Method for Diagnostics, Treatment and Prevention of Parkinson's Disease

Info

Publication number: 20190137493
Type: Application
Filed: Nov 12, 2018
Publication Date: May 9, 2019
Inventors: Filip Scheperjans (Helsinki), Petri Auvinen (Helsinki), Velma Aho (Helsinki), Pedro Pereira (Helsinki), Kaisa Koskinen (Graz), Lars Paulin (Helsinki)
Application Number: 16/186,663

Abstract

The present invention relates to the field of medicine and in particular to Parkinson's disease (PD). Specifically the present invention relates to methods and means for early detection of PD. The invention relates also to methods and means for treatment or prophylaxis of PD. In the method of the invention a probability of a subject developing or having Parkinson's disease (PD) is determined by measuring the relative abundances of one or multiple microbial taxa in a sample from a subject; and the probability of the subject developing or having PD is determined based on the measured abundances. The present invention provides a novel approach for the diagnostics of PD.

Description

Description

RELATED APPLICATION DATA

This application is a divisional application which claims priority to U.S. patent application Ser. No. 15/314,240, filed on Nov. 28, 2016, which is a National Stage Application under 35 U.S.C. 371 of co-pending PCT application PCT/FI2015/050374 designating the United States and filed May 28, 2015; which claims the benefit of FI application 20145492 and filed May 28, 2014 each of which are hereby incorporated by reference in their entireties.

SEQUENCE LISTING

The instant application contains a Sequence Listing which has been submitted electronically in ASCII format and is hereby incorporated by reference in its entirety. Said ASCII copy, created on Jan. 14, 2019, is named 001942_00086_ST25.txt and is 1 kilobyte in size.

FIELD OF THE INVENTION

The present invention relates to the field of medicine and in particular to Parkinson's disease (PD). Specifically the present invention relates to methods and means for early detection of PD. The invention relates also to methods and means for treatment or prophylaxis of PD.

BACKGROUND OF THE INVENTION

Parkinson's disease is the most frequent movement disorder. The prevalence in people above 65 years is approximately 1% and it has a great negative effect on quality of life. The cause of Parkinson's disease (PD) is unknown and there are no disease modifying treatments available. In Parkinson's disease (PD), the cardinal motor symptoms are mainly related to the loss of dopaminergic neurons in the substantia nigra. However, neuropathologic changes are much more widespread involving the autonomic nervous system, olfactory structures, lower brainstem and cerebral cortex. Extranigral pathology is related to a broad spectrum of non-motor symptoms (NMS) that have been increasingly recognized as an important feature of PD. Gastrointestinal dysfunction, in particular constipation, affects up to 80% of PD-patients and may precede the onset of motor symptoms by years (Savica et al. 2009). Idiopathic constipation is one of the strongest risk-factors for PD (Noyce et al. 2012).

It is not known what factors initiate the pathophysiological cascade leading to neurodegeneration in PD, but an environmental factor likely plays a key role in PD pathogenesis probably against a background of genetic vulnerability (Kieburtz et al. 2013). The early involvement of the gastrointestinal tract in PD lends support to the hypothesis that this environmental factor exerts its influences primarily via the gut. However, recent studies have revealed that, changes of the complex equilibrium of the entire microbiome may be related to human disease.

Intestinal microbiota have gained a lot of attention in research in recent years and dysequilibrium of the gut microbiome has been associated with several diseases, including autism, bowel disease and cancer, rheumatoid arthritis, diabetes, and obesity. The gut microbiome in PD has not been previously investigated.

Currently there is no method available for an early diagnostics of PD. Biomarkers for PD, especially for the premotor phase, are urgently needed since future disease modifying therapies should be initiated as early as possible in the disease process to maximize their effect. Moreover, there is no disease modifying treatment available for PD.

BRIEF DESCRIPTION OF THE INVENTION

An object of the present invention is thus to provide methods and means for early detection of PD. A further object is to provide methods and means for treatment or prophylaxis of PD.

The objects of the application are achieved by a method for measuring the probability of a subject developing or having Parkinson's disease (PD), the method comprising

a. obtaining a sample from a subject;

b. determining relative abundances of at least Prevotellaceae taxa in the sample; and

c. determining probability of the subject developing or having PD based on the abundances measured in b,

wherein a high relative abundance of said taxa indicates a low probability of the subject developing or having PD.

Further the present invention provides a method for determining a motor subtype of a PD patient, wherein method comprises:

obtaining a sample from a subject;

determining the abundance of one or multiple of the following taxa: Enterobacteriaceae, Clostridium XVIII, Anaerofilum, Papillibacter, Succiniclasticum, Klebsiella, Escherichia-Shigella and Paludibacter in said sample; and determining by statistical methods the motor subtype based on the measured abundances, wherein low relative abundance of the said taxon or taxa indicates a tremor dominant subtype and high relative abundance indicates a non-tremor subtype.

Additionally, the present invention provides a kit for detection and risk assessment of PD.

Further, the present invention provides a method for treatment of patients with PD preferably before the appearance of motor symptoms of PD, or for prophylaxis of PD in subjects with conditions or symptoms indicating an increased risk for PD, the method comprising administering an effective amount of a composition that increases the relative abundance of Prevotellaceae in the intestines to a subject in need of such treatment.

Additionally, the present invention provides a composition increasing abundance of Prevotellaceae in the intestine for use in treatment or prevention of PD.

The preferred embodiments of the invention are disclosed in the dependent claims.

The inventors of the present application surprisingly noticed an association of microbiota, especially gut microbiota and neurodegenerative disease. The inventors found that PD patients did not only have an altered microbiome when compared to matched control subjects, but microbiota were associated also with the motor phenotype of PD. These findings were independent of controlled confounders such as degree of constipation, disease duration, medication, comorbidities, gender, or age.

The present invention provides a novel approach for the diagnostics of PD. The method is rapid, non-invasive and easy to use. The present invention provides a considerable advantage of enabling the individuals having a risk of PD being diagnosed at an early stage preferably before motor symptoms appear e.g. in patients with severe constipation of unknown cause or irritable bowel syndrome. Once diagnosed at an early stage as belonging to the risk group, the onset of PD in the individual can be prevented by modifying the community structure of the gut microbiota. The present invention specifically provides novel means for preventing PD in a subject and also means for slowing the disease process or even stopping it. The combination of the method and the composition of the invention enable development of personalized treatment and possibly personalized dietary guidance. A further advantage of the invention is that the present method to measure the risk of a person to develop PD provides also a new tool for investigation of patients with irritable bowel syndrome (IBS) that currently is a diagnose of exclusion (other possible sources of symptoms are excluded). A considerable fraction of these patients do have premotor PD and the present invention the first test that can be used to identify PD at this stage. IBS symptoms are frequently severe and impairing quality of life (pain, constipation, diarrhea) and treatment is purely symptomatic. Diagnosing PD in these patients using the methods of the present invention opens a way for disease modifying treatment. The same applies to patients with idiopathic constipation.

Still a further advantage of the invention is that selecting donors for fecal transplantation to PD subjects based on high Prevotellaceae abundance ensures effective treatment and is a clear improvement to a transplantation of feces with unknown microbiome structure.

The present invention may improve gut function in PD patients leading to relief of gastrointestinal symptoms including constipation and IBS, improved nutritional status. It may increase levels of thiamine, folate, short chain fatty acids, hydrogen sulfide, and gut hormones like ghrelin in the body of PD patients leading to alleviation of symptoms and slowing of disease progression. Prevotella enterotype is associated with high production of these molecules and those have been shown to be decreased in PD and effective in PD treatment.

The present invention may provide a tool for differentiating patients with PD from patients with another disease mimicking PD (such as Progressive supranuclear palsy, Multiple system atrophy, Corticobasal degeneration, vascular parkinsonism, Alzheimer's disease, Essential tremor, Normal pressure hydrocephalus)

The method of the invention improves barrier function of the gut mucosa reducing exposure to bacterial endotoxin, inflammation and oxidative stress in the patient organism leading to better health.

Since microbiota composition is associated with certain medications e.g. COMT-inhibitors, analysis of microbiota in PD patients can be useful to select the optimal medications for a patient, minimizing adverse side effects.

BRIEF DESCRIPTION OF THE DRAWINGS

In the following the invention will be described in greater detail by means of preferred embodiments with reference to the attached drawings, in which

FIG. 1 shows relative abundances of the 10 most abundant bacterial families in the subsampled dataset;

FIG. 2 shows distributions of Prevotellaceae abundance in both study groups

Box plots show the distributions of Prevotellaceae abundance in both study groups. Black horizontal lines indicate the median values and the boxes around them delineate the IQR. Whiskers extend to the highest value within 1.5 IQR of the upper quartile. Circles represent outliers beyond the whisker limit and asterisks represent extreme outliers beyond 3 IQR of the upper quartile. Median [IQR]: Parkinson 0.16% [0.00%-1.66%]; Control: 0.77% [0.00%-18.18%];

FIG. 3 shows ROC curves of discriminators of study group membership

The ROC curve analysis of Prevotellaceae abundance was somewhat hampered by the considerable number of samples in both groups being devoid of this family (AUC=0.591 [95% CI 0.496-0.685], P=0.060). Since samples devoid of Prevotellaceae were equally frequent in PD and control groups, we performed a second ROC curve analysis only including subjects positive for Prevotellaceae (n=103; AUC=0.664 [95% CI 0.556-0.771]; P=0.004; not shown). Logistic regression classifier including abundances of Prevotellaceae, Lactobacillaceae, Bradyrhizobiaceae and Clostridiales Incertae Sedis IV: AUC=0.722 [95% CI 0.641-0.804], P<0.001. Wexner score: 0.747 [95% CI 0.666-0.828], P<0.001. Logistic regression classifier as above, but including also Wexner constipation score: AUC=0.832 [95% CI 0.766-0.897], P<0.001;

FIG. 4 shows distribution of Enterobacteriaceae in phenotypic subgroups of PD

Box plots showing the distributions of Enterobacteriaceae abundances in tremor dominant (TD) and postural instability/gait difficulty (PIGD) phenotypes. Black horizontal lines indicate the median values and the boxes around them delineate the IQR. Whiskers extend to the highest value within 1.5 IQR of the upper quartile. Asterisks represent extreme outliers beyond 3 IQR of the upper quartile. Median [IQR]: TD (0.04 [0.00-0.27]; n=23); PIGD (0.46 [0.07-1.84]; n=40);

FIG. 5 shows microbial taxa that can be used as biomarkers for the diagnosis of PD. We chose 10 PD fecal samples with lowest abundances of Prevotellaceae and highest abundances of Ruminococcaceae. The other 10 were control samples with high abundances of Prevotellaceae. For identification of taxa, DNA from fecal samples was shotgun sequenced using Illumina MiSeq. Sequences equal or above 75 nucleotides were taxonomically identified using MGRast. Biomarkers for PD were selected using the LEfSe method (Segata et al, 2011). The plots show the biomarkers found by LEfSe ranking them accordingly to their effect size and associating them with the class with the highest median. Green columns indicate taxa more abundant in PD patients and red columns indicate taxa more abundant in control subjects.

FIGS. 6A-6F show plots and data generated using the partitioning around the medoid (PAM) and Dirichlet multinomial mixture (DMM) methods to classify samples into groups based on similarity of their microbiome community structure on bacterial family level (fecal samples of 72 PD patients and 72 control subjects with microbiome community structure determined as described previously (Arumugam et al. 2011). FIG. 6A: Results of principal component analysis demonstrating that samples are classified into 3 groups (P, R, B). FIGS. 6B and 6D: Only 5 PD subjects, but 19 control subjects were assigned to group P. Chi Square test confirming significant association between study group and classification group. FIG. 6C Non-metric Multidimensional Scaling arranging individual subject samples according to similarity of microbiome community structure. The closer two samples, the more similar the microbiome. Dots=control subjects, Triangles=PD patients. The plot also shows the dominating families. FIGS. 6E and 6F Relative abundance of the five most abundant families in the three groups based on the DMM (E) and PAM (F) methods. Both methods show that the P group shows particularly high abundance of Prevotellaceae, whereas the R group is dominated by Ruminococcaceae and the B group mainly by Bacteroidaceae.

FIGS. 7A-7F show plots and data generated using the same techniques and samples as for FIGS. 6A-6F, but for the bacterial genus level. FIG. 7A: Results of principal component analysis demonstrating that samples are classified into 3 groups. FIGS. 7B and 7D: Only 1 PD subject, but 16 control subjects were assigned to group P (=group 1 from DMM) whereas 45 PD subjects, but only 28 subjects were assigned to group 0 (=group 3 from DMM). Chi Square test confirming significant association between study group and classification group. FIG. 7C Non-metric Multidimensional Scaling arranging individual subject samples according to similarity of microbiome community structure. The closer two samples, the more similar the microbiome. Dots=control subjects Triangles=PD patients. FIGS. 7E and 7F Relative abundance of the ten most abundant genus in the three groups based on the DMM (E) and PAM (F) methods. The P group (=group 2 in E and group 1 in F) shows particularly high abundance of Prevotella, whereas the B group (=group 1 in E and group 2 in F) is dominated by Bacteroides and Lachnospiraceae incertae sedis. Group 0 (=group 3) shows relatively high abundance of Oscillibacter and Saccharofermentans.

FIG. 8 shows microbial taxa that can be used as biomarkers for the diagnosis of PD based on analysis of 72 PD and 72 control fecal samples. Biomarkers for PD were selected using the LEfSe method. The plots show the biomarkers found by LEfSe ranking them accordingly to their effect size and associating them with the class with the highest median. Light grey columns indicate taxa more abundant in PD patients and dark grey columns indicate taxa more abundant in control subjects.

DETAILED DESCRIPTION OF THE INVENTION

The present invention relates to a method, wherein the probability of a subject developing or having Parkinson's disease (PD) is determined by a method wherein a sample is obtained from a subject; relative abundances of one or more microbial taxa in the sample are measured; and probability of the subject developing or having PD is determined based on the measured relative abundances of one or multiple microbial taxa in the sample.

The present invention concerns generally Parkinsons's disease and also Parkinsons's disease where the definition of PD includes a prodromal and a premotor period of the disorder in which the subject has no motor symptoms. A prodromal period means a period of early symptoms that might indicate the start of a disease before specific symptoms occur. In a pre-motor period typical motor symptoms of PD have not yet developed although neurons in the autonomic nervous system or brain have started to degenerate. Generally, PD is suspected based on the subject having one or more of the following conditions that are associated with PD: family history of PD, family history of tremor, constipation, irritable bowel syndrome, hyposmia, REM-sleep behavior disorder, mood disorder, hyperechogenicity of the substantia nigra on transcranial ultrasound examination.

In the present invention relative abundances of one or more microbial taxa in a sample are measured. A sample is a microbial sample taken from an individual. Preferably it is a gut microbiota sample obtained from lower gastrointestinal tract, more preferably it is e.g. feces sample. The sample may also be an oral or nasal mucosal swab sample.

Gut flora or, more appropriately, gut microbiota, consists of a complex community of microorganism species that live in the digestive tracts of human and animals and is the largest reservoir of microorganisms mutual to humans. In this context gut is synonymous with intestinal, and flora with microbiota and microflora. The gut microbiome refers to the genomes of the gut microbiota, but is also used synonymously with microbiota and microflora.

In the present invention the probability of the subject developing or having PD is determined based on the measured relative abundances of one or multiple microbial taxa in the sample. The high relative abundance is defined as being higher as a reference value, which is a predefined threshold value, and vice versa. Relative abundance means the abundance of a microbial taxon relative to the sum of the abundances of all taxa detected in a sample.

A reference sample is a sample from an individual not having PD and having a low risk of developing PD. Indicators of low risk are e.g. no strong family history of PD, no strong family history of tremor, no constipation, no irritable bowel syndrome, no hyposmia, no REM-sleep behaviour disorder, no mood disorder. In the method of the invention taxa abundances in reference samples including samples from PD patients can be used to generate a logistic regression model, wherein the probability of developing or having PD is modelled as a function of taxa abundances. The value of a logistic regression classifier is calculated based on said logistic regression model and the values of the subject. The value of said logistic regression classifier is then used to determine the probability of the subject developing or having PD.

The relative abundance of the microbial taxa can be measured using techniques based on DNA sequencing, quantitative PCR, DNA microarray, by using test beds utilizing arrays or other suitable platforms including microfluidistic solutions, techniques based on droplet PCR, or any other suitable method wherein the abundance of taxa is expressed as read count, percentage, cell count, or value expressing intensity of any other suitable signal.

A microbial taxon as used herein refers to a taxonomic unit, whether named or not: i.e. a population, or group of populations of organisms which are usually inferred to be phylogenetically related and which have characters in common which differentiate the unit (e.g. a genus, a family) from other such units. A taxon encompasses all included taxa of lower rank and individual organisms. The term as used herein furthermore includes termed species-level phylotypes or operational taxonomic units that are identified only by their complete 16S rRNA sequence and usually defined as sharing 97% or less sequence identity with other entries in the ribosomal databases. In the present invention the measurement of relative abundances of microbial taxa may include all taxa that can be identified in a sample.

In the method of the invention the microbial taxa is preferably Prevotellaceae, and more preferably Prevotella. According to one embodiment of the invention the taxon is one or more of the following: Prevotella amnii, Prevotella bergensis, Prevotella bivia, Prevotella bryantii, Prevotella buccae, Prevotella buccalis, Prevotella copri, Prevotella disiens, Prevotella marshii, Prevotella melaninogenica, Prevotella oralis, Prevotella oris, Prevotella ruminicola, Prevotella salivae, Prevotella sp oral taxon 299, Prevotella sp oral taxon 317, Prevotella sp oral taxon 472, Prevotella tannerae, Prevotella timonensis and Prevotella veroralis.

The basic idea behind the present invention is the finding of the reduced abundance of Prevotellaceae in PD patients. According to the present study, a person with a relative abundance of Prevotellaceae of more than 6.5% is very unlikely to have PD (86.1% sensitivity). In other words, high relative abundance of Prevotellaceae indicates a low probability of the subject developing or having PD.

Prevotella is a genus of Gram-negative bacteria. It is a commensal microbe in the human large intestine and has the ability to degrade a broad spectrum of plant polysaccharides. It not only plays a key role in digesting carbohydrate-rich food and degrading of mucin glycoproteins in the mucosal layer of the gut, but may also interact with the immune system. Prevotella is the main contributor of one of the recently suggested gut microbiome enterotypes (Arumugam et al. 2011) in addition to Bacteroides, and Ruminococcus. The composition of the gut microbiome of an individual is established in early childhood and the enterotype of an individual is thereafter very stable. Therefore the present invention measuring Prevotella, Bacteroides, and Ruminococcaceae abundances as determinants of the enterotype can also be used to determine the risk of a subject for future development of PD.

In addition to Prevotellaceae, also the abundances of Lactobacillaceae, Verrucomicrobiaceae, Bradyrhizobiaceae, and Clostridiales Incertae Sedis IV were independently associated with PD in the studies of the present invention. Thus, in one aspect of the invention one or more of the following taxa may be included in the measurement: Lactobacillaceae, Verrucomicrobiaceae, Bradyrhizobiaceae, Clostridiales Incertae Sedis IV and Ruminococcaceae.

In one embodiment of the invention low relative abundance of Prevotellaceae and high relative abundances of Lactobacillaceae, Verrucomicrobiaceae, Bradyrhizobiaceae, Clostridiales Incertae Sedis IV and/or Ruminococcaceae indicates a high probability of the subject developing or having PD.

According to another embodiment one or more of the following taxa are included in the determination in addition to Prevotella: Sutterella, Saccharofermentans, Mahella, Lactobacillus, Phaeovibrio, Agromonas, and/or Anaerotruncus. In addition to low relative abundance of Prevotella low relative abundance of the taxon Sutterella and high relative abundance of the taxa Saccharofermentans, Mahella, Lactobacillus, Phaeovibrio, Agromonas, and/or Anaerotruncus indicates a high probability of the subject developing or having PD.

According to another embodiment one or more of the following taxa are included in the determination in addition to Prevotellaceae and Prevotella: Bacteroidia, Clostridiaceae, Clostridium sensu stricto, Faecalibacterium, Mogibacterium, Oscillibacter, Prevotella, Prevotellaceae, Roseburia, Acetivibrio, Actinobacteria, Allobaculum, Anaerotruncus, Blautia, Butyricicoccus, Clostridium IV, Collinsella, Coriobacteriaceae, Coriobacteriales, Eggerthella, Firmicutes, Parabacteroides, Porphyromonadaceae, Ruminococcaceae, Sporobacterium. In addition to low relative abundance of Prevotella and/or Prevotellaceae low relative abundance of the taxa Bacteroidia, Clostridiaceae, Clostridium sensu stricto, Faecalibacterium, Mogibacterium, Oscillibacter, and/or Roseburia and high relative abundance of the taxa Acetivibrio, Actinobacteria, Allobaculum, Anaerotruncus, Blautia, Butyricicoccus, Clostridium IV, Collinsella, Coriobacteriaceae, Coriobacteriales, Eggerthella, Firmicutes, Parabacteroides, Porphyromonadaceae, Ruminococcaceae, and/or Sporobacterium indicates a high probability of the subject developing or having PD.

According to one embodiment of the invention in addition to the low relative abundance of Prevotella and the taxa from one or more of the further bacterial species, which may be selected from the group of Bacteroides barnesiae, Bacteroides coprocola, Bacteroides coprophilus, Clostridium argentinense, Dialister invisus, Dialister micraerophilus, Fibrobacter succinogenes, Haemophilus influenzae, Haemophilus parainfluenzae, Leadbetterella byssophila, Shigelladysenteriae, Shigella and Sporanaerobacteracetigenes, and the high relative abundance of one or more of the taxa, which may be selected from the following group: Achromobacter denitrificans, Alistipes putredinis, Alistipes shahii, Alistipes sp HGB5, Alkaliphilus metalliredigens, Bacteroides dorei, Bacteroides eggerthii, Bacteroides fragilis, Bacteroides intestinalis, Bacteroides, Bacteroides sp 1 1 14, Bacteroides sp 2 1 16, Bacteroides sp 2 1 7, Bacteroides sp 2 2 4, Bacteroides sp 3 1 33FAA, Bacteroides sp 3 2 5, Bacteroides sp 4 1 36, Bacteroides sp 9 1 42FAA, Bacteroides sp D20, Bacteroides uniformis, Bifidobacterium longum, Bifidobacterium scardovii, Bilophila wadsworthia, Caldicellulosiruptor saccharolyticus, Clostridium cellulolyticum, Clostridium hathewayi, Clostridium methylpentosum, Clostridium symbiosum, Coprococcus catus, Desulfovibrio, Desulfovibrio vulgaris, Enterococcus faecalis, Enterococcus faecium, Erysipelotrichaceae bacterium 3 1 53, Erysipelotrichaceae bacterium 5 2 54FAA, Ethanoligenens harbinense, Eubacterium cylindroides, Eubacterium siraeum, Eubacterium ventriosum, Holdemania filiformis, Lachnospiraceae bacterium 5 1 63FAA, Lachnospiraceae bacterium 8 1 57FAA, Paenibacillus sp TS12, Paracoccus aminophilus, Ruminococcaceae bacterium D16, Thermoanaerobacter sp X514 indicates a high probability of the subject developing or having PD.

The method of the present invention may include a quantitative measurement of constipation, wherein values indicating constipation indicate a high probability of the subject developing or having PD. The constipation may be measured e.g. by using a questionnaire or an objective assessment of colonic transit time.

In the present invention clinical measures of patient symptoms that may include, but are not restricted to quantitative measurements of constipation, may be included in the logistic regression model. Also information on the presence of PD risk alleles in the genome of the subject may be included in the model.

Specifically, in the method of the present invention taxa abundances from reference samples including samples from PD patients and healthy controls and abundances from the sample of the subject are subjected to a statistical classification method such as, but not restricted to “partitioning around the medoid” and “Dirichlet multinomial mixture”, that assigns samples into groups based on their taxonomic composition (Arumugam et al., 2011). If the method from a statistical classification assigns the subject's sample to a group containing few PD samples, the probability of the subject developing or having PD is determined to be low and vice versa. Also if the method from a statistical classification assigns the subject's sample to a group where the samples have higher abundances of one or more of the taxa Prevotellaceae or Prevotella as compared to the other groups the probability of the subject developing or having PD is determined to be low.

PD is a clinically heterogenic disorder and it has been suggested that different pathophysiologic mechanisms underlie the observed differences in expression of tremor and non-tremor symptoms between patients. In comparison to tremor dominant (TD) patients, patients with a non-TD phenotype progress faster, have a worse prognosis, and show more severe alpha-synuclein pathology in the colonic enteric nervous system.

The present invention relates also to a method for determining a clinical subtype of a PD patient, especially a motor subtype The method comprises obtaining a sample from a subject; determining the abundance of one or multiple of the following taxa: Enterobacteriaceae, Clostridium XVIII, Anaerofilum, Papillibacter, Succiniclasticum, Klebsiella, Escherichia/Shigella and Paludibacter in said sample; and by using statistical methods determining a motor subtype based on the measured abundances, wherein low relative abundance of the said taxon or taxa indicates a tremor dominant subtype and high relative abundance indicates a non-tremor subtype. The determination of clinical subtypes can be a useful tool to improve research methods and thus find a cure or better treatments, or help to counsel patients or direct existing therapies.

In one aspect the present invention relates to a kit for use in a method of the invention i.e for detection and risk assessment of PD, the comprising means for determining alterations in the microbiota, especially gut microbiota. The kit comprises reagents and means for determining the abundance of one or more of the taxa mentioned in claims of the present invention in a sample.

The present invention relates further to a composition comprising a gut microbiome altering agent that, when administered to an individual, increases the abundance of Prevotellaceae or supports growth of these taxa in the intestine. The composition may comprise live or killed microbes and optionally a pharmaceutically acceptable carrier. Preferably the microbes are from the taxon Prevotellaceae. A composition may also be such that when administered to an individual, it decreases the abundance of one or more of the taxa Lactobacillaceae, Verrucomicrobiaceae, Bradyrhizobiaceae, Clostridiales Incertae Sedis IV, Ruminococcaceae, Lachnospiraceae, Enterobacteriaceae, Bacteroidaceae, Alistipes, Coprococcus, Anaerotruncus, Porphyromonadaceae, Parabacteroides, Butyricicoccus, Collinsella, Eggerthella, Coriobacteriaceae, Sporobacterium, Actinobacteria, Blautia, Enterococcus, Acetivibrio, Allobaculum Agromonas, Firmicutes, Mahella, Phaeovibrio, Porphyromonadaceae, Oscillibacter, and/or Saccharofermentans in the intestine.

The composition of the invention may be in the form of a food composition, pharmaceutical composition, nutraceutical, supplement or an anaerobical microbiota culture. It may include probiotics and other micro-organisms, prebiotics, antibiotics, growth factors, bacteriophages etc. The composition may be administered mixed in food or drink, for example, or separately in the form of a tablets, capsules, microcapsules, powders, solutions, pastes, etc. Food composition may be any kind of food (functional, conventional and novel), food supplement, formula for nutritional purposes, or nutraceutical and it may contain any suitable additives and excipients. The effect of the bacterial supplementation may be enhanced by adding prebiotics such as fibre or oligosaccharides to the composition. The composition may also be a prebiotic and optionally contain a pharmaceutically acceptable carrier. The composition may contain live microbes of the taxon Prevotellaceae and a prebioticial agent or compound supporting growth of these taxa and optionally a pharmaceutically acceptable carrier. Generally composition of gut microbiome can be changed by following a long-term diet.

The present invention relates further to a method for treatment or prevention of PD in an individual by administering said individual a composition comprising a gut microbiome altering agent that, when administered to an individual, increases the abundance of Prevotellaceae in the intestine. The composition to be administered may be the composition of the present invention as disclosed above. The composition may also be a stool transplant. The administration of a gut microbiome altering agent can be done e.g via a bacterial composition given orally, via a nasogastric tube, as a suppository, or by fecal microbiota transplantation. Fecal microbiota transplantation (FMT) or a stool transplant is the process of transplantation of fecal bacteria from a healthy individual into a recipient. It involves restoration of the colonic flora by introducing healthy bacterial flora through infusion of stool, e.g. by enema.

When (FMT) or a stool transplant is used the feces donor is preferably previously selected according to the present invention based on a high abundance of Prevotellaceae and/or low abundance of one or more of the taxa Enterobacteriaceae, Lactobacillaceae, Ruminococcaceae, Bacteroidaceae, Lachnospiraceae, Verrucomicrobiaceae, Bradyrhizobiaceae, Clostridiales Incertae Sedis IV, Alistipes, Coprococcus, Anaerotruncus, Porphyromonadaceae, Parabacteroides, Butyricicoccus, Collinsella, Eggerthella, Coriobacteriaceae, Sporobacterium, Actinobacteria, Blautia, Enterococcus, Acetivibrio, Allobaculum Agromonas, Firmicutes, Mahella, Phaeovibrio, Porphyromonadaceae, Oscillibacter, and/or Saccharofermentans in feces. It is also preferred that the donor feces is previously selected based on method of the present invention showing that the donor does not have PD and has a low risk of developing PD.

Preferably the treatment or prevention of PD in an individual is initiated before the diagnosis of PD based on results of the microbial analysis of the present invention. The goal of the treatment is to slow the disease process or to even stop it, or even prevent PD. The term “treatment” may refer to both therapeutic treatment and prophylactic treatment or preventative measures, wherein the goal of the treatment is to slow the disease process or even stop it, or prevent PD.

It will be obvious to a person skilled in the art that, as the technology advances, the inventive concept can be implemented in various ways. The invention and its embodiments are not limited to the examples described below, but may vary within the scope of the claims.

EXAMPLES

The study was approved by the ethics committee of the Hospital District of Helsinki and Uusimaa and all participants gave informed consent. The study was registered at clinicaltrials.gov (NCT01536769).

Study Subjects

This case-control study compared patients with a diagnosis of PD according to the Queen Square Brain Bank criteria with gender- and age-matched (±five years) control subjects without any signs of parkinsonism or potential premotor symptoms. The chance of recruiting patients with monogenic parkinsonism was minimized by restricting the age of motor-symptom onset to above 50 years and by excluding subjects with more than one relative or one first-degree relative with PD. From the control group subjects with symptoms associated with premotor PD such as hyposmia, REM sleep behaviour disorder (RBD), and accumulation of other NMS were excluded. Further exclusion criteria covered a broad range of conditions and medications that could independently affect the fecal microbiome.

Subjects were recruited from in- and outpatient departments of the participating hospitals as well as via referrals from cooperating neurologists. We also accepted referrals via participants and invited subjects via announcements in patient journals and meetings. Between November 2011 and December 2012, 277 subjects were screened for the study. Of these, 152 (76 patients, 76 controls) were included. All patients were under regular follow-up by a neurologist who was confident in the diagnosis. Patient records were reviewed if available. Due to sample processing 72 patients and 72 controls were included in the final analysis.

Clinical Data

Parkinsonian symptoms were measured using the Unified Parkinson's Disease Rating Scale (UPDRS) and the modified Hoehn & Yahr scale (H&Y). The PD patients were classified into postural instability and gait difficulty (PIGD), tremor dominant (TD), and mixed phenotypes (MX) as described by Jankovic et al 1990. In addition, an akinetic-rigid symptom score was calculated as described by Poletti et al. 2011, but items 29 (gait) and 30 (postural stability) were excluded since those are included in the PIGD score. Overall NMS severity was assessed using the Non-Motor Symptoms Scale (NMSS). The degree of constipation was quantified in more detail using the Wexner constipation score. Diagnosis of active irritable bowel syndrome (IBS) was an exclusion criterion in this study since it is associated with alterations of gut microbiota.

Analysis of Fecal Microbiota

Sample Collection

The subjects collected the fecal samples at home into collection tubes pre-filled with Stool DNA Stabilizer (PSP® Spin Stool DNA Plus Kit, STRATEC Molecular). Within three days, the tubes were transferred to a storing temperature of −80° C. until further processing.

DNA Extraction and PCR

Total DNA was extracted using the PSP® Spin Stool DNA Plus Kit (STRATEC Molecular). PCR amplification was carried out in an ARKTIK Thermal Cycler (Finnzymes Diagnostics/Thermo Scientific). In the first stage, the V1-V3 regions of the bacterial 16S rRNA gene were amplified in three replicate reactions with universal bacterial primers pA (AGAGTTTGATCMTGGCTCAG)(SEQ ID NO:1) and pD′ (GTATTACCGCGGCTGCTG)(SEQ ID NO:2) with 18-mer overhangs added to the 5′ ends of the primers (Edwards et al., 1989; Lane et al. 1991). The replicate PCR products were pooled and purified with Agencourt® AMPure® XP magnetic beads (Agencourt Bioscience) and subjected to a second PCR round with barcoded forward primers and a reverse primer, both of which attached to the respective 18-mer overhang sequences from the primers of the first PCR amplification. Phusion polymerase (Thermo Fisher Scientific/Finnzymes) with HF buffer and 2.5% DMSO were used. Cycling conditions for both PCR reactions consisted of an initial denaturation at 98° C. for 30 s, followed by 15 cycles at 98° C. for 10 s, 65° C. for 30 s, and 72° C. for 10 s, and then a final extension for 5 min. Between 3.6 and 60 ng of template DNA were used in the initial reaction. DNA concentration and quality were measured with Qubit (Invitrogen) and Bioanalyzer 2100 (Agilent).

Sequencing and Sequence Quality Control

PCR products were sequenced using 454-GS FLX Titanium chemistry, with an average read length of ˜400 bp (Roche Diagnostics). The dataset consisted of 2549217 raw reads (mean read count 17224 per subject). Basic sequence quality control and taxonomical assignment were performed with mothur, following the standard operating procedure for 454 sequenced 16S data (Schloss et al., 2011). The phylotype based approach was selected for further analysis. The final dataset included 1131504 reads, with a mean read count of 7645 per subject, and a mean read length of 218 bases. The sequences have been deposited in the Sequence Read Archive at the European Bioinformatics Institute (accession no. PRJEB4927).

Statistical Analysis

From each subject, random subsamples of 4500 sequences were used for statistical analysis of family level data. Alpha diversity indices were calculated and beta diversity dendrograms describing the between subject differences were generated and compared with mothur. Differences in bacterial communities between patients and controls and between PIGD and TD patients (a priori defined subgroup analysis) were analyzed using Metastats (White et al., 2009). The following analyses were performed using IBM® SPSS® Statistics Version 19.0.0.1 (IBM Corp.): Group differences of clinical parameters were analyzed using T-tests for normally distributed variables, otherwise non-parametric tests were used. Differences regarding categorical variables were tested using Fisher's exact test. Associations of microbiome features with clinical parameters were explored using generalized linear models (GLM) employing linear, negative binomial, or logistic distributions depending on the distribution of the target variable.

In the Metastats results, differences with Q-values below 0.05 were considered significant and are the basis of the main conclusions of this paper. We also performed exploratory analyses on bacterial families with higher Q-values if P-values were below 0.05. In the remaining analyses (SPSS), P-values below 0.05 (two-sided if not indicated otherwise) were considered significant.

Results

Demographics and Clinical Data

The patient and control groups were similar with respect to most of the studied variables (Tables 1).

TABLE 1 Selected demographic and clinical parameters of the cohort including all those parameters that showed significantly different distributions between groups. Patients Controls P-Value Demographics n 72 72 female subjects 48.6% 50.0% 1.000 Age (years, mean ± SD) 65.3 ± 5.5 64.5 ± 6.9 0.448 Medical History Atrial Fibrillation 4.2% 18.1% 0.015 TIA or Ischaemic Stroke 7.0% 37.5% <0.001 Non-motor Symptoms Overall NMS severity 40 [25.25-55.00] 8 [4.00-11.75] <0.001 (NMSS Constipation 5 [3-9] 2 [1-4] <0.001 (Wexner score) Medication Levodopa 54.2% 0% <0.001 COMT Inhibitor 15.3% 0% 0.001 Dopamine Agonist 77.8% 0% <0.001 MAO Inhibitor 70.8% 0% <0.001 Anticholinergic 8.3% 0% 0.028 Warfarin 1.4% 15.3% 0.004 Statin 20.8% 54.2% <0.001

A dopamine transporter SPECT study was documented for 26 (36.1%) patients (TD: 56.5%, MX: 22.2%, PIGD: 27.5%). For 14 (19.4%) patients it was not known whether a SPECT study had been performed at time of diagnosis. The median of the time from motor symptom onset to study visit was 5 [3-8] years (TD: 5 [3-6], MX: 6 [2-12.5], PIGD: 6 [3-10]). The median H&Y stage was 2.5 [2.0-2.5] (Table 2).

TABLE 2 Distribution of PD patients into the modified Hoehn & Yahr (H&Y) stages H&Y Stage 1 1.5 2 2.5 3 n (%) 4 (5.6%) 2 (2.8%) 23 (31.9%) 27 (37.5%) 16 (22.2%)

The mean (±SD) scores for total UPDRS and UPDRS-III were 45.5±13.6 and 31.6±9.0 points, respectively. 97.3% of PD patients were using antiparkinsonian medication (Table 1). Two patients (2.8%) were treated by deep brain stimulation (DBS). 55.6% of our patients had a PIGD phenotype, 31.9% were TD, and 12.5% were classified as MX. PD patients had significantly more severe NMS overall (median NMSS score 40 [25.25-55.00] vs. 8 [4.00-11.75] points; P<0.001) and constipation in particular than control subjects (median Wexner constipation score 5 [3-9] vs. 2 [1-4] points; P<0.001).

Microbiome

The full dataset included bacteria from 360 genera, 125 families, 60 orders, 29 classes and 18 phyla. The majority of the reads (94%) represented the phyla Firmicutes and Bacteroidetes, which are typically the dominant phyla in the gut microbiome. In the subsampled dataset the 10 and 5 most common families accounted for 91.6% and 81.3% of all reads, respectively (FIG. 1). While no statistically significant differences were found with respect to commonly used alpha diversity indices (Chaol, ACE, Shannon, Inverse Simpson, data not shown), comparisons of the clustering of patient and control samples in dendrograms based on beta diversity metrics (Yue & Clayton theta, Morisita-Horn index and Bray-Curtis index, calculated with family-level data) showed a significant difference between groups (unweighted UniFrac P<0.02 and weighted UniFrac P<0.001 for all three indices).

The mean abundance of Prevotellaceae in the feces of PD subjects was reduced by 77.6% in comparison to control subjects (Q=0.031; Table 3). High levels of Prevotellaceae were rare in the PD group whereas low levels were found in both groups (FIG. 2). Samples devoid of Prevotellaceae were equally frequent in both groups (PD: 29.2%; control: 27.8%; P=1.000; Fisher's exact test). The explorative analysis suggested that five families were more abundant in patients than in controls, but the absolute differences between groups were smaller than for Prevotellaceae (Table 3).

TABLE 3 Bacterial families showing different abundances between PD and control groups with a P-value less than 0.05. Taxonomy Patients* Controls* P-Value Q-Value Prevotellaceae 2.70 ± 0.32 12.06 ± 3.73 0.001 0.031 Lactobacillaceae 0.44 ± 0.04 0.02 ± 0.00 0.004 0.063 Verrucomicrobiaceae 0.06 ± 0.00 0.02 ± 0.00 0.014 0.146 Bradyrhizobiaceae 0.16 ± 0.00 0.03 ± 0.00 0.021 0.151 Clostridiales Incertae Sedis IV 2.49 ± 0.30 1.01 ± 0.03 0.025 0.151 Ruminococcaceae 33.63 ± 1.99 28.54 ± 1.81 0.029 0.151 *mean % ± variance

Mean abundance and variation of taxa at the genus level (Table 4a) and family level (Table 4b) that showed different abundances in PD and control groups (Metastats analysis) are shown in Table 4a and b. Data is based on same sample as family data.

TABLE 4a Mean abundance and variation of taxa at the genus level that showed different abundances in PD and control groups Parkinson Control mean variance stderr mean variance stderr Genus (group1) (group1) (group1) (group2) (group2) (group2) p-value q-value Prevotella 0.013423 0.001995 0.005264 0.104250 0.035548 0.022220 0.000999 0.021104 Lactobacillus 0.004358 0.000390 0.002329 0.000216 0.000001 0.000087 0.000999 0.021104 Saccharofennentans 0.047272 0.002385 0.005755 0.029738 0.000912 0.003560 0.016983 0.198570 Anaerotruncus 0.024364 0.000787 0.003307 0.015238 0.000310 0.002074 0.018981 0.198570 Mahella 0.025262 0.003024 0.006480 0.010022 0.000322 0.002114 0.020979 0.198570 Sutterella 0.005485 0.000075 0.001019 0.010182 0.000217 0.001735 0.020979 0.198570 Agromonas 0.001469 0.000021 0.000545 0.000306 0.000001 0.000144 0.022977 0.198570 Phaeovibrio 0.003355 0.000175 0.001561 0.000438 0.000003 0.000203 0.041958 0.332389

TABLE 4b Mean abundance and variation of taxa at the family level that showed different abundances in PD and control groups Parkinson Control mean variance stderr mean variance stderr Taxonomy (group1) (group1) (group1) (group2) (group2) (group2) p-value q-value Prevotellaceae 0.0270 0.0032 0.0066 0.1206 0.0373 0.0228 0.0010 0.0313 Lactobacillaceae 0.0044 0.0004 0.0023 0.0002 0.0000 0.0001 0.0040 0.0627 Verrucomicrobiaceae 0.0006 0.0000 0.0002 0.0002 0.0000 0.0000 0.0140 0.1462 Bradyrhizobiaceae 0.0016 0.0000 0.0006 0.0003 0.0000 0.0001 0.0210 0.1514 Clostridiales_Incertae_Sedis_IV 0.0249 0.0030 0.0065 0.0101 0.0003 0.0022 0.0250 0.1514 Ruminococcaceae 0.3363 0.0199 0.0166 0.2854 0.0181 0.0158 0.0290 0.1514

It was investigated whether antiparkinsonian medication was a possible confounder by testing each of the abovementioned families for significantly different abundances between users and non-users of each antiparkinsonian drug class in the PD group. Patients using COMT-inhibitors had significantly higher levels (median % [IQR]) of Lactobacillaceae (0.22 [0-2.07] vs. 0 [0-0.02]; P=0.001) and lower levels of Clostridiales Incertae Sedis IV (0.00 [0-0.20] vs. 0.84 [0.02-2.73]; P=0.005) than those not using this class of drugs. No significant differences were found for any other antiparkinsonian drug class.

To estimate effects of possible confounders (Table 1) on the observed group differences GLMs was applied to model the distribution of bacterial abundances (Table 5). For those bacterial families that were more abundant in PD subjects, Prevotellaceae abundance was also included in the model to disentangle a group related effect from an unspecific effect compensating low Prevotellaceae levels. Study group was the only factor that was significantly associated with Prevotellaceae abundance (Table 5). Thus, the decreased abundance of Prevotellaceae was not explained by e.g. more severe constipation in the PD group or differences in medications or medical history. The abundances of all the other families except for Ruminococcaceae were independently related to PD diagnosis. Ruminococcaceae abundance was significantly associated only with levels of Prevotellaceae suggesting that the higher levels of Ruminococcaceae in the PD group were not related to PD itself, but rather compensating lower levels of Prevotellaceae (Table 5). Abundances of Verrucomicrobiaceae and Bradyrhizobiaceae were associated with the severity of NMS and in particular constipation. Abundances of Lactobacillaceae, Bradyrhizobiaceae, and Clostridiales Incertae Sedis IV were associated with statin medication and abundance of Lactobacillaceae also with previous TIA/ischaemic stroke. The independent associations of bacterial abundances with PD diagnosis remained significant when subjects fulfilling Rome-III criteria for IBS were excluded from the analysis.).

TABLE 5 GLM results Results of the GLMs for bacterial abundances (sequence counts) based on the group factor and possible confounders. Normal distribution for Ruminococcaceae. Negative binomial distribution with log link for all others. COMT-inhibitor medication effect was nested within the study group effect for Lactobacillaceae and Clostridiales Incertae Sedis IV. Results shown as: B [95% CI], Wald Chi-Square, P-Value TIA or Atrial ischaemic fibrillation stroke Warfarin Statin Family Control vs. PD No vs. Yes No vs. Yes No vs. Yes No vs. Yes Prevotellaceae 1.490 0.432 0.025 −0.145 −0.096 [1.005-1.976], [−0.489-1.353], [−0.517-0.567], [−1.218-0.928], [−0.542-0.349], 36.179, <0.001 0.845, 0.358 0.008, 0.927 0.070, 0.791 0.180, 0.671 Lactobacillaceae −5.095 0.017 1.754 −0.258 −1.859 [−6.190-−4.000], [−1.502-1.535], [0.958-2.549], [−1.943-1.427], [−2.373-−1.344], 83.151, <0.001 0.000, 0.983 18.667, <0.001 0.090, 0.764 50.134, <0.001 Verrucomicrobiaceae −1.126 1.031 0.228 −1.768 0.081 [−1.784-−0.468], [−0.714-2.775], [−0.475-0.931], [−3.654-0.117], [−0.496-0.658], 11.254, 0.001 1.341, 0.247 0.405, 0.525 3.378, 0.066 0.076, 0.783 Bradyrhizobiaceae −2.368 −0.109 −0.440 −0.213 0.687 [−3.069-−1.666], [−1.307-1.088], [−1.237-0.356], [−1.631-1.204], [0.004-1.370], 43.748, <0.001 0.032, 0.858 1.174, 0.279 0.087, 0.768 3.891, 0.049 Clostridiales 1.441 0.173 −0.080 0.481 −0.856 Incertae Sedis IV [0.613-2.269], [−0.418-0.764], [−0.549-0.388], [−0.208-1.170], [−1.303-−0.409], 11.628, 0.001 0.330, 0.565 0.113, 0.737 1.871, 0.171 14.116, <0.001 Ruminococcaceae −52.526 −243.780 96.856 190.885 −112.421 [−317.204-212.151], [−719.394-231.834], [−175.166-368.877], [−347.816-729.585], [−339.258-114.416], 0.151, 0.697 1.009, 0.315 0.487, 0.485 0.482, 0.487 0.944, 0.331 Total NMSS Total Wexner Prevotellaceae COMT- Score (Z- Score (Z- abundance inhibitor Family transformed) transformed) (Z-transformed) No vs. Yes Prevotellaceae 0.038 −0.040 [−0.154-0.230], [−0.270-0.191], 0.152, 0.696 0.114, 0.736 Lactobacillaceae −0.199 −0.015 0.377 −3.998 [−0.590-0.193], [−0.280-0.250], [0.142-0.611], [−4.756-−3.241], 0.988, 0.320 0.012, 0.911 9.925, 0.002 107.025, <0.001 Verrucomicrobiaceae −0.340 0.406 −0.954 [−0.626-−0.054], [0.141-0.672], [−1.523-−0.385], 5.444, 0.020 9.001, 0.003 10.801, 0.001 Bradyrhizobiaceae −0.376 −0.595 −0.285 [−0.703-−0.049], [−0.888-−0.301], [−0.618-0.048], 5.076, 0.024 15.769, <0.001 2.813, 0.094 Clostridiales −0.144 0.153 −0.712 2.237 Incertae Sedis IV [−0.397-0.109], [−0.074-0.380], [−0.938-−0.485], [1.557-2.918], 1.239, 0.266 1.742, 0.187 37.966, <0.001 41.535, <0.001 Ruminococcaceae −27.605 95.836 −212.358 [−157.321-102.112], [−16.063-207.736], [−311.798-−112.917], 0.174, 0.677 2.818, 0.093 17.519, <0.001

A receiver operating characteristic (ROC) curve analysis of Prevotellaceae abundance was performed with respect to discrimination between PD patients and controls (FIG. 3). Using the optimal inclusive cut-off of 6.5% for classifying a subject with low Prevotellaceae abundance into the PD group, PD patients were identified with 86.1% sensitivity but only 38.9% specificity (LR+=1.41, LR−=0.36). It was evaluated whether inclusion of other bacterial families improved the discriminative power by performing a logistic regression analysis with study group as the dependent variable and all six bacterial families as covariates (likelihood ratio based backward elimination method). This model retained Prevotellaceae, Lactobacillaceae, Bradyrhizobiaceae and Clostridiales Incertae Sedis IV as predictors of study group (Table 6).

TABLE 6 Logistic Regression Results of logistic regression analysis with study group as the dependent variable. Initially all six bacterial families were included as covariates. We employed a backward elimination method based on the likelihood ratio. OR per 1% increase in abundance for belonging to PD group [95% Covariate Wald P-Value CI] Prevotellaceae 7.478 0.006 0.945 [0.907-0.984] Lactobacillaceae 1.805 0.179 23.441 [0.235-2337.377] Bradyrhizobiaceae 3.965 0.046 15.788 [1.044-238.716] Clostridiales Incertae Sedis 2.166 0.141 1.112 IV [0.966-1.280]

This classifier (FIG. 3) achieved a significantly higher AUC than Prevotellaceae alone (P=0.020 one-sided) and provided 90.3% specificity and 47.2% sensitivity (optimal probability cut-off 0.54; LR+=4.86, LR−=0.58). When Wexner total score as a clinical measure of constipation was included (Table 7) the discriminative power was further increased significantly (P=0.022 one-sided) due to better sensitivity (66.7%) and preserved 90.3% specificity (optimal probability cut-off 0.55; LR+=6.86, LR−=0.37; FIG. 3). The AUC of this model was also larger than that of Wexner score alone, but this difference just missed the level of significance (P=0.055 one-sided).

TABLE 7 Logistic Regression Results of logistic regression analysis with study group as the dependent variable after adding Wexner constipation score to the model. Based on Z-transformed values. OR per 1 unit Z-score increase for belonging to Covariate Wald P-Value PD group [95% CI] Prevotellaceae 7.053 0.008 0.429 [0.230-0.801] Lactobacillaceae 1.752 0.186 156.372 [0.088-277179.530] Bradyrhizobiaceae 3.562 0.059 2.590 [0.964-6.957] Clostridiales Incertae 1.084 0.298 1.349 Sedis IV [0.768-2.371] Wexner total score 17.742 <0.001 2.969 [1.789-4.927]

Enterobacteriaceae were significantly more abundant in patients with a PIGD phenotype than in TD patients (Q=0.018; FIG. 4, Table 8).

TABLE 8 Bacterial families showing differences in abundances between TD and PIGD phenotypes with a p-value less than 0.05. Taxonomy TD* PIGD* P-Value Q-Value Enterobacteriaceae 0.28 ± 0.00 2.31 ± 0.26 0.004 0.018 Erysipelotrichaceae 1.17 ± 0.02 2.19 ± 0.04 0.024 0.103 *mean % ± variance

Potential confounders such as gender ratio, medical history, medication, Wexner score, or time from motor symptom onset did not differ significantly between TD and PIGD subgroups. However, PIGD subjects tended to be older (66.3±5.7 vs. 63.4±5.2 years; mean±SD; P=0.051) and to have higher NMSS total scores (42.0 [32.0-69.5] vs. 33.0 [19.0-51.0]; P=0.053) than TD subjects. The only antiparkinsonian drug class significantly associated with Enterobacteriaceae abundance was COMT-inhibitors (users: 0.56 [0.24-7.40]; non-users: 0.09 [0.02-0.62]; median % [IQR]; P=0.043). A GLM was used to quantify effects of these possible confounders versus the effect of PD phenotype on Enterobacteriaceae abundance. In this model, PIGD and akinetic-rigid subscores, were positively associated with Enterobacteriaceae abundance whereas the negative association with tremor subscore slightly missed the level of significance (Table 9). When patients with motor fluctuations or DBS treatment were excluded from the analysis (n=43; Table 9), the association with PIGD subscore remained highly significant, but not the association with akinetic-rigid subscore. However, in this model the COMT-inhibitor medication showed a positive association with Enterobacteriaceae abundance.

TABLE 9 GLM results Results of the GLMs for Enterobacteriaceae abundance (sequence counts) based on phenotype subscores and possible confounders. Negative binomial distribution with log link. Results shown as: B [95% CI], Wald Chi-Square, P-Value Tremor PIGD Akinetic rigid NMSS total subscore (Z- subscore (Z- subscore (Z- score (Z- Age (Z- COMT-inhibitor transformed) transformed) transformed) transformed) transformed) No vs. Yes Enterobacteriaceae −0.254 0.931 0.445 −0.237 0.156 −0.712 (all patients [−0.517-−0.009], [0.500-1.362], [0.134-0.757], [−0.638-0.164], [−0.150-0.462], [−1.495-0.071], included; n = 72) 3.572, 0.059 17.955, <0.001 7.857, 0.005 1.340, 0.247 0.996, 0.318 3.173, 0.075 Enterobacteriaceae −0.143 2.643 −0.210 −0.344 −0.366 −3.502 (motor fluctuations [−0.619-0.332], [1.698-3.588], [−0.695-0.274], [−1.065-0.377], [−0.920-0.188], [−5.836-−1.167], and DBS excluded; 0.349, 0.555 30.075, <0.001 0.724, 0.395 0.876, 0.349 1.676, 0.196 8.643, 0.003 n = 43)

TABLE 10 Mean abundance and variation of taxa at the genus level that showed different abundances in TD and PIGD groups TD PIGD mean variance stderr mean variance stderr Taxonomy (group1) (group1) (group1) (group2) (group2) (group2) p-value q-value Clostridium XVIII 0.001585 0.000002 0.000292 0.004567 0.000046 0.001068 0.000999 0.002519 Anaerofilum 0.000213 0 0.000038 0.000456 0 0.000071 0.002997 0.007337 Papillibacter 0.000976 0.000001 0.000215 0.003389 0.000036 0.000953 0.006993 0.016638 Succiniclasticum 0 0 0 0.000617 0.000014 0.000588 0.006993 0.016638 Klebsiella 0 0 0 0.001206 0.000018 0.000677 0.008991 0.021095 Escherichia_Shigella 0.002966 0.000044 0.00139 0.02115 0.002558 0.007997 0.017982 0.03997 Paludibacter 0.000348 0.000001 0.000162 0.003461 0.000147 0.001916 0.020979 0.04486 Anaerostipes 0.00257 0.000016 0.00083 0.00055 0.000001 0.000183 0.020979 0.04486

It was found that, although no subject had a known diagnosis of active irritable bowel syndrome (IBS), 18 of 72 (25.0%) of PD patients but only 4 of 72 (5.6%) of control subjects fulfilled Rome-III criteria for IBS (Fisher's exact test p=0.002). While this prevalence in the control group matched that reported in previous studies, this is the first documentation of a clearly increased prevalence of IBS symptoms in PD patients indicating that IBS may be a non-motor symptom of PD.

REFERENCES

Savica R, Carlin J M, Grossardt B R et al. Medical records documentation of constipation preceding parkinson disease: A case-control study. Neurology 2009; 73(21):1752-1758.
Noyce A J, Bestwick J P, Silveira-Moriyama L et al. Meta-analysis of early nonmotor features and risk factors for parkinson disease. Ann Neurol 2012; 72(6):893-901.
Kieburtz K, Wunderle K B. Parkinson's disease: Evidence for environmental risk factors. Mov Disord 2013; 28(1):8-13.
de Vos W M, de Vos E A. Role of the intestinal microbiome in health and disease: From correlation to causation. Nutr Rev 2012; 70 Suppl 1:S45-56.
Segata N, Izard J, Waldron L, Gevers D, Miropolsky L, Garrett W S, Huttenhower C. Metagenomic biomarker discovery and explanation. Genome Biol. 2011 Jun. 24; 12(6):R60. doi: 10.1186/gb-2011-12-6-r60.
Arumugam M, Raes J, Pelletier E et al. Enterotypes of the human gut microbiome. Nature 2011; 473 (7346): 174-180.
Jankovic J, McDermott M, Carter J et al. Variable expression of parkinson's disease: A base-line analysis of the DATATOP cohort. the parkinson study group. Neurology 1990; 40(10): 1529-1534.
Poletti M, Frosini D, Pagni C et al. The association between motor subtypes and alexithymia in de novo parkinson's disease. J Neurol 2011; 258(6):1042-1045.
Edwards U, Rogall T, Blocker H et al. Isolation and direct complete nucleotide determination of entire genes. characterization of a gene coding for 16S ribosomal RNA. Nucleic Acids Res 1989; 17(19):7843-7853.
Lane D J. 16S/23S rRNA sequencing. In: Stackebrandt,E. Goodfellow, M. Eds. Nucleic Acid Techniques in Bacterial Systematics. 1991; p 115-175.
Schloss P D, Westcott S L, Ryabin T et al. Introducing mothur: Open-source, platform-independent, community-supported software for describing and comparing microbial communities. Appl Environ Microbiol 2009; 75(23):7537-7541.
White J R, Nagarajan N, Pop M. Statistical methods for detecting differentially abundant features in clinical metagenomic samples. PLoS Comput Biol 2009; 5:e1000352.

Claims

1.-13. (canceled)

14. A composition for use in treatment or prevention of Parkinson's disease (PD) in an individual, wherein the composition increases abundance of intestinal Prevotellaceae in said individual.

15. The composition of claim 14, wherein the composition comprises live or killed microbes and optionally a pharmaceutically acceptable carrier.

16. The composition of claim 14, wherein the composition comprises live or killed microbes and a pharmaceutically acceptable carrier.

17. The composition of claim 14, wherein the composition comprises one or more members selected from the group consisting of probiotics, other micro-organisms, prebiotics, antibiotics, growth factors, and bacteriophages.

18. The composition of claim 14 wherein the composition comprises Prevotellaceae.

19. The composition of claim 14, wherein the composition decreases intestinal abundance of one or more of taxa selected from the group consisting of Lactobacillaceae, Verrucomicrobiaceae, Bradyrhizobiaceae, Clostridiales Incertae Sedis IV, Ruminococcaceae, Lachnospiraceae, Enterobacteriaceae, Bacteroidaceae, Alistipes, Coprococcus, Anaerotruncus, Porphyromonadaceae, Parabacteroides, Butyricicoccus, Collinsella, Eggerthella, Coriobacteriaceae, Sporobacterium, Actinobacteria, Blautia, Enterococcus, Acetivibrio, Allobaculum Agromonas, Firmicutes, Mahella, Phaeovibrio, Porphyromonadaceae, Oscillibacter, and Saccharofermentans.

20. The composition of claim 14, wherein the composition is in the form of a food composition, pharmaceutical composition, nutraceutical, supplement or an anaerobical microbiota culture.

21. A method for treatment or prevention of Parkinson's disease (P D) in an individual comprising administering to said individual a composition that increases the abundance of intestinal Prevotellaceae in said individual when administered to said individual.

22. The method of claim 21, wherein the composition comprises live or killed microbes and optionally a pharmaceutically acceptable carrier.

23. The method of claim 21, wherein the composition comprises one or more members selected from the group consisting of probiotics, other micro-organisms, prebiotics, antibiotics, growth factors, and bacteriophages.

24. The method of claim 21, wherein the composition comprises Prevotellaceae.

25. The method of claim 21, wherein the composition is a stool transplant.

26. The method of claim 25, wherein a donor of the stool transplant is selected based on high abundance of Prevotellaceae and/or low abundance of one or more of taxa selected from the group consisting of Enterobacteriaceae, Lactobacillaceae, Ruminococcaceae, Bacteroidaceae, Lachnospiraceae, Verrucomicrobiaceae, Bradyrhizobiaceae, Clostridiales Incertae Sedis IV, Alistipes, Coprococcus, Anaerotruncus, Porphyromonadaceae, Parabacteroides, Butyricicoccus, Collinsella, Eggerthella, Coriobacteriaceae, Sporobacterium, Actinobacteria, Blautia, Enterococcus, Acetivibrio, Allobaculum Agromonas, Firmicutes, Mahella, Phaeovibrio, Porphyromonadaceae, Oscillibacter, and Saccharofermentans.