FR2935712A1 - METHOD FOR DETECTING THE DIFFERENTIAL EXPRESSION OF A MOLECULAR MARK SET ASSOCIATED WITH PARKINSON'S DISEASE - Google Patents

Abstract

The present invention provides a method for detecting the differential expression of a set of molecular markers that is associated with the presence or risk of developing Parkinson's disease, and in particular a sporadic form of this disease. The invention also relates to a kit or necessary for the implementation of this method.

Description

The present invention relates to the field of diagnosing Parkinson's disease in a subject. More specifically, the present invention essentially relates to a method for detecting the differential expression of a set of molecular markers that is associated with the presence or risk of developing Parkinson's disease, and in particular a sporadic form of this illness. Parkinson's disease is a common neurodegenerative disease characterized by degeneration of neurons of the nigrostriatal axis and following a slow and irreversible progression. The average age of diagnosis of the disease is around 57, at a time when individuals are still working. Parkinson's disease is the second neurodegenerative disease after Alzheimer's disease and affects about 2% of the population of industrialized countries beyond 65 years. The prevalence of this disease increases with the age of the population and the current demographic evolution.

The majority of patients with Parkinson's disease have no family history and develop a form of the so-called sporadic disease. The etiology of their disease is poorly understood. It is assumed that its development is the result of a multitude of environmental and genetic factors, most of which have not been identified to date.

On the other hand, certain hereditary forms (also called familial forms) have been identified during the last ten years, and are currently being studied. These family forms account for between 5 and 15% of cases. They have an autosomal dominant or recessive mode of transmission. More than 13 loci have been linked to the disease and the inventors identified a new locus on chromosome 3 in a familial form of autosomal dominant inheritance. The gene of this locus responsible for the disease is involved in the mRNA translation mechanism and is called EIF4G1. It is in addition to the 10 other genes already described in Parkinson's disease carrying deleterious mutations, here the most frequently mutated genes. The parkin gene (PARK2 or PRKN) is responsible for 50% of early forms (development of symptoms before age 40) of the disease. However, mutations of this gene represent only 0.4% of all cases. The most frequent mutations occur in autosomal dominant forms and concern the SNCA (alpha-synuclein), SCA2 and especially LRRK2 genes. The latter is thought to be responsible for 3% of Caucasian familial forms and about 1% of sporadic cases of the disease. In rare cases, this proportion can reach 18% of patients in some Arab-North African populations and 39% of Ashkenazi Jewish populations. These high percentages are related to cultural habits and a significant percentage of consanguinity. The LRRK2 gene encodes a kinase that occurs at a metabolic junction in different parkinsonian syndromes and related diseases (synucleopathies and tauopathies, which are characterized by aggregates of alpha-synuclein and Tau proteins respectively). This connection between parkinsonian syndromes and other neurodegenerative diseases is also illustrated by the mutations affecting the SCA2 gene; some mutations are responsible for a typical Cerebellar Ataxia (SCA2-C) while others give rise to Parkinson's disease (SCA2-P). In the same way the SNCA gene codes for alpha-synuclein which can aggregate and form inclusions like Lewy bodies; its abnormal aggregation characterizes the family of diseases called synucleopathies including Parkinson's Disease, Lewy Body Dementia (DLB), and Multi-Systemic Atrophy (MSA).

Although 10 genes responsible for rare forms of Parkinson's disease have already been identified and their identification is important to better understand the mechanisms leading to the disease, these genes do not explain the majority of sporadic cases of the disease. In fact, the impact of environmental faders is preponderant in Parkinson's disease as shown by the absence of a difference in agreement between monozygotic and dizygotic twins in the epidemiological studies carried out. If the concordance rate was higher in the monozygotic than dizygotic twins, this would show an impact of genetic factors in the development of Parkinson's disease. These studies highlight the important role played by environmental factors for the majority of patients. Several animal models of the disease have been made using pesticides as an environmental causative agent of the disease. However, it is possible that although less, the impact of genetic factors may act through gene-environment interactions. Indeed, the inventors have demonstrated in the past that such interactions were possible. For example, some people exposed to pesticides had an increased risk of developing the disease when they had the CYP2D6 genotype, whereas this risk was zero for subjects with different genotypes exposed to the same factors. During his life, the individual will be exposed to many environmental faders, which depending on its genetic heritage, may be favorable or deleterious. Since Parkinson's disease is a disease of the elderly, many environmental factors will affect certain metabolic pathways before the onset of symptoms. Moreover, when these first symptoms emerge, more than 50% of the dopaminergic neurons of the substantia nigra are dead and the deficiency of dopamine in the striatum exceeds 70%. This chronic condition is characterized by a long silent period during which the individual has no symptoms. However, on the molecular and cellular level, minute modifications occur and add up, particularly within the nigrostriatal pathway.

The diagnosis of Parkinson's disease is made only by the insidious onset of asymmetrical symptoms that have progressed slowly for several years. There is no diagnosis at the early stage of the disease. At this stage, we often bet on parkinsonian syndromes. Indeed, the development of DLB or MSA for example can mimic Parkinson's disease or the onset of this disease. Three cardinal signs define a parkinsonian syndrome: - resting tremor - akinesia / bradykinesia - hypertonia / muscle rigidity Parkinson's syndrome is present in Parkinson's disease but can also be found in other conditions such as DLB, some forms of MSA and Progressive Supranuclear Palsy (PSP) ... However Parkinson's disease differs from other parkinsonian syndromes by a fourth cardinal sign postural instability of late onset and a positive response to treatment with levodopa (L-dopa). Most people with Parkinson's Syndrome have resistance to this treatment. In addition, several exclusion criteria have been defined in order to establish the diagnosis of Parkinson's disease. These similarities of symptoms contribute to the difficulty of establishing an initial diagnosis of Parkinson's disease. An evolution of clinical manifestations over a period of at least five years is necessary to confirm or refute the initial diagnosis of Parkinson's disease. A definite diagnosis of this disease can only be made after a post-mortem neuropathological examination.

Several sets of clinical criteria have been proposed to try to make a diagnosis of Parkinson's disease during the patient's lifetime. It should be noted, however, that at present there are no common international criteria for this diagnosis. The most commonly used diagnostic criteria in clinical practice are those of the UK Parkinson's Disease Society Brain Bank (UK PDSBB), Hughes A.J., Journal of Neurology, Neurosurgery, and Psychiatry, 1992, 55, 181-184; Gelb D, JArch Neurol. 1999 Jan; 56 (1), 33-9. They often classify the certainty of the diagnosis into three levels: defined, probable and possible. A definite diagnosis holds the four cardinal criteria (bradykinesia / akinesia, resting tremor, stiffness, postural instability) and confirmation by autopsy.

