Lactobacillus casei Shirota Supplementation Does Not Restore Gut Microbiota Composition and Gut Barrier in Metabolic Syndrome: A Randomized Pilot Study

Metabolic syndrome is associated with disturbances in gut microbiota composition. We aimed to investigate the effect of Lactobacillus casei Shirota (LcS) on gut microbiota composition, gut barrier integrity, intestinal inflammation and serum bile acid profile in metabolic syndrome. In a single-centre, prospective, randomised controlled pilot study, 28 subjects with metabolic syndrome received either LcS for 12 weeks (n = 13) or no LcS (n = 15). Data were compared to healthy controls (n = 16). Gut microbiota composition was characterised from stool using 454 pyrosequencing of 16S rRNA genes. Serum bile acids were quantified by tandem mass spectrometry. Zonulin and calprotectin were measured in serum and stool by ELISA. Bacteroidetes/Firmicutes ratio was significantly higher in healthy controls compared to metabolic syndrome but was not influenced by LcS. LcS supplementation led to enrichment of Parabacteroides. Zonulin and calprotectin were increased in metabolic syndrome stool samples but not influenced by LcS supplementation. Serum bile acids were similar to controls and not influenced by LcS supplementation. Metabolic syndrome is associated with a higher Bacteroidetes/Firmicutes ratio and gut barrier dysfunction but LcS was not able to change this. LcS administration was associated with subtle microbiota changes at genus level. Trial Registration ClinicalTrials.gov NCT01182844


Introduction
Obesity and metabolic disorders (type 2 diabetes and insulin resistance) are tightly linked to inflammation (1) Obesity, a pandemic affecting 30-50% of the adult population, is mediated by a variety of genetic and environmental factors. (2) It is well described that cytokines cause insulin resistance which causes hyperinsulinemia and excessive fat storage in adipose tissue and the liver. (3) However, the triggering factor, linking inflammation to metabolic syndrome has not been fully elucidated yet.

Gut flora interactions
Recently it has been hypothesized that the gut flora is an important factor in this vicious cycle of obesity, metabolic disease and inflammation. Firstly, metabolic activities of the gut microbiota facilitates the extraction of calories from ingested dietary substances and helps to store these calories in host adipose tissue for later use. (4)(5)(6) Second, the gut bacterial flora of obese mice and humans include fewer Bacteroidetes and correspondingly more Firmicutes than that of their lean counterparts, suggesting that differences in caloric extraction of ingested food substances may be due to the composition of the gut microbiota. (7-9) Furthermore, bacterial lipopolysaccharide derived from the intestinal microbiota may trigger inflammation, linking it to high-fat diet-induced metabolic syndrome. High-fat diet induces insulin resistance and oxidative stress in mice and is associated with increased gut permeability. (10) high fat diet induces a low-grade endotoxemia in mice ("metabolic endotoxemia) and infusing endotoxin causes weight gain and insulin resistance. (11) This has also been shown in humans, where patients with fatty liver had a susceptibility to higher gut permeability, possibly causing increased endotoxin levels. (12)

Role of endotoxin
Endotoxin and Lipopolysaccharide-binding protein (LBP) is elevated in obese patients, patients with type 2 diabetes and patients with liver steatosis. (13,14) Endotoxin causes a significant increase in proinflammatory cytokine production in adipocytes via a TLR mediated pathway, contribution to the proinflammatory state in obesity. (13,14) Endotoxin levels correlate with adiponectin (15) and insulin (13,14) suggesting a pathophysiological link between obesity, inflammation and metabolic disease.

