt, which is a fermented dairy product, is generally known a fermented dairy product contributes to a reduced risk of
to have promising health benefits for certain gastrointestinal infectious diseases or to enhanced immune cell activity is
conditions, including lactose intolerance, constipation, inflam- lacking. Previous literature is limited due to differences in the
matory bowel diseases, and Helicobacter pylori infection.1,2 In strains of lactic acid-producing bacteria studied or due to
recent studies, yogurt has been expected to enhance human study design (i.e., small-scale sample size, the nature of the
immune function through changing the balance of the intesti- study population, and investigating only subjective symptoms).
nal microbiota and stimulating the intestinal immune system Most notably, no functional food has been shown in random-
ized controlled trials to enhance the activity of natural killer
(NK) cells in humans, which are considered major innate
a
Department of Community Health System Nursing, Ehime University Graduate immune cells combating against pathogenic viruses and
School of Medicine, Toon, Ehime, Japan. E-mail: [email protected]; tumor cells.5,6
Fax: +81-89-960-5284; Tel: +81-89-960-5283
b A previous study on yogurt fermented with Lactobacillus del-
Institute of Community Life Sciences Co., Ltd., Matsuyama, Ehime, Japan
c
Special Course of Food and Health Science, Department of Bioscience Graduate brueckii ssp. bulgaricus OLL1073R-1 (OLL1073R-1) showed that
School of Agriculture, Ehime University, Matsuyama, Ehime, Japan this strain reduced the risk of respiratory infections in the
d
Department of Fundamental and Clinical Nursing, Ehime University Graduate elderly.7 In animal experiments, the exopolysaccharides (EPSs)
School of Medicine, Toon, Ehime, Japan produced from OLL1073R-1 had anti-influenza virus effects
e
Department of Public Health and Epidemiology, Faculty of Medicine, Oita
and enhanced NK cell activity.8–10
University, Yufu, Japan
† Electronic supplementary information (ESI) available. See DOI: 10.1039/ Therefore, the preventive effect of OLL1073R-1 yogurt intake
C9FO02128K against influenza infection during winter could be elucidated

This journal is © The Royal Society of Chemistry 2019 Food Funct., 2019, 10, 8129–8136 | 8129
 View Article Online

Paper Food & Function

via the enhancement of immune cell activity using an appro- consume any yogurt or fermented dairy foods from the day of
priate study design. Given that women healthcare employees agreement to that of random assignment. The 961 women
are exposed to infectious diseases during shift work and thus were assigned to either the yogurt group (n = 479) or the
their immune function might be deteriorated by their irregular control group (n = 482) by block randomization within the
lifestyles,11 they were chosen as subjects for this trial. strata of the institutions they worked for, age, and having a
plan of influenza vaccination or not. The yogurt group con-
sumed the test yogurt daily for 16 weeks (from November 14,
2. Methods 2016, to March 5, 2017), whereas the control group consumed
Published on 24 October 2019. Downloaded by Vanderbilt University Library on 1/2/2020 9:02:19 PM.

no yogurt during this period. Furthermore, all participants

2.1. Power analysis and sample size
were instructed not to consume any other yogurt or fermented
In this randomized controlled study, sample size was calculated dairy products throughout this trial. In addition, they were
based on the incidence rates of influenza infection in Japanese directed not to change their lifestyles during this trial except
observational studies. Studies conducted in the 2013–2014 and for yogurt intake. The participants underwent blood tests at
2014–2015 seasons reported influenza infection rates of 5.1% the baseline and after 16 weeks. In addition, participants were
and 7.0%, respectively, among pregnant Japanese women.12,13 given a “health notebook” to record any incidence of influenza
In 2013–2014, the infection rate of vaccinated women was lower and the common cold, changes in their health condition,
than that of non-vaccinated women.12 However, in 2014–2015, changes in their lifestyles, intake of the test yogurt or not (for
the overall infection rate did not differ significantly between the yogurt group), and intake of any other yogurt or fermented
those with and without vaccination (7.3% vs. 6.7%, respect- dairy foods or not (for both groups) during the trial period.
ively).13 The contract rate of the control group was hypothesized Whether the participants received influenza vaccination or not
to be 6.5% during this winter trial and 2.6% for the yogurt was left to their individual discretion. If they planned to vacci-
group, based on a previous clinical trial.7 Additionally, assum- nate, they were prompted to receive vaccination between the
ing an alpha error of 5%, beta error of 20%, and dropout rate baseline blood test and the first day (Nov. 14, 2016) of test
during the trial of approximately 20%, the number of partici- food consumption. During the trial, four participants in the
pants was calculated to be 600 in each group. yogurt group and 12 in the control group declined partici-
pation. In the control group, two were absent from the examin-
2.2. Participants
ation session and one did not answer the self-administered
Participants were recruited from 23 medical institutions in questionnaires 16 weeks later. In addition, two participants in
Ehime Prefecture, Japan, from August to September, which are the yogurt group and four in the control group revealed that
not the months of flu season, 2016. Women aged 20 or older they were pregnant during the trial, thereby infringing the
who were currently medical or welfare-related professionals at exclusion criteria. Fig. 1 shows the sampling scheme through-
medical institutions in Ehime Prefecture and were capable of out this study.
understanding the study purpose were selected, with written
informed consent provided by all participants. The exclusion 2.4. Test yogurt
criteria were as follows: (1) pregnant women, (2) women who The test food was “Meiji Probio Yogurt R-1” drink type (Meiji
contracted influenza during the period of July 2016 to the date Co., Ltd, Tokyo), which is currently on the market. This yogurt
of giving written informed consent, (3) women who show aller- is manufactured using two lactic acid bacterial species,
gic responses to dairy products, (4) lactose intolerant women, L. bulgaricus OLL1073R-1 and a strain of Streptococcus thermo-
(5) women who are instructed to restrict calories by a phys- philus, originally isolated from traditional Bulgarian yogurt.
ician, (6) women with a history of diseases involving the One bottle contains 112 mL drinkable yogurt and provides
immune system (e.g., rheumatism, cancer, thyroid disorder, 76 kcal, 13.9 g carbohydrate, 0.67 g fat, 3.6 g protein, and 1.12
systemic lupus erythematosus, myasthenia gravis, Graves’ × 109 CFU (colony forming unit) or more of L. bulgaricus and
disease, and scleroderma), (7) women who had participated in S. thermophilus. The total count of lactic acid bacteria is con-
other clinical trials within the past three months, and (8) firmed by colony counting using a culture medium in which
women who were judged inadequate by the principal doctor both L bulgaricus and S. thermophilus can grow. Lactic acid bac-
for other reasons. Of the 1026 women who agreed to partici- teria in the test drink were counted after culture in bromocre-
pate in this study, 20 infringed the exclusion criteria and 24 sol purple broth at 35–37 °C for 72 ± 3 h. Although these
declined to participate after the agreement. Therefore, 982 culture conditions are not the most appropriate for the growth
women were examined at the screening session. During the of both bacterial species, both can be detected and that count-
session, 20 women infringed the exclusion criteria and one ing method is universally used.
declined participation. Consequently, 961 women aged 20–71
years were enrolled in the present study. 2.5. Outcome measurement
2.5.1. Cumulative incidence rates of influenza and the
2.3. Study design common cold. The primary objective of the present study was
This trial was designed as a randomized, controlled, open- to examine the cumulative incidence rate of influenza within
labeled study. Participants were first instructed not to the two groups. The participants were instructed to record

8130 | Food Funct., 2019, 10, 8129–8136 This journal is © The Royal Society of Chemistry 2019
 View Article Online

Food & Function Paper

Published on 24 October 2019. Downloaded by Vanderbilt University Library on 1/2/2020 9:02:19 PM.

Fig. 1 Sampling scheme throughout this study.

influenza incidence in their health record only if diagnosed by medical institution and examined within the sampling day.
a physician. The common cold was defined as having a fever of Blood sampling was conducted in the morning (before starting
38 °C or more, accompanied by one or more of the following their work) after 12 hours of fast considering the diurnal vari-
symptoms: cough, sputum, throat pain, snivel, joint pain, ation of blood markers. The indices were NK cell activity,
headache, and diarrhea with abdominal pain. Cumulative inci- serum high-sensitivity C-reactive protein (hs-CRP), serum
dence rates of influenza and the common cold were deter- interleukin (IL)-2, IL-4, IL-5, IL-10, IL-12, IL-13, granulocyte-
mined using the health records during the trial period (from macrophage colony stimulating factor (GM-CSF), interferon
November 14, 2016, to March 5, 2017). However, the first week gamma (IFN-γ), and tumor necrosis factor-α (TNF-α). NK cell
of the trial was not subject to evaluation. activities were assessed by the 51Cr release method targeting
2.5.2. Immunological markers. Blood tests examining K562 cells at Shikoku-Chuken Co. Ltd (Kagawa, Japan), hs-CRP
immunological markers such as cytokines were conducted at was gauged by latex nephelometry (N-latex CRP II, Siemens
the baseline and after 16 weeks. Blood was sampled in each Healthcare) at Shikoku-Chuken, and other serum immunologi-

This journal is © The Royal Society of Chemistry 2019 Food Funct., 2019, 10, 8129–8136 | 8131
 View Article Online

Paper Food & Function

cal biomarkers were examined by using a BioPlex14 at Meiji Table 1 The baseline characteristics of participants
Corporation. These markers were chosen based on the pre-
vious animal study,9 which highlighted the effects of Yogurt group Control group
(n = 479) (n = 482) P value
OLL1073R-1 strain on enhancing natural killer cell activation
and on induction of producing IFN-γ. IFN-γ is an important Age (years) 39.3 (11.5) 39.4 (11.4) 0.91*
BMI (kg m−2) 22.0 (3.5) 21.8 (3.5) 0.42*
factor in eliminating infected cells, and thus we evaluated the Carrier
levels of IFN-γ and other cytokines. When the measurement Nurse 310 (64.7%) 312 (64.7%) 0.62**
Care worker 43 (9.0%) 33 (6.9%)
values were below the detection limit, half of the lowest values Physical therapist/ 36 (7.5%) 40 (8.3%)
Published on 24 October 2019. Downloaded by Vanderbilt University Library on 1/2/2020 9:02:19 PM.

detected in this trial were applied. In this trial, we did not occupational therapist
examine nutritional indices, such as total protein or albumin, Others 90 (18.8%) 97 (20.1%)
Influenza vaccination
because this study focused on the changes in immunological Received 452 (94.4%) 457 (94.8%) 0.76**
functions. Did not receive 27 (5.6%) 25 (5.2%)
Irregular shift worker 220 (46.0%) 228 (47.3%) 0.67**
2.5.3. Safety. Adverse events occurring during this trial Living by herself 118 (24.6%) 120 (24.9%) 0.93**
period were evaluated for the two groups. Based on the health Current smoker 47 (9.8%) 43 (8.9%) 0.64**
record and the blood test after 16 weeks, the number of NK cell activitya (%) 26.8 (15.9–45.2) 26.7 (15.6–45.6) 0.90***
hs-CRPa (mg/dl) 0.031 (0.0086–0.11) 0.028 (0.0084–0.092) 0.21***
adverse events was counted and classified into 10 categories: IL-2a,b (pg ml−1) 0.82 (0.56–1.20) 0.80 (0.71–0.92) 0.33***
indefinite complaint (e.g. headache, fatigue, irritability, etc.), IL-4a,b (pg ml−1) 0.063 (0.016–0.25) 0.063 (0.015–0.26) 0.94***
IL-5a,b (pg ml−1) 1.16 (0.14–9.60) 1.20 (0.14–10.25) 0.83***
menstrual pain, digestive symptoms, body aches, skin symp- IL-10a,b (pg ml−1) 4.62 (1.66–12.82) 4.20 (1.49–11.88) 0.15***
toms, sleep and mental condition, allergy, respiratory symp- IL-12a,b (pg ml−1) 0.14 (0.017–1.14) 0.14 (0.017–1.21) 0.80***
toms, injury, and others. For the yogurt group, the principal IL-13a,b (pg ml−1) 2.12 (0.46–9.90) 2.20 (0.50–9.70) 0.71***
GM-CSFa,b (pg ml−1) 0.46 (0.33–0.65) 0.47 (0.31–0.70) 0.57***
doctor judged whether the adverse event had a causal relation- IFN-γa,b (pg ml−1) 2.69 (0.61–11.84) 2.65 (0.58–12.16) 0.85***
ship with the intake of the test yogurt. TNF-αa,b (pg ml−1) 0.75 (0.038–14.79) 0.74 (0.036–15.17) 0.94***

Values are expressed as means. Standard deviations or percentages are

indicated in parentheses. a Geometric mean values. Lower 1SD and higher
2.6. Statistical analysis 1SD are indicated in parentheses. b Two participants were excluded from
All statistical analyses were performed with SAS version 9.4 the analysis as they were unmeasured due to the lack of blood samples at
the baseline (Yogurt: n = 478 and Control: n = 481). *Wilcoxon’s rank sum
(SAS Institute Inc., Cary, NC, USA) based on intention-to-treat test. **Chi-square test. ***Unpaired t-test.
analysis. The difference in cumulative incidence rates of influ-
enza and the common cold between the two groups was ana-
lyzed by a chi-square test using the Cochran–Mantel–Haenszel
method. However, Fisher’s exact test was used when the 3.1. Cumulative incidence rates of influenza and the
numbers of incidence were less than five. The variable strata common cold
were as follows: the institutes, the yogurt group or control Fig. 2 indicates the Kaplan–Meier curves of the cumulative
group, and the outcome. If a participant dropped out during incidence rates of influenza (A) and the common cold (B) for
the trial period, influenza incidence was evaluated during the each group during the 16 week trial period. No statistically sig-
participation period since it was the primary objective, and the nificant difference was found between the curves of the yogurt
common cold was excluded from evaluation. and control groups.
For the adverse events, incidence rates in both groups were The total cumulative incidence rate of influenza was 7.6%
analyzed in 10 categories by a chi-square test. (73/961), and all incidences were attributed to the influenza A
In all analyses, p < 0.05 (two-sided) was considered statisti- virus. The rate was 7.5% (36/479) for the yogurt group and
cally significant. 7.7% (37/482) for the control group (Table 2). Between the two
groups, there was no significant difference in the cumulative
2.7. Ethics statement incidence rate of influenza ( p = 0.91). With respect to the
common cold, the cumulative incidence rate was 18.4% (87/
The present study was approved by the Institutional Review
473) for the yogurt group and 20.2% (94/466) for the control
Board (IRB) of Ehime University Hospital and the IRB of Meiji
group. Between the groups, a significant difference was not
Corporation. Written informed consent was obtained from all
found ( p = 0.49).
participants. This trial was registered at the clinical trial regis-
Table 2 shows the stratified analysis related to the cumulat-
try of Japan as UMIN000023932.
ive incidence rates of influenza. Twenty-seven (5.6%) partici-
pants in the yogurt group did not receive influenza vaccination
and five (18.5%) of them developed influenza, while 25 (5.2%)
3. Results participants in the control group did not receive influenza vac-
cination and three (12.0%) of them developed influenza.
Table 1 shows the baseline characteristics of the participants. During the 16 weeks, 87 (18.2%) participants from the yogurt
There were no significant differences identified between the group took a preventive agent for influenza (e.g., Tamiflu®,
two groups for any of the indices at the baseline. Relenza®, and Inavir®) in accordance with the rules of the

8132 | Food Funct., 2019, 10, 8129–8136 This journal is © The Royal Society of Chemistry 2019
 View Article Online

Food & Function Paper

cell activity increased in both groups (yogurt: 26.8% to 32.1%;

control: 26.7% to 33.4%), but the intervention effect was not
statistically significant ( p = 0.11). IFN-γ production increased
in both groups (yogurt: 2.69 pg ml−1 to 6.21 pg ml−1; control:
2.65 pg ml−1 to 4.70 pg ml−1), and a significant intervention
effect was found ( p = 0.03). The corresponding significant
increase in IFN-γ production was confirmed, in case that we
regarded the IFN-γ below the detection limit as zero. For the
Published on 24 October 2019. Downloaded by Vanderbilt University Library on 1/2/2020 9:02:19 PM.

other biomarkers, significant intervention effects were not

found.

Fig. 2 Kaplan–Meier curves of the cumulative incidence rates of 3.3. Safety

influenza (A) and the common cold (B) for each group during the 16 ESI table† shows the number of adverse events throughout this
week. The straight line denotes the yogurt group, while the dotted line
trial in each group. In the yogurt group, digestive symptoms
denotes the control group.
were observed significantly compared to the control group. No
serious side effects were considered to be caused by the
yogurt.
Table 2 Stratified analysis of the cumulative incidence rate of influenza

Yogurt group Control group P value

Total cumulative 36/479 (7.5%) 37/482 (7.7%) 0.91

4. Discussion
incidence of Influenza
This trial examined the effects of OLL1073R-1 yogurt intake on
Incidence in January–March, 33/479 (6.9%) 31/482 (6.5%) 0.82 the prevention of influenza and the common cold during
2017 (excluded the incidence winter and on the enhancement of immune function among
within 2016)
women who work in healthcare. Of the immunological para-
Influenza vaccination meters analyzed, IFN-γ production demonstrated a significant
Vaccinated 31/452 (6.9%) 34/457 (7.4%) 0.72 intervention effect.
Unvaccinated 5/27 (18.5%) 3/25 (12.0%) 0.59*
IFN-γ is secreted by NK cells with other cytokines and
Preventive agent for influenza enhances NK cell activity.10,15,16 Thus, IFN-γ is considered the
Took once or more 3/87 (3.5%) 6/91 (6.6%) 0.30* most critical cytokine for NK cell activity. Although IFN-γ
Did not take at all 33/392 (8.4%) 31/391 (7.9%) 0.77
might have enhanced NK cell activation to some extent,
Working shift various environmental factors, such as a change in tempera-
Irregular 13/220 (5.9%) 23/228 (10.1%) 0.10 ture with seasons,17,18 could have greatly influenced this
Day shift 23/259 (8.9%) 14/254 (5.5%) 0.08
increase. While IFN-γ levels were elevated in the OLL1073R-1
Living yogurt group, other cytokines such as GM-CSF and TNF-α were
By herself 10/118 (8.5%) 6/120 (5.0%) 0.35 not increased. Thus, it is unlikely that the elevated levels of
With family 26/361 (7.2%) 31/362 (8.6%) 0.47
IFN-γ were attributed to inflammation or infection by some
Smoking pathogens. The mechanisms of increased IFN-γ production
Current smoker 4/47 (8.5%) 4/43 (9.3%) 0.85* should be further investigated. It is known that T cells are a
Non smoker 32/432 (7.4%) 33/439 (7.5%) 0.94
major source of IFN-γ other than NK cells.19 Furthermore, pre-
Chi-square test using the Cochran–Mantel–Haenszel method. *Fisher’s vious studies have reported increased T cell activity stimulated
exact test. by OLL1073R-1 yogurt or extracellular polysaccharides pro-
duced from OLL1073R-1.7–10 In line with the association, the
consumption of the OLL1073R-1 yogurt might possibly stimu-
late IFN-γ production via T cell activation. It has been shown
institutions where they worked, and three (3.5%) of them that cell-mediated immunity evaluated by the production of
developed influenza. In the control group, 91 (18.9%) partici- IFN-γ, probably by influenza virus-specific T cells, contributes
pants took the preventive medicine and six (6.6%) developed to protection against clinical influenza infection in young chil-
influenza. In each stratum, no significant difference in the dren.20 Although we did not investigate whether it was
cumulative incidence rate of influenza was found between the antigen-specific, the increased production of IFN-γ might at
two groups. least partly contribute to protection against infection with
pathogens.
Although the preventive effect of OLL1073R-1 yogurt
3.2. Immunological markers against influenza was the primary objective, no significant
Table 3 shows the measurement values of immunological effect was observed in this trial. The cumulative incidence rate
markers at the baseline and after 16 weeks in each group. NK of influenza in the control group was 7.7%, which was 1.2%

This journal is © The Royal Society of Chemistry 2019 Food Funct., 2019, 10, 8129–8136 | 8133
 View Article Online

Paper Food & Function

Table 3 Changes in the immunological indicators during the trial period

Yogurt group (n = 479) Control group (n = 482)

Baseline 16 weeks Baseline 16 weeks P value*

NK cell activity (%) 26.8 (15.9–45.2) 32.1 (18.9–54.7) 26.7 (15.6–45.6) 33.4 (20.3–54.9) 0.11
hs-CRP (mg dl−1) 0.031 (0.0086–0.11) 0.031 (0.0086–0.11) 0.028 (0.0084–0.092) 0.028 (0.0078–0.097) 0.88
IL-2a (pg ml−1) 0.82 (0.56–1.20) 0.82 (0.54–1.24) 0.80 (0.71–0.92) 0.81 (0.68–0.95) 0.77
IL-4a (pg ml−1) 0.063 (0.016–0.25) 0.11 (0.021–0.60) 0.063 (0.015–0.26) 0.11 (0.021–0.54) 0.60
IL-5a (pg ml−1)
Published on 24 October 2019. Downloaded by Vanderbilt University Library on 1/2/2020 9:02:19 PM.

1.16 (0.14–9.60) 2.21 (0.32–15.15) 1.20 (0.14–10.25) 2.04 (0.30–14.02) 0.47

IL-10a (pg ml−1) 4.62 (1.66–12.82) 6.05 (2.27–16.11) 4.20 (1.49–11.88) 5.68 (2.26–14.28) 0.64
IL-12a (pg ml−1) 0.14 (0.017–1.14) 0.31 (0.029–3.19) 0.14 (0.017–1.21) 0.28 (0.029–2.75) 0.43
IL-13a (pg ml−1) 2.12 (0.46–9.90) 2.36 (0.57–9.71) 2.20 (0.50–9.70) 2.27 (0.51–10.15) 0.53
GM-CSFa (pg ml−1) 0.46 (0.33–0.65) 0.46 (0.33–0.65) 0.47 (0.31–0.70) 0.46 (0.34–0.64) 0.27
IFN-γa (pg ml−1) 2.69 (0.61–11.84) 6.21 (1.06–36.45) 2.65 (0.58–12.16) 4.70 (0.84–26.19) 0.03
TNF-αa (pg ml−1) 0.75 (0.038–14.79) 0.82 (0.034–19.68) 0.74 (0.036–15.17) 0.61 (0.025–15.04) 0.24

Values are expressed as geometric means. Standard deviations (±1SD) are indicated in parentheses. a Two participants were excluded from the
analysis as they were unmeasured due to the lack of blood samples at the baseline (Yogurt: n = 478 and Control: n = 481). *Two-way analysis of
variance with repeated measures.

higher than the hypothesized 6.5% from the power calcu- the randomization was conducted among institutions but not
lation. On the other hand, the rate in the yogurt group of 7.5% among individuals. The present trial was conducted using stra-
was quite higher than the hypothesized 2.6%. The Kaplan– tified randomization in a large-scale population. Furthermore,
Meier curves seemed to slightly show the preventive effects of the effect of OLL1073R-1 yogurt on increased IFN-γ production
OLL1073R-1 yogurt; thus, it is possible that if the present in women healthcare workers provided new evidence in the
study had gone on for longer (e.g. 24–32 weeks) then a signifi- field of nutritional immunology. Another strength is that this
cant effect may have been observed; however, a larger scale large-scale study was executed under the stringent condition
study would be required in order to detect a significant effect that the participants do not consume any other fermented
during 16 weeks. dairy products during the 16 week trial. Despite this restric-
The use of antiviral drugs for influenza (e.g., Tamiflu®, tion, the dropout rate was less than 3%, and over 900 partici-
Relenza®, and Inavir®) should affect the cumulative incidence pants completed the trial.
rate of influenza. In the yogurt group, 87 (18.2%) participants Nonetheless, this study has several limitations. First, the
took an influenza preventive agent, whereas 91 (18.9%) partici- participants were not blinded. Ideally, a placebo yogurt should
pants took an influenza preventive agent in the control group. In be used as the control food. However, since it was impossible
both groups, the participants who took the preventive medicine to prepare a placebo similar to the yogurt in appearance, taste,
showed low cumulative incidence rates (3.5% and 6.5%, respect- and flavor, an open-label design was used. As a result, if all
ively). Therefore, taking the preventive medicine seems to have participants were eager to be in the yogurt group, someone
contributed to influenza prevention. However, in the analysis assigned to that group might be motivated to complete the
among the participants who did not take the preventive medi- trial period, whereas a person assigned to the control group
cine, no significant difference in the cumulative incidence rate might feel some stress due to complete yogurt restriction.
was found. Thus, the preventive agents would not have affected Consequently, there is a possibility that the differences in the
the difference in the cumulative incidence rate of both groups. mood or stress levels between the groups might affect the
In the yogurt group, digestive symptoms were observed sig- results. Second, since all participants were instructed not to
nificantly compared to the control group. The symptoms were consume any fermented dairy products except for the test food
mainly loose stools and concentrated within 2 weeks from the for the yogurt group, there is a possibility that this rule intro-
start of yogurt intake. We consider that since they had not duced selection bias. In the informed consent session, some
eaten yogurt for more than one month after they agreed to par- participants refused to participate due to this restriction. To
ticipate in this study, a habituation period to the dairy product eliminate this influence, it might be preferable to choose par-
was needed for some participants. Those would mean yogurt ticipants who do not have a habit of probiotic intake in order
intolerance, not lactose tolerance, because lactose tolerant to evaluate the effects of yogurt.
women were excluded by participants’ criteria.
The strength of this study lies in the use of a randomized
controlled design and the sample size calculation based on 5. Conclusion
reported influenza infection rates12,13 in Japanese women.
Previous studies evaluating the effects of probiotics on the pre- This large-scale clinical trial among women healthcare
vention of influenza or enhancing immune response21–26 had workers did not show any effects of OLL1073R-1 yogurt intake
limitations in study design, including small-scale, inappropri- on influenza prevention and NK cell activity enhancement.
ate intake season, lack of basis in sample size calculation, or However, a significant increase in IFN-γ production was found

8134 | Food Funct., 2019, 10, 8129–8136 This journal is © The Royal Society of Chemistry 2019
 View Article Online

Food & Function Paper

with daily intake of OLL1073R-1 yogurt. The current findings dietary intake of yoghurt fermented with Lactobacillus del-
present novel evidence in the field of nutritional immunology brueckii ssp. bulgaricus OLL1073R-1, Br. J. Nutr., 2010,
and contribute to a better understanding of the benefits of 104, 998–1006.
probiotics. Further clinical trials, using an appropriate sample 8 T. Nagai, S. Makino, S. Ikegami, H. Itoh and H. Yamada,
size and a randomized controlled study design, are needed to Effects of oral administration of yogurt fermented with
verify the effects of fermented dairy products. Lactobacillus delbrueckii, ssp. bulgaricus OLL1073R-1 and
its exopolysaccharides against influenza virus infection in
mice, Int. Immunopharmacol., 2011, 11, 2246–2250.
Published on 24 October 2019. Downloaded by Vanderbilt University Library on 1/2/2020 9:02:19 PM.

Author contributions 9 S. Makino, A. Sato, A. Goto, M. Nakamura, M. Ogawa,

Y. Chiba, J. Hemmi, H. Kano, K. Takeda, K. Okamura and
The authors’ responsibilities were as follows: TK, KM, and IS Y. Asami, Enhanced natural killer cell activation by exopoly-
conceived and designed the study; TK, KS, MN, KA, AJ, and KK saccharides derived from yogurt fermented with
led and conducted research; KM and IS analyzed the data; TK, Lactobacillus delbrueckii, ssp. bulgaricus OLL1073R-1,
KM, an IS wrote the paper; and all authors read and approved J. Dairy Sci., 2016, 99, 915–923.
the final manuscript. 10 S. Makino, S. Ikegami, H. Kano, T. Sashihara, H. Sugano,
H. Horiuchi, T. Saito and M. Oda, Immunomodulatory
effects of polysaccharides produced by Lactobacillus del-
Conflicts of interest brueckii, ssp. bulgaricus OLL1073R-1, J. Dairy Sci., 2006,
89, 2873–2881.
The test yogurt was obtained from Meiji. Tetsu Kinoshita (first
11 Y. Morikawa, K. Kitaoka, C. Tanimoto, M. Hayashi,
author) is the head of the Institute of Community Life
R. Oketani, K. Miura, M. Nishijo and H. Nakagawa, A cross-
Sciences Co., Ltd; this study was conducted based on the con-
sectional study on the relationship of job stress with
tract agreement between Meiji and the Institute of Community
natural killer cell activity and natural killer cell subsets
Life Sciences. Meiji had no input in the data analysis and
among healthy nurses, J. Occup. Health, 2005, 47, 378–383.
interpretation. The other authors had no personal or financial
12 T. Yamada, K. Abe, Y. Baba, E. Inubashiri, K. Kawabata,
conflict of interest.
T. Kubo, Y. Maegawa, N. Fuchi, M. Nomizo, M. Shimada,
A. Shiozaki, H. Hamada, S. Matsubara, N. Akutagawa,
S. Kataoka, M. Maeda, H. Masuzaki, N. Sagawa, A. Nakai,
Acknowledgements S. Saito and H. Minakami, Vaccination during the
The authors thank all participants, doctors and chief nurses of 2013–2014 influenza season in pregnant Japanese women,
23 medical institutions for cooperation in experiments. This Eur. J. Clin. Microbiol. Infect. Dis., 2015, 34, 543–548.
study was supported by funding from the Meiji Corporation 13 T. Yamada, S. Kawakami, Y. Yoshida, H. Kawamura,
(Tokyo, Japan). S. Ohta, K. Abe, H. Hamada, S. Dohi, K. Ichizuka,
H. Takita, Y. Baba, S. Matsubara, J. Mochizuki, N. Unno,
Y. Maegawa, M. Maeda, E. Inubashiri, N. Akutagawa,
References T. Kubo, T. Shirota, Y. Oda, T. Yamada, E. Yamagishi,
A. Nakai, N. Fuchi, H. Masuzaki, S. Urabe, Y. Kudo,
1 O. Adolfsson, S. N. Meydani and R. M. Russell, Yogurt and M. Nomizo, N. Sagawa, T. Maeda, M. Kamitomo,
gut function, Am. J. Clin. Nutr., 2004, 80, 245–256. K. Kawabata, S. Kataoka, A. Shiozaki, S. Saito, A. Sekizawa
2 K. R. George, J. K. Patra, S. Gouda, Y. Park, H. S. Shin and and H. Minakami, Influenza 2014–2015 among pregnant
G. Das, Benefaction of probiotics for human health: A Japanese women: primiparous vs multiparous women,
review, J. Food Drug Anal., 2018, 26, 927–939. Eur. J. Clin. Microbiol. Infect. Dis., 2016, 35, 665–671.
3 S. N. Meydani and W. K. Ha, Immunologic effects of 14 E. K. Riccio, L. R. Pratt-Riccio, C. Bianco-Júnior,
yogurt, Am. J. Clin. Nutr., 2000, 71, 861–872. V. Sanchez, P. R. Totino, L. J. Carvalho and C. T. Daniel-
4 L. V. Hooper, D. R. Littman and A. J. Macpherson, Ribeiro, Molecular and immunological tools for the evalu-
Interactions between the microbiota and the immune ation of the cellular immune response in the neotropical
system, Science, 2012, 336, 1268–1273. monkey Saimiri sciureus, a non-human primate model for
5 E. Vivier, D. H. Raulet, A. Moretta, M. A. Caligiuri, malaria research, Malar. J., 2015, 14, 166.
L. Zitvogel, L. L. Lanier, W. M. Yokoyama and S. Ugolini, 15 S. H. Jang, M. S. Oh, H. I. Baek, K. C. Ha, J. Y. Lee and
Innate or adaptive immunity? The example of natural killer Y. S. Jang, Oral administration of silk peptide enhances the
cells, Science, 2011, 331, 44–49. maturation and cytolytic activity of natural killer cells,
6 S. Paul and G. Lal, The molecular mechanism of natural Immune Netw., 2018, 18, e37.
killer cells function and its importance in cancer immu- 16 R. Wang, J. J. Jaw, N. C. Stutzman, Z. Zou and P. D. Sun,
notherapy, Front. Immunol., 2017, 8, 1124. Natural killer cell-produced IFN-γ and TNF-α induce target
7 S. Makino, S. Ikegami, A. Kume, H. Horiuchi, H. Sasaki cell cytolysis through up-regulation of ICAM-1, J. Leukocyte
and N. Orii, Reducing the risk of infection in the elderly by Biol., 2012, 91, 299–309.

This journal is © The Royal Society of Chemistry 2019 Food Funct., 2019, 10, 8129–8136 | 8135
 View Article Online

Paper Food & Function

17 S. Makino, J. Hemmi, H. Kano, M. Kashiwagi, K. Hojo and 22 K. Namba, M. Hatano, T. Yaeshima, M. Takase and
Y. Asami, Anti-fatigue effects of yogurt fermented with K. Suzuki, Effects of Bifidobacterium longum, BB536
Lactobacillus delbureckii subsp. bulgaricus, OLL1073R-1 in administration on influenza infection, influenza vaccine
healthy people suffering from summer heat fatigue: a ran- antibody titer, and cell-mediated immunity in the elderly,
domized, double-blind, placebo-controlled trial, Nutrients, Biosci. Biotechnol. Biochem., 2010, 74, 939–945.
2018, 10, 798. 23 N. Waki, M. Matsumoto, Y. Fukui and H. Suganuma,
18 F. A. Lévi, C. Canon, Y. Touitou, A. Reinberg and G. Mathé, Effects of probiotic Lactobacillus brevis KB290 on inci-
Seasonal modulation of the circadian time structure of cir- dence of influenza infection among schoolchildren: an
Published on 24 October 2019. Downloaded by Vanderbilt University Library on 1/2/2020 9:02:19 PM.

culating T and natural killer lymphocyte subsets from open-label pilot study, Lett. Appl. Microbiol., 2014, 59, 565–
healthy subjects, J. Clin. Invest., 1988, 81, 407– 571.
413. 24 T. Sugimura, H. Takahashi, K. Jounai, K. Ohshio,
19 F. M. Barakat, V. McDonald, J. P. Di Santo and D. S. Korbel, M. Kanayama, K. Tazumi, Y. Tanihata, Y. Miura,
Roles for NK cells and an NK cell-independent source of D. Fujiwara and N. Yamamoto, Effects of oral intake of
intestinal gamma interferon for innate immunity to plasmacytoid dendritic cells-stimulative lactic acid bacterial
Cryptosporidium parvum infection, Infect. Immun., 2009, strain on pathogenesis of influenza-like illness and
77, 5044–5049. immunological response to influenza virus, Br. J. Nutr.,
20 B. D. Forrest, M. W. Pride, A. J. Dunning, M. R. Capeding, 2015, 114, 727–733.
T. Chotpitayasunondh, J. S. Tam, R. Rappaport, 25 H. Zhang, C. Yeh, Z. Jin, L. Ding, B. Y. Liu, L. Zhang and
J. H. Eldridge and W. C. Gruber, Correlation of cellular H. K. Dannelly, Prospective study of probiotic supplemen-
immune responses with protection against culture-con- tation results in immune stimulation and improvement of
firmed influenza virus in young children, Clin. Vaccine upper respiratory infection rate, Synth. Syst. Biotechnol.,
Immunol., 2008, 15, 1042–1053. 2018, 3, 113–120.
21 L. S. Harbige, E. Pinto, J. Allgrove and L. V. Thomas, 26 B. Wang, T. Hylwka, M. Smieja, M. Surrette,
Immune response of healthy adults to the ingested probio- D. M. E. Bowdish and M. Loeb, Probiotics to prevent respir-
tic Lactobacillus casei, Shirota, Scand. J. Immunol., 2016, atory infections in nursing homes: a pilot randomized con-
84, 353–364. trolled trial, J. Am. Geriatr. Soc., 2018, 66, 1346–1352.

8136 | Food Funct., 2019, 10, 8129–8136 This journal is © The Royal Society of Chemistry 2019