A probable diagnosis uses three criteria on the four previous ones, often with additional prospective criteria. Indeed, this diagnosis based on clinical arguments is difficult to establish with certainty, especially at the early stage of the disease because of the similarity of symptoms between Parkinson's disease and Parkinson's syndromes. As a result, most of the scientific studies carried out employ strict criteria for the diagnosis of Parkinson's disease, taking into account three of the four criteria including bradykinesia and the positive response to levodopa and the absence of the 16 exclusion criteria defined by Hughes ( above). A possible diagnosis usually begins with two of the four cardinal signs of the disease, the criterion of bradykinesia being required to hold the diagnosis as possible. In order to quantify therapeutic improvement in patient follow-up, the Unified Parkinson's Disease Rating Scale (UPDRS) is the most widely used by clinicians. This scale is an objective assessment of the severity, fluctuation and progression of clinical symptoms and their impact on the patient's daily life. This scale is not essential for diagnosis and follow-up but may be necessary for certain therapeutic decisions, particularly surgical decisions. This multidimensional scale has four parts and is assigned scores: - the mental state of the patient and his behavior, (part I: maximum score 16) his activities of daily living during the period of pharmacological treatment (ON period); and absence of pharmacological treatment (OFF period) (part II, maximal score: 52) motor examination performed with and more rarely in the absence of treatment (ON and OFF) (part III, maximum score: 108) treatment complications . (part IV: maximum score 23) The UPDRS scale includes, in addition to these four parts, a ranking of the 5 patients according to the Hoehn and Yahr (HY) scale. The HY scale has 8 stages (Table 1) that estimate the severity of Parkinson's disease in a patient. Stage Clinical events Stage 0 No evidence of illness Stage 1 Unilateral signs causing no handicap in daily life Stage 1.5 Unilateral attack and axial involvement Stage 2 Predominantly unilateral signs leading to a certain handicap Stage 2.5 Bilateral mild attack to moderate; some postural instability Stage 3 Bilateral impairment with postural instability; Autonomous patient Stage 4 Severe handicap but possibility of walking; partial loss of autonomy Stage 5 Sick in wheelchairs or bedridden; is no longer autonomous. Table 1: Ladder of Hoehn and Yahr

The age of onset and extent of functional impairment are the two main factors that guide therapeutic choices. Although there is currently no treatment to stop or slow the progression of this disease, new treatments for protective purpose are being evaluated in clinical trials (antioxidants, glutamatergic antagonists, neurotrophic factors, ligands neuro-immunophilins or anti-apoptics, etc.). Ideally, these new therapies should be adapted to each stage of the disease to aim for optimal efficacy. It is therefore very important to effectively target patients who may benefit from neuroprotective treatments. As such, the diagnosis of Parkinson's disease in its early phase is essential and would make it possible to provide a treatment adapted to the patient. The development of new functional genomics technologies has led to the first transcriptomic studies of brain tissue from patients with neurodegenerative diseases and controls. In particular, these data show that there are a number of genes whose expression is specifically deregulated in patients with Parkinson's disease (WO2005 / 067391). However, these studies only reflect the final changes in the disease since these data are obtained from brain tissue samples of deceased patients. These studies also include in their results the many molecular perturbations related to cell death and post-mortem delays. Under these conditions, it is difficult to say that the observed variations in gene expression mirror the early stage of Parkinson's disease.