Consequences of chronic inflammation in obesity
As described above, endotoxin is related to increased inflammation and oxidative stress, causing insulin resistance. Adipocytes have been shown to play a dynamic role in regulation of inflammation by producing cytokines via a Toll-like receptor (TLR)/Nuclear Factor kappa B (NFkB) mediated pathway. (16) But not only adipocytes are in a proinflammatory state -also circulating mononuclear cells have been described to be activated (17) Clinical evidence suggests immune dysfunction in obesity, since obese patients are more prone to infections after surgery, higher incidence of lower respiratory infection which is also underlined by impairment of cell-mediated immune responses in vivo and in vitro and a reduced intracellular killing by neutrophils. (Reviewed in (18)) A similar situation has been recently described in alcoholic cirrhosis and alcoholic hepatitis, which is also a proinflammatory condition with impaired innate immunity, leading to infection. Endotoxin has been described as a key mediator and inadequate activation of neutrophils leading to high oxidative burst and energy depletion of the cells with consecutive impaired phagocytic capacity has been described. (19)

Effects of modulating gut flora
The most effective therapy of obesity -weight loss -leads to significant improvement of mononuclear cell activation. (20) However, there is no data available on the effect of weight loss on gut flora, gut permeability and endotoxin.
Since weight loss is usually very hard to achieve, other therapeutic strategies have been tested. Since gut flora seems to be crucial in the development of the vicious cycle of obesity, inflammation and metabolic disease, several studies tried to modify the composition of gut microbiota. In mice treatment with antibiotics improved glucose tolerance by altering expression of genes involved in inflammation and metabolism. (10,21) A similar result was found in mice treated with a probiotic that increases the number of Bifidobacterium spp., which leads to improved glucose tolerance, insulin secretion and a decrease in inflammatory tone. (22) Finally treatment of mice with a probiotic (VSL#3) decreased hepatic insulin resistance via a JNK and NFkB pathway, supporting the concept that intestinal bacteria induce endogenous signals that play a pathogenic role in hepatic insulin resistance. (23) Which probiotic?
Among the vast amount of bacteria described to alter gut flora and exert positive effects on the host, we have chosen to study Lactobacillus casei Shirota several reasons: Firstly this commercially available preparation delivers a high bacterial number in a relatively small volume and is available as a palatable milk drink. Furthermore Lactobacillus casei Shirota has been proven to survive the passage through the stomach and is present in the lower intestinal tract (24)(25)(26). It has also been shown that this bacterial strain can increases the amount of Lactobacilli and decreases the number of gram-negative organisms in the bacterial flora (24,27,28). This bacterial strain has been shown to be effective in modulating natural killer cell function (29) and neutrophil function. (30)

Hypothesis
We hypothesize that Lactobacillus casei Shirota is able to decrease metabolic endotoxemia by altering gut flora composition and gut permeability which leads to an improvement in neutrophil function and insulin resistance in obesity

1) To investigate intestinal permeability and its relation to systemic inflammation in
patients with metabolic syndrome.
2) To investigate the effect of Lactobacillus casei Shirota supplementation over 12 weeks on neutrophil function (phagocytosis, oxidative burst and TLR expression) in patients with metabolic syndrome.
3) To investigate the effect of Lactobacillus casei Shirota supplementation over 12 weeks on glucose tolerance, insulin resistance, inflammation, gut flora composition, and endotoxemia in metabolic syndrome

Plan of investigations
Patients 30 Patients with metabolic syndrome will be randomized to either receive food supplementation with a milk drink containing Lactobacillus casei Shirota (3 bottles a day, 65 ml each, containing Lactobacillus casei Shirota at a concentration of 10 8 /ml) for twelve weeks or standard medical therapy.

Exclusion Criteria:
• Drug treatment for diabetes mellitus • Liver cirrhosis (biopsy proven) or elevated transaminases (≥2x ULN) • Inflammatory bowel disease (Crohns disease, ulcerative colitis) • Celiac disease • Alcohol abuse (more than 40g alcohol per day in the history) • clinical evidence of active infection • antibiotic treatment within 7 days prior to enrolment • use of immunomodulating agents within previous month (steroids etc.) • concomitant use of supplements (pre-, pro-, or synbiotics) likely to influence the study • Any severe illness unrelated to metabolic syndrome • malignancy • pregnancy

Randomization
The patients will be randomised using the "Randomizer" (IMI Graz) software.

Study flow and management of patients
Patients will be identified from the outpatient clinic at the Department of Medicine and enrolled into the study on an outpatient basis.
After fulfilling inclusion criteria, patients will be randomized to either receive food supplementation with Lactobacillus casei Shirota (3 bottles of Yakult™ light per day) or standard medical care.
Patients will be advised not to consume any other probiotic supplements during the study period. This will be ensured by handing out a checklist containing all commercially available probiotics in Austria to the patients.
At day 0 and at day 84 (12 weeks) patients will be seen in the outpatients clinic for a detailed examination. On day 28 and 56 the patients will be seen for a routine blood test and a physical examination.
Every two weeks the patients in the treatment group will visit the outpatients clinic to receive the milk drink.
Patients will be managed for their diabetes mellitus type 2 and the metabolic syndrome according to the current national guidelines (www.oedg.org). If patients have elevated liver function tests, a complete hepatological workup will be performed prior to the study to exclude any underlying liver disease other than steatosis.

End of the study
For the individual patient the study will end at day 84 or at the occurrence of a severe adverse event. The study will terminate after the last patient has finished the study. No interim analysis is planned

Endpoints
Neutrophil phagocytosis, neutrophil oxidative burst and TLR expression, glucose tolerance, insulin resistance, plasma and ex vivo stimulated cytokines, endotoxin, gut flora composition, gut permeability After analysis of the first set of data, the remaining material will be used to assess gut permeability modulation and changes in gut microbiota in depth.

Neutrophil phagocytosis
The Phagotest ® (Orpegen Pharma, Heidelberg, Germany) is used to measure phagocytosis by using FITC-labelled opsonized E. coli bacteria as described before. (19) Neutrophil oxidative burst

Glucose Metabolism
After an overnight fast a 3h-oral glucose tolerance test will be performed. Insulin, Cpeptide and glucose will be measured before and 15, 30, 60, 120 and 180 minutes after ingestion of 75g glucose.  OTU's (collector-, rarefaction curves) by using DOTUR. Unbound material is washed from the bead and chemiluminescent substrate was added.

Plasma cytokines and ex vivo cytokine production
Light emission is read with a high-sensitivity photon counter. For cytokine ELISA (IL8, IL10, TNFα and sTNFαR1 and 2) analyte-specific antibodies (capture antibodies) are precoated onto a microplate. A 100-µl serum sample is added and any analyte present is bound by the immobilized antibody. An enzyme-linked analyte-specific detection antibody binds to a second epitope on the analyte, forming the analyte-antibody complex.
Substrate is added and optical density is read on a microplate reader.
Ex vivo stimulation of cytokines will be analysed in a whole blood assay after stimulation with endotoxin derived from E.coli (E.coli 0111:B4 Lot 085K4068, Sigma, Poole, UK) for 4 h at 37 °C.
In addition hsCRP, sICAM-1, sVCAM-1, von Willebrand factor will be determined by ELISA as described above. According to the Hoorn-study publication (35), a summarizing z score for inflammation and endothelial dysfunction will be calculated: (individual value−the mean value for the study population)/standard deviation. The summarising score for inflammation is calculated as followed: (z score for CRP+z score for sICAM-1)/2.

Measurements of choline metabolites
TMAO, trimethylamine, choline, betaine, and their d9-isotopologues are quantified with the use of a stable-isotope-dilution as-say and high-performance liquid chromatography with online electrospray ionization tandem mass spectrometry; d4(1,1,2,2)-choline, d3(methyl)-tri-methylamine-N-oxide, and d3(methyl)-trimethyl-amine are used as internal standards. Levels of TMAO in urine are adjusted for urinary dilution by analysis of the urine creatinine level.

Endotoxin (Limulus amoebocyte lysate assay) and related proteins
Endotoxin Heparinized whole blood is drawn with pyrogen-free needles into pyrogen-free tubes and the serum separated at 4°C and stored at -80°C in pyrogen-free polyethylene cryotubes

Statistical methodology and data analysis
Since this study is a pilot study, no formal sample size calculation can be performed.
All the data will be described as mean and standard errors. All clinical data will be analysed on an intention to treat basis but will also be described on 'as treated' basis.
The primary analysis will be based on a t-test. Analysis of the secondary endpoints will be done as descriptive statistics, by t-test, Mann-Whitney test, Pearson and/or Spearman correlation as appropriate.