The diagnosis of Parkinson's disease is therefore complex and can be called into question during the evolution of the disease. There is currently no specific biological signature of Parkinson's disease that can diagnose or predict development of this early stage disease. The provision of an effective and simple diagnostic test, for an early stage, would enable patients to be treated early in their illness and thus benefit from a more effective treatment, more adapted to their needs. and offer them neuroprotective strategies. There is therefore an important need for a diagnostic / prognostic test to determine, while alive, whether subjects have Parkinson's disease and / or to predict whether subjects are at risk of developing this disease. There is also an important need to be able to distinguish patients likely to develop Parkinson's disease from those who are likely to develop another Parkinson's syndrome, in particular of the DLB type, MSA who can develop in the early phase the same clinical symptoms. The present invention provides solutions to these needs. By analyzing the differential expression of a set of molecular markers that are differentially expressed specifically in familial forms of the disease, a tool is available to detect the presence or risk of developing this disease, even in asymptomatic or pauci-symptomatic subjects, or to confirm a diagnosis in patients with symptoms. The invention also makes it possible to identify patients with Parkinson's syndrome from those with Parkinson's disease. The international application published under the number W02006 / 050475 describes a method for diagnosing neurodegenerative diseases from a blood sample in which the expression of one or more genetic markers is measured and then compared with that of said marker (s) ( s) in a healthy subject to determine if the subject has a neurodegenerative disease. A list of 8 genes would differentiate subjects at risk of developing a sporadic form of Parkinson's disease. As in all current studies to determine which genes are deregulated in sporadic forms of neurodegenerative diseases, the familial criterion has been excluded from the diagnosis of Parkinson's disease. Indeed, the authors seek to identify molecular markers that would be involved in the development of this disease whose etiology is unknown. Family hereditary forms do not fit into this definition since, in essence, their cause is known; the development of the disease being linked to the presence of a mutation. It is generally accepted by the scientific community that each carrier of pathogenic mutation develops a particular case (a variant) of Parkinson's disease. Indeed, a pathogenic mutation of a gene will generate a gain or loss of function of the encoded protein. These modifications will preferentially target a biological mechanism and lead to specific modifications in biological networks. For simplicity, it could be said that each mutated gene will result in a distinct form of disease in relation to another mutated gene in the sense that it only highlights a small part of the potentially important mechanisms for understanding all cases. of the disease. In addition, it is recognized that the role of environmental factors is preponderant for the occurrence of sporadic cases and that familial forms account for only 5 to 15% of the total forms of Parkinson's disease. In addition, SNCA, LRRK2, SCA2 and PRKN gene mutations account for less than 2% of all cases. Family forms are thus rare forms appearing at an often early age (often before 40 years) while sporadic forms are observed to develop later (on average after 57 years). Therefore, statistically the changes observed for each of these distinct forms of Parkinson's disease can not fully explain all cases, especially sporadic forms. The inventors refute this misconception and they show on the contrary, and this constitutes the foundation of the present invention, that the taking into account of the combined data of the genes whose expression is deregulated in subjects with familial forms, thus rare forms Parkinson's disease, can identify a series of genes whose expression is deregulated in subjects who are sick or at risk of developing Parkinson's disease, regardless of the stage of severity of the disease, and even asymptomatic cases, and in particular sporadic forms. The method devised by the inventors based on the set of genes whose expression is deregulated in familial forms due to rare mutations in the general Caucasian population makes it possible to identify subjects suffering from Parkinson's disease presenting with or not symptoms with a low error rate. Preferably, the method according to the invention makes it possible to determine the presence or the risk of developing Parkinson's disease with an error rate of less than or equal to 30%, and in particular to identify sporadic cases of this disease. The diagnostic criteria for Parkinson's disease according to the invention are based on those accepted in the field by the French Federation of Neurology, National Agency for Accreditation and Health, 2000. Specifically, the cumulative criteria 1) 2) and 3 ) following were used: 1) at least 3 of the 4 cardinal signs of Parkinson's disease, namely resting tremor, bradykinesia, muscle rigidity or postural instability; bradykinesia is necessarily part of one of the 3 cardinal signs selected for diagnosis, 2) the positive response to anti-parkinsonian levodopa treatment, and 3) the absence of the exclusion criteria defined by UK PDSBB, Hughes A.1 Journal of Neurology, Neurosurgery, and Psychiatry, 1992, 55, 181-184. The 16 th criterion of family history was not retained. Thus, within the meaning of the present invention, is meant by: - subject with a family form or family case, any subject of which at least one other member of his family is affected by Parkinson's disease. The concept of family history is that defined in Hughes et al, 1992 (supra). - subject suffering from a sporadic form or sporadic case, any subject having no hereditary form of Parkinson's disease. - asymptomatic subject, any subject, suffering from hereditary or sporadic form, having no symptoms of Parkinson's disease. - pauci-symptomatic subject, person with mild disorders but insufficient to make a diagnosis of Parkinson's disease according to the diagnostic criteria mentioned above. - symptomatic subject, person who developed Parkinson's disease whose symptoms are visible and unequivocal to make a diagnosis according to these same criteria. - early stage Parkinson's disease, the silent phase (asymptomatic subject) and the symptomatic pauci phase of the disease. In the context of the invention, the inventors have therefore not retained the notion of familial form of Parkinson's disease as being an exclusion criterion. People at risk are people with a family history, or people with known risk factors for developing Parkinson's disease (eg exposure to pesticides, dysregulation of biomarkers, etc.). By deregulated expression or deregulated genes is meant any gene whose level of expression in a subject is increased or decreased compared to that of a control subject. The expression variation threshold considered significant depends on each comparison technique. For example, for cDNA microarray analysis, an expression variation threshold of at least 1.2 with a Welch T-test during Volcano plot filtering is conventionally considered significant. Thus, a person skilled in the art is able to determine the expression variation threshold to be taken into account as a function of the detection and expression comparison technique used. In the context of the present invention, mutation is understood to mean any genetic disorder consisting of the replacement of a nucleotide base, the deletion of one or more nucleotide base (s), the deletion or the multiplication of exons. (1, 2, 3 or advantages of copies of exons), etc., but also variations in the number of gene copies such as the presence of a copy (deletion), a number of copies greater than 3 (duplication) or 4 copies (triplication) or more of a chromosomal region. According to a first aspect, the present invention therefore relates to a method for detecting in a biological sample to be tested the differential expression of a set of molecular markers associated with the presence or risk of developing Parkinson's disease, characterized in that it comprises the following steps: (a) in vitro extraction of molecular markers from the biological sample to be tested, (a2) in vitro extraction of molecular markers from biological samples of subjects with different familial forms of Parkinson's, subjects with other neurodegenerative diseases or parkinsonian syndromes and healthy subjects, b) determine all the molecular markers whose expression is significantly deregulated in common in subjects with different familial forms regardless of the stage of the disease but whose expression is not significantly deregulated in patients with other neurodegenerative diseases or parkinsonian syndromes, compared to that measured in healthy subjects, c) verify that the expression of all these molecular markers is well deregulated in the sample to be tested. The originality of the present invention therefore lies in the use of data (genes whose expression is deregulated relative to healthy individuals) that are derived from patients with familial forms of Parkinson's disease to determine a set of genes whose dysregulated expression is specifically associated with this disease, compared with the methods described in the prior art and in particular in WO2006 / 050475 and Scherzer et al., PNAS 2007, vol. 104, 3, 955-960 that do not specifically consider the data from these types of patients to predict the existence or risk of developing a sporadic form of Parkinson's disease. Indeed, the patients who are part of these studies meet the clinical diagnostic criteria of the United Kingdom Parkinson's Brain Bank Disease Society which excludes patients with a familial form of the disease. Thus, the inventors believe that this notion is to be questioned, and that is why they did not retain this last criterion of exclusion (exclusion of family forms) in the present invention. However, they used strict criteria for Parkinson's disease including the presence of three of the four cardinal criteria of the disease including bradykinesia, and the positive response to levodopa treatment and the absence of exclusion criteria other than the one concerning the exclusion of family forms.

In addition to the general diagnosis of Parkinson's disease, the major advantage of the method according to the invention is to be able to detect the presence or the risk of developing a sporadic form of Parkinson's disease. In the case of a positive diagnosis, the distinction between a family form and a sporadic form of Parkinson's disease will be determined on the basis of information already available about other family members (ie whether at least one other family member has already developed or not the disease).

The implementation of the present invention also makes it possible to determine the presence of Parkinson's disease even in persons suffering from asymptomatic forms of the disease. Preferably, the method according to the invention makes it possible to determine the presence or the risk of developing Parkinson's disease with an error rate of less than or equal to 30%. Patients with different familial forms of Parkinson's disease are referred to as patients with at least one member of their family with Parkinson's disease. Significantly deregulated gene expression data in subjects with familial forms of Parkinson's disease that may be used in the context of the present invention may in particular be derived from patients with different familial forms, and preferably carriers of mutations in the SNCA gene and / or patients carrying mutations in the PRKN gene and / or patients carrying mutations in the SCA2 gene and / or patients carrying mutations in the LRRK2 gene. According to a particularly advantageous embodiment, the gene expression data come from subjects suffering from the four familial forms, each of which is caused by mutation in the SNCA, PRKN, SCA2 and LRRK2 genes respectively. Preferably, when the expression data of the deregulated genes are in particular derived from patients carrying a mutation in the LRRK2 gene, the patients preferably have the G2019S mutation. Likewise, when the expression data of the deregulated genes are in particular derived from patients carrying mutations responsible for Parkinson's disease in the SNCA gene, these patients preferably have a multiplication of the SNCA gene, and preferably a duplication of the SNCA gene. . More preferably, when the expression data of the deregulated genes are in particular derived from patients carrying a mutation in the SCA2 gene, the patients preferably have a mutation of 39 interrupted CAG / CAA repeat triplet type. Finally, when the expression data of the deregulated genes are in particular derived from patients carrying a mutation in the PRKN gene, the patients preferably have a homozygous type mutation or a composite heterozygous mutation Trp445stop or ex3.4dupl or ex6del /c.1385insA or ex6dupl / c.225delA.

Moreover, the expression data of the deregulated genes in subjects suffering from other neurodegenerative diseases or parkinsonian syndromes can in particular be obtained from patients with different neurodegenerative diseases or different parkinsonian syndromes such as Spino-Cerebellar Ataxia (SCA). ), MSA, DLB, PSP, amyotrophic lateral sclerosis (ALS), and in particular these expression data come from both patients with ACS, patients with ASM and patients with DLB. More particularly, these expression data come from patients with ACS, MSA patients, DLB patients, PSP patients, and ALS patients. According to a particularly preferred embodiment, the set of biological markers whose expression is deregulated consists of the markers that are common to at least 1 patient carrying a multiplication of the SNCA gene, at least 4 patients carrying a mutation. in the PRKN gene, at least 6 parkinsonian patients carrying a mutation in the SCA2 gene and at least 9 patients carrying a mutation in the LRRK2 gene (in the case of patients with familial forms) and not at least 6 patients with SCA, at least 1 patient with MSA and at least 4 patients with DLB (for patients with other neurodegenerative diseases or parkinsonian syndromes). The biological sample that is used in the context of the present invention consists more particularly of any material from the patient, which may contain a biological marker that can detect the expression of a gene or its product. In particular, the biological sample may be a biological fluid (whole blood, serum, saliva, smears of the mucous membranes, circulating cells of the patient (blood or immune, such as platelets or mononuclear blood cells). or a tissue such as skin or cells derived from the skin such as fibroblasts, cells derived from muscle, hair, or hair), as long as it contains molecular markers that may reflect the expression of genes for example, such as RNAs or proteins. Preferably, the biological sample from the person to be tested is of the same type as that derived from the reference subjects (subjects suffering from various familial forms of Parkinson's disease, subjects suffering from other neurodegenerative diseases or Parkinson's syndromes and healthy subjects).

Various studies, and in particular those of the inventors, have shown that mononuclear cells can be used to study the changes involved in the brain tissue of subjects suffering from degenerative diseases, and in particular Parkinson's disease. Thus, the biological sample used in the context of the method according to the invention preferably consists of mononuclear cells derived from whole blood.

The biological markers according to the present invention are represented by any molecule whose detection corresponds or makes it possible to demonstrate the expression of genes, such as in particular a ribonucleic acid or a protein. According to a preferred embodiment of the invention, the molecular markers used are RNAs (in particular mRNAs) or proteins.

In the context of the present invention, it will be particularly preferred to use, as a biological sample, mononuclear cells isolated from blood, from which the RNAs which will be used as a biological marker for the implementation of the method can be extracted in particular. In this case, the biological sample is preferably obtained from a fasting patient.

The biological markers can be extracted from the biological samples by any conventional means well known to those skilled in the art and specifically adapted to the nature of the biological marker chosen. Furthermore, any technique for quantitatively measuring gene expression and comparing expression levels well known to those skilled in the art can be used for the implementation of the present invention. The measurement and comparison techniques will be chosen to be specifically adapted to the type of biological markers chosen. As examples of quantification method, mention may be made especially of techniques based on microchips, real-time PCRs or any other technique allowing quantification (sequencing, Western blotting, etc.). However, according to a preferred variant of the present invention, the deregulated expression of the molecular markers is determined by a quantitative measurement of the expression of the genes or the quantity of the proteins chosen from microchips and quantitative RT-PCR.

Thus, the study of the differential expression of the molecular markers carried out in step c) of the method makes use of the knowledge of a person skilled in the art who will know how to prepare the specific reagents (for example, the design of primers, chips, of antibodies, etc.) or choose these reagents from those already available commercially and analyze the differential expression of these molecular markers.

In particular, during the implementation of step c) of the method according to the invention, the specific reagents of the molecular markers identified in step b) can detect homologs of these markers. By homologues is meant any molecular marker whose sequence homology is at least 70%, and preferably at least 80%, at least 85% or even at least 90%, or at least 90%. minus 95% with the sequence of the molecular marker concerned which will have been identified in step b). The method according to the present invention may further comprise a step of validating the results obtained by means of a predictive analysis in order to determine the rate of error of prediction to be reached or to develop Parkinson's disease. step that can be performed using any method known to those skilled in the art is able to best predict the phenotype of new individuals from their expression data. Known methods include, but are not limited to, KNN (K Nearest Neighbors) method, closest centroid method (closest Shrunken centroid method also known as Prediction Analysis for Microarray PAM), SVM (Support Vector Machine) method, Linear Discriminant Analysis (LDA) method, ...

According to a particularly advantageous embodiment, the set of genes whose dysregulated expression is associated with Parkinson's disease is mentioned in the following Table 2: No. Chip Code Gene Name Code Genbank 1 A_232216712 Transient potential receptor of the channel to NM_017662 cation, subfamily M, member 6 (TRPM6) 2 A_24_P245838 Mannosyl (beta-1,4-) - glycoprotein beta-1,4-AK125361 N-acetylglucosaminyltransferase (MGAT3) 3 A_232369316 Vang-like 1 (VANGL1) NM 138959 4 A 2331904 Transmembrane domain 4, subfamily NM_000139 A, member 2 (MS4A2) 5 A 233349321 Xylosyltransferase type I (XYLT1) NM_022166 6 A_232117662 Histidine decarboxylase (HOC) NM_002112 7 11_232133916 Component C2 of complement (C2) NM_000063 8 A_322162187 Component C2 complement (C2) NM_000063 9 A_2328961 Interleukin 7 (IL7) NM_000880 10 11_322175313 EST PM1-NN1207-071200-003-a12 NM1207 BF960555 11 A_3226433 EST yr57d04.rl Complementary DNA of H64096 clone IMAGE: 209383 soares fetal liver spleen, similar to the protein of the 71KD12 human heat shock family A_322162183 C2 component of the complement (C2) NM_000063 13 A_322104572 EST UI-E-DW1-ahe-d-15-0-UL1 BM670971 clone Complementary DNA UI-E-DW1-ahe-d-15-O-UI3 '14 A_322234485 EST UI-1-BB1p-a-g-04-0-UI.s1 BQ024427 NCI_CGAP_Pl6 Complementary DNA of clone UI-1-BB1p-auq- g-04-0-UI 3 'A_23260047 Protein 1 containing bi-spiral acid structures NM_006283 (TACC1) 16 A_32245875 EST ws88f04.xl Homo sapiens, AW009873 Complementary DNA of clone IMAGE: 2505055 3', similar to the sequence close to the phospholipase C TR: Q62084 Q62084 17 / 1_232352435 Signaling G protein regulator NM_002926 12 (RGS12) 18 11_232387649 DNAJA4 homologous to DnaJ (Hsp40), sub-AF116663 family A, member 4 (DNAJA4) or PRO1472 19 A _23_P200252 Open reading frame 57, chromosome 1 NM_032324 (C1orf57) A_23266174 SH3 binding domain of protein kinase NM_00102440 1 (SBK1) 21 A_232339 27 FKSG43 AF334945 A_23_P333640 Papiline (PAPLN) sulfated glycoprotein NM_173462 similar to a proteoglane chromosomal region of lateral sclerosis 23 A_24_P323682 amyotrophic 2 (juvenile) NM_020919 Table 2; Example of a set of genes whose dysregulated expression is associated with Parkinson's disease.

In a second aspect, the present invention relates to the use of specific reagents of a set of molecular markers identified according to step b) of the method described above for determining the presence or risk of developing Parkinson's disease. In the context of this use, the identification of the set of molecular markers, as well as the use of the specific reagents of these markers in order to study their differential expression, are implemented ex vivo and therefore not applied to the human body. animal. Preferably, reagents specific to all 23 genes mentioned in Table 2 above will be used. In this variant of the invention, the probes indicated in Table 3 below can in particular be used. Codes probes No. indicated Identifiers Sequence microchips probes in the sequence table Agilent 2 A_32_P45875 16 SEQ ID NO 1 CCAGAAATGAAGGCCTCTGTGGGIIIGTAACAGTCAACCAGCAGCTCCTTGACCGTGGCA A_32_P175313 10 SEQ ID NO 2 TTACTCAAGGGGTGTGTATGTGTGTGTCTGAACCAAGAAAGGTTGAGAGCTCCAGGTATT A_24_P245838 2 SEQ ID NO 3 ACATTCTGAAATCAIIIICTCTGTAAATGGTTGGATTTCATTTCACCCTTAAAGGGATGC A_23_P200252 19 SEQ ID NO: 4 CAAAGAGTGTGCGTCATCGATGAGATTGGGAAGATGGAGCTCTTCAGTCAGCIIiiCATT A_23_P133916 7 SEQ ID NO 5 TTCTACCTCTGAATGGCCACCCTTAGACCCTGTGATCCATCCTCTCTCCTAGCTGAGTAA A_32_P162183 12 SEQ ID NO 6 GAATGAACTGCTGAACAAACAGAGTGTTCCTGCTCATTTTGTCGCCTTGAATGGGAGCAA A_32_P162187 8 SEQ ID NO 7 GGTTGACTTGACTCATGCTTGIIICACTTTCACATGGAATTTCCCA (a1ATGAAATTAAT A_23_P33927 21 SEQ ID NO 8 TGCCCTTCTATGGGTGTGCCTTCTTCCACGGTGAGGTTGACAAGCCAGCCCAAGGLIIII A _23_P117662 6 SEQ ID NO 9 CCGAGGGTAGACAGGCAGCTTCTGTGGTTCAGCTTGTGACATGATATATAACACAGAAAT A_23_P8961 9 SEQ ID NO 10 TAAACCTTAAGTAAGCAACAGCATAACAAGGTCCAAGATACCTAAAAGAGATTTC AAGAG A_23_P1904 4 SEQ ID NO 11 TCAATGTCATCTTCTCCATGAAGACCACTGAATGAACACUIIIICATCCAGCCTTAATTT A_23_P333640 22 SEQ ID NO 12 TGACIIIIATTACCAACACACTGGCTGGAAAAGGGACAAACCACATCACGGGTGAGTGAT A_23_P387649 18 SEQ ID NO 13 AGTGAGCTGTATCACAACATGTATTGGGTGTGCCATGAATTCCTCTT (, IIICCCAGTCTT A_23_P352435 17 SEQ ID NO 14 TTCAGTGAGTGTGAAGGTGAACAGGATGGCCTGGCAGCAGGGATGIIICCGTGAGCCACA A_23_P66174 20 SEQ ID NO 15 CCTTCGGCGTGCTCATCTTCTGCGTGCTCACCGGCAACTTCCCGTGGGAGGCGGCGTCGG A_23_P60047 15 SEQ ID NO 16 TCCCCCTGTTAGCTCAACAGATCTGCATTTGGCTGCTTCTCTTGTGACCACAATTATCTT A_23_P216712 1 SEQ ID NO 17 TGCCCAATTGCCAAACTAAAGACATCAGTTCATTGGTCAAATATTI GTTACCTGGAATGG A_23_P369316-3 SEQ ID NO 18 CTGTCCACCCTGAAAAAGAGTGACTGATGACATCTGACIIIIGTCGATGGGACTTCTCAA A_23_P349321 5 SEQ ID NO 19 TGTGAGCACCAGACTCAAGACCATGACCTTC111GATCCTTGAAAATGGGAACTTTGACA A_32_P234485 14 SEQ ID NO 20 GCAACATAAATACAGAAGCGATGAACAGAAGACTCATAACCAATACTGGAACAGGGCCAA A_322104572 13 SEQ ID NO 21 GTTCAGGGAGGAAAAAACAGGGGCTAGTTGAATTTAGCCTTGGAAAAATCCATCTCCGGA A_32_P6433 SEQ ID NO: 22 ATGCCCAGTAACTTACTATAGCAGCTTAAC 1 1 1 1 1AAAACTGCCACAGAATTTGCTACAA A_24 P323682 23 SEQ ID NO. 23 TTGGGTTATTTCTCCATTTATGGCCTATGGTTCAGCAAGTCAACTCTTGAGGACCTGIII Table 3: Examples of probes specific for 23 genes whose expression is significantly deregulated in subjects with Parkinson's disease.

According to a third aspect, the subject of the present invention is a kit or kit necessary for carrying out the method described above comprising at least the following elements: specific reagents (such as primers or probes) of the set of molecular markers as identified according to step b) of the method of the invention described above, - optionally the reagents and / or devices necessary and adapted according to the type of molecular markers chosen to carry out the extraction molecular markers to be analyzed and / or the relative detection and / or quantification of these markers relative to the amount of molecular markers of the same type in a sample from a healthy subject (negative control), and - optionally a sample containing molecular markers of the type chosen from a patient with Parkinson's disease in a proven way (positive control).

According to a preferred embodiment, the invention relates to a kit or necessary for detecting in a biological sample to be tested the differential expression of a set of molecular markers associated with the presence or risk of developing Parkinson's disease, comprising at least one minus: - reagents specific for the 23 genes indicated in Table 2 above, - optionally the necessary reagents and / or devices adapted according to the type of molecular markers chosen to carry out the extraction of the molecular markers to be analyzed and / or the detection and / or relative quantification of these markers relative to the amount of molecular markers of the same type in a sample from a healthy subject (negative control), and - optionally a sample containing molecular markers of the chosen type from a patient with Parkinson's disease in a proven way (positive control). In particular, the invention relates to a kit or necessary for detecting in a biological sample to be tested the differential expression of a set of molecular markers associated with the presence or the risk of developing Parkinson's disease, comprising at least: specific reagents of the 23 genes indicated in Table 2 above, - optionally the reagents and / or devices necessary to carry out the extraction of the RNAs and / or the detection and / or the relative quantification of RNA with respect to the amount of RNA in a sample from a healthy subject (negative control), and possibly an RNA sample from a patient with Parkinson's disease in a proven way (positive control). Preferably, the specific reagents of the molecular markers identified according to step b) of the above method are chosen from primers, probes, antibodies, etc. Preferably, the positive control provided in the kits according to the invention is a sample from a subject having a pathogenic mutation responsible for Parkinson's disease at a disease stage HY of greater than or equal to 3. Preferably also, the negative control provided in the kits according to the invention is a sample derived from a subject free from any personal or family history of Parkinson's disease, without any other apparent disease (and in particular without other neurodegenerative disease or Parkinson's syndrome), nor taken medical treatment. According to a particularly advantageous embodiment, the kit according to the invention contains reagents specific for the 23 genes of Table 2, and in particular the probes which are indicated in Table 3 above. The present invention will be described in more detail in the nonlimiting examples which follow and which refer to the appended figures in which: FIG. 1 represents the correlation between the expression measurements obtained by microchips and those obtained by RT-PCR quantitative. The y-axis represents the difference in expression (FC = Fo / d Change) of the micro-chip data expressed in log2, and the abscissa axis that obtained in quantitative RT-PCR of the data expressed in log2. The names of the genes whose expression has been measured by these two techniques are shown on the graph.

Figure 2A shows the ranking of the training group samples based on the probability that subjects could belong to the control group (t, white circle) or sick group (m, black square) from all 233 significantly deregulated genes in patients with PRKN gene mutations.

The training group includes healthy subjects (Ti to T20) and subjects carrying mutations in the PRKN gene and developing Parkinson's disease (P13 to P16). The x-axis represents the number of subjects in the training group.

The y-axis represents the probability rate that a subject belongs to the control or diseased group.

FIGS. 2B, 2C, 2D show the ranking of the test group samples according to the probability that subjects could belong to the control group (t, white circle) or to the sick group (m, black square) from the whole group. 233 genes significantly deregulated in patients with PRKN gene mutations. Figure 2B shows the ranking of patients with neurodegenerative diseases other than Parkinson's disease (A1 to A11). Figure 2C shows the ranking of subjects with familial forms of Parkinson's disease other than familial cases carrying mutation of the PRKN gene (P1 to P12 and P17 to P29). Figure 2D shows the ranking of subjects with sporadic forms of Parkinson's disease (Si to S17). The x-axis represents the number of subjects (A, P or S) in the test group. The y-axis represents the probability rate that a subject belongs to the control or diseased group. Figure 3A shows the ranking of training group samples based on the probability that subjects could belong to the control group (t, white circle) or sick group m (black square) from all 176 genes. significantly deregulated in patients with mutations in the SCA2 gene. The training group includes healthy subjects (Ti to T20) and subjects carrying mutations in the SCA2 gene and developing Parkinson's disease (P17 to P22). The x-axis represents the number of subjects in the training group. The y-axis represents the probability rate that a subject belongs to the control group or the sick group. FIGS. 3B, 3C, 3D represent the ranking of the test sample as a function of the probability that subjects could belong to the control group (t, white circle) or to the sick group (m, black square) from the set of 176 genes significantly deregulated in patients with SCA2 gene mutations. Figure 3B represents the ranking of patients with neurodegenerative diseases other than Parkinson's disease (Al to Ail). Figure 3C shows the ranking of subjects with familial forms of Parkinson's disease other than familial cases carrying SCA2 mutation (P1 to P16 and P23 to P29). Figure 3D represents the ranking of subjects with sporadic forms of Parkinson's disease (Si to S17).

The x-axis represents the number of subjects (A, P or S) in the test group. The y-axis represents the probability rate that a subject belongs to the control or diseased group.

Figure 4A shows the ranking of the training group samples based on the probability that subjects could belong to the control group (t, white circle) or sick group (m, black square) from all 438 significantly deregulated genes in patients with the LRRK2 mutation, except for P6 and P12 subjects.

The training group includes healthy subjects (Ti to T20) and subjects carrying the LRRK2 mutation (P2 to P5 and P7 to P11). The x-axis represents the number of subjects in the training group. The y-axis represents the probability rate that a subject belongs to the control or diseased group. FIGS. 4B, 4C, 4D show the ranking of the test group samples according to the probability that subjects could belong to the control group (t, white circle) or to the sick group (m, black square) from the whole group. 438 genes significantly deregulated in patients with the LRRK2 mutation. Figure 4B shows the ranking of patients with neurodegenerative diseases other than Parkinson's disease (A1 to A11). Figure 4C represents the ranking of subjects with familial forms of Parkinson's disease other than familial cases carrying mutations of the LRRK2 gene (except for two of them P6 and P12) (Pi, P6, P12 to P29). Figure 4D represents the ranking of subjects with sporadic forms of Parkinson's disease (Si to S17). The x-axis represents the number of subjects (A, P or S) in the test group. The y-axis represents the probability rate that a subject belongs to the control group or the sick group. Figure 5A shows the ranking of the training group samples according to the probability that subjects could belong to the control group (t, white circle) or sick group (m, black square) from all 23 significantly deregulated genes in patients with different familial forms of Parkinson's disease.

The training group includes healthy subjects (T1 to T20) and subjects carrying mutations of PRKN, SCA2, LRRK2 or SNCA genes developing familial and hereditary forms of Parkinson's disease (P1 to P22, except for subjects P6 and P12).

The x-axis represents the number of subjects in the training group. The y-axis represents the probability rate that a subject belongs to the control or diseased group. Figures 5B, 5C, 5D represent the ranking of the test group samples according to the probability that subjects could belong to the control group (t, white circle) or to the sick group (m, black square) from the whole group. 23 genes significantly deregulated in patients with different familial forms of Parkinson's disease. Figure 5B shows the ranking of patients with neurodegenerative diseases other than Parkinson's disease (A1 to A11). FIG. 5C represents the classification of subjects suffering from familial forms of Parkinson's disease other than familial cases carrying mutations of the PRKN, SCA2, LRRK2 genes (except for two of them P6 and P12) or SNCA (P6, P12, P23 to P29). Figure 5D represents the ranking of subjects with sporadic forms of Parkinson's disease (Si to S17).

The x-axis represents the number of subjects (A, P or S) in the test group. The y-axis represents the probability rate that a subject belongs to the control or diseased group. EXAMPLES A - PATIENT POPULATIONS Patients were recruited at the University Hospital Center of Lille as part of the PARKFANORD study (PARKinson FAmilial NORD). Patients who were treated or symptomatic patients with Parkinson's disease were diagnosed according to the following cumulative criteria 1) 2) and 3) accepted in this respect, namely 1) at least 3 of the 4 cardinal signs of Parkinson's disease, at resting tremor, bradykinesia, muscle rigidity or postural instability; bradykinesia is necessarily part of one of the 3 cardinal signs selected for diagnosis, 2) the positive response to anti-parkinsonian levodopa treatment, and 3) the absence of the exclusion criteria defined by UK PDSBB, Hughes A.3 Journal of Neurology, Neurosurgery, and Psychiatry, 1992, 55, 181-184. The 16th criterion of family history was not retained. Three types of patient populations were recruited - subjects (P) related to at least one person with Parkinson's disease (Table 4), with family or non-familial status (relationship) determined in function of the family tree (hereditary or family forms). In this population, there are both subjects who have symptoms (treated, symptomatic and pauci- symptomatic patients) or who do not have symptoms of Parkinson's disease (asymptomatic). The so-called pauci-symptomatic subjects are familial forms of the disease and which have, in addition to the positive response to anti-parkinsonian levodopa treatment and the absence of the exclusion criteria defined by UK PDSBB, at least two of the four cardinal signs including bradykinesia. subjects (S) with Parkinson's disease according to criteria 1) 2) and 3) as defined above for subjects said to be treated or symptomatic patients, but not related to another person with Parkinson's disease of Parkinson's disease (Table 5), subjects (A) developing Parkinson's syndrome or other related neurodegenerative disease (Table 6). These subjects do not meet the diagnostic criteria for Parkinson's disease as defined above (subjects referred to as treated or symptomatic patients) and in particular as regards the response to levodopa treatment and the exclusion criteria.

In parallel with these three types of populations, forty healthy individuals with no personal or family history of Parkinson's disease or any other neurodegenerative disease were also recruited as controls (T). Nineteen RNA samples from nineteen control individuals were included in the microarray expression study, as well as a pool of equal amounts of RNA from twenty men and twenty women. between the ages of 22 and 82 (average = 52.9 years) in order to have a representative sample of these forty witnesses. The blood samples were taken from each type of population (sick or control individuals) after obtaining the informed consent of the subjects. Case Name of Clinical Status Stage Aspen Brady- Rigidity Instabi Begin Family Score gene HY -mentine kinesic -lity Asymmetic UPDRS mutated postu cally III rest line P1 SNCA paucisymptomatic 0 1 1 0 1 5 P2 LRRK2 treated patient 4 1 1 1 1 1 20 P3 LRRK2 asymptomatic 0 0 0 0 0 0 0 P4 LRRK2 treated patient 2 1 1 1 0 1 18 P5 LRRK2 symptomatic 1.5 1 1 1 0 1 15 P6 LRRK2 treated patient 4 1 1 1 1 1 34 P7 LRRK2 treated patient 2 1 1 1 0 1 34 P8 LRRK2 patient treated 2 1 1 1 1 1 20 P9 LRRK2 patient treated 3 0 1 1 1 1 16 P10 LRRK2 patient treated 3 1 1 1 1 1 56 P11 LRRK2 treated patient 2.5 1 1 1 1 1 36 P12 LRRK2 patient treated 2 1 1 1 0 1 33 P13 PRKN patient treated 3 1 1 1 1 1 30 P14 PRKN unhealthy 3 1 1 1 1 1 65 P15 PRKN treated patient 2 1 1 1 0 1 10 P16 PRKN treated patient 2 1 1 1 0 1 15 P17 SCA2 patient treated 2 1 1 1 1 1 27 P18 SCA2 paucisymptomatic 0 0 1 1 0 0 2 P19 SCA2 patient treated 2 0 1 1 0 1 - P20 SCA2 patient treated 2 0 1 1 0 1 - P21 SCA2 sickness 3 1 1 1 1 0 22 P 22 SCA2 paucisymptomatic 0 0 1 1 0 0 5 P23 unknown patient treated 3 1 1 1 1 1 22 P24 unknown patient treated 2 1 1 1 0 1 30 P25 unknown patient treated 3 1 1 0 1 1 14 P26 EIF4G1 treated patient 2 1 1 1 1 1 32 P27 EIF4G1 treated patient 1.5 1 1 1 0 1 22 P28 EIF4G1 paucisymptomatic 1 0 1 1 0 1 9 P29 EIF4G1 paucisymptomatic 1 0 1 1 0 1 3 Table 4: Description of symptoms and stage of severity of cases When the mutated gene is identified, its name is indicated. Patients P17 to P22 carry a SCA2-P mutation. The absence or presence of treatment or symptoms is indicated by 0 and 1 respectively. The score of Part III of the UPDRS scale is indicated. The higher the score, the more the motor examination is unfavorable and the more the patient presents disorders in the speech, the gestures of the everyday life and a disturbed gait, annoying for his everyday life. case Status Stage Tremor Brady- Rigidity Instability Early Sporadic UPDRS score clinic HY resting postural kinesia Asymmetric III S1 patient treated 3 1 1 1 1 1 47 S2 patient treated 4 0 1 1 1 1 32 S3 patient treated 2 1 1 1 1 1 21 S4 patient treated 3 1 1 1 1 1 26 55 patient treated 3 0 1 1 1 0 34 S6 patient treated 2 1 1 1 0 1 20 S7 patient treated 2 0 1 1 1 0 18 58 patient treated 2 1 1 1 0 1 11 59 patient treated 3 0 1 1 1 1 21 S10 patient treated 2 1 1 1 0 1 13 511 patient treated 2 0 1 1 1 1 19 S12 patient treated 2 1 1 1 0 1 12 S13 patient treated 3 1 1 1 1 1 26 514 patient treated 2 1 1 1 1 1 31 S15 patient treated 2 1 1 1 0 1 16 S16 patient treated 2 1 1 1 1 1 16 S17 patient treated 2 1 1 1 0 1 24 Table 5: Description of symptoms and stage of severity of sporadic cases of Parkinson's disease The absence or presence of symptoms is indicated by 0 and 1, respectively. The score of Part III of the UPDRS scale is indicated. The higher the score, the more the motor examination is unfavorable and the more the patient presents disorders in the speech, the gestures of the everyday life and a disturbed gait, annoying for his everyday life. case Name of Other disease mutated neurodegenerative gene SCA2 SCA A2 SCA2 SCA A3 SCA2 SCA A4 SCA2 SCA A5 SCA2 SCA A6 SCA2 SCA A7 unknown MSA A8 unknown DLB A9 unknown DLB A10 unknown DLB Unknown diluted DLB Table 6: Description of patients with another Neurodegenerative disease or parkinsonian syndrome meeting the criteria in force Patients Al to A6 carry an SCA2-C mutation.

B - BIOLOGICAL SAMPLES AND MOLECULAR MARKERS 1. Recovery of blood mononuclear cells The blood samples, taken in CPT (BD Vacutainer) tubes containing a FICOLL gradient, are centrifuged for 20 minutes at 1750 g at 23 ° C. Mononucleated cells are collected. After washing, aliquots containing 0.5 × 10 7 cells are made by adding to each tube 700 μl of lysis buffer (RNeasy mini kit QIAGEN ') containing 1% of [3-mercaptoethanol. 2. Total RNA Extraction RNA extraction is performed using the RNeasy mini kit and the RNasefree DNase kit from QIAGEN ° according to the manufacturer's recommendations. Briefly, the cells are mechanically lysed. A volume of 70% ethanol is then incorporated into each sample and the mixture is homogenized with a pipette. The mixture thus obtained is deposited on a column containing a silica membrane which will retain the nucleic acids. After washing, DNase I treatment is applied to remove residual DNA. Three other washes eliminate DNase, DNA and traces of ethanol. The RNA is recovered in a volume of 30 μl of RNase-free water by centrifuging for one minute at 10,000 g. The purified total RNA sample is stored at -80 ° C. 3. Quantification and verification of the quality of the RNA The RNA assay is carried out using the Nano-Drop ° ND-1000 spectrophotometer by measuring the absorbance at 260 nm from a sample lpl. 260nm / 280nm and 260nm / 230nm absorbance ratios are calculated to evaluate RNA purity. RNA quality analysis is performed using Agilentc Bioanalyzer RNA 6000 nano . The samples to be analyzed (ipl) and a molecular weight marker are deposited in each well of the RNA 6000 nano chip composed of a gel mixed with a fluorophore (intercalating single-stranded nucleic acids). Wells are analyzed by Agilent Bioanalyzer 2100 °. The data used to verify the quality of the RNAs are the ratio of 285 / 18S ribosomal RNA fractions and the RIN for RNA Integrity Number which is an algorithm developed by Agilent to score RNA degradation on a scale of 1 at 10. C - STUDY OF GENE EXPRESSION 1. Complementary DNA microarrays The technology of cDNA microarrays makes it possible to simultaneously analyze all the transcripts of the genome. The expression chips used are developed by the firm Agilent ° and covers the entire human genome thanks to nearly 44,000 probes (44K whole human genome expression arrays (Agilent, 4 X 44K slide formats) .These probes are based on in situ synthesis of 60-mer oligonucleotides on a glass slide that will be used to capture labeled complementary DNA (cDNA) targets., from the total RNA (300 ng), the cDNA targets are prepared by retro-transcription To each sample, were added spike-in which are RNAs as positive controls during the amplification and labeling steps.Then the T7 RNA polymerase amplifies the sample and incorporates PTCs labeled with the Cyanine 3. The labeled cDNAs are then fragmented into oligomers of 60 to 100 nucleotides allowing optimal hybridization with the probes The hybridization of the samples on the chip is carried out in a sealed hermetic chamber and placed in an oven for 17 hours at 65 ° C. The slides are then washed, dried and scanned.

The scanner has a laser that excites cyanines 3 to measure the fluorescence intensity thus emitted. Agilent Feature Extraction software eliminates background noise and verifies spot quality on slides. Each probe on the slide corresponds to a fluorescence intensity. Data normalization is necessary because experimental biases exist. Indeed, there may be differences in the rate of incorporation of cyanines into the cDNA, a poor quality of fluorescence detection by the scanner; other biases may be caused by deposition, hybridization. 2. Quality controls Each of the RNA samples used for the chips showed a satisfactory quality of the absorbance ratios 260 nm / 280 nm (between 1.8 and 2), and 28S / 18S (> 1.5) and RIN (> 8.9). The analysis of the chip quality ratios by Feature Extraction showed fluorescence intensities of the spots which respected a normal law. 3. Data processing from chips Data analysis is performed using GeneSpring GX 7.3 (Agilent), a collection of statistical analysis functions. Firstly, the analysis of the results of the chips comprises, conventionally, a step of normalization of the data per chip and per gene on the median fluorescence intensities of the control subjects. Normalized values are expressed in log ratio.

Expression data are then classified according to the subject's HY stage of Parkinson's disease. Series A: subjects with Parkinson's disease whose HY stage is 0 to 1.5 inclusive. Series B: subjects with Parkinson's disease whose HY stage is from 2 to 2.5 inclusive. Series C: subjects with Parkinson's disease whose HY stage is from 3 to 5 inclusive. Each series (A, B, C) is compared independently to all of the expression data of the control subjects. This comparison is carried out by a Volcano plot filtering by applying the statistical analysis by the Welch T test which is defined by the formula: r F1 .... 2 t = 2 where Xi, 9, i and Ni are respectively the nth of the sample mean, sample variance and sample size, with a significance threshold p = 0.05 and an expression variation threshold (Fo / d Change) of 1.2 . Thus the deregulated genes whose expression variation threshold is less than 1.2 will not be retained for the rest of the analysis. At the end of this step, we obtain for each series a list of genes specifically deregulated for a given HY stage (list_A, list_B, list_C). All of the expression data of subjects with neurodegenerative diseases or Parkinson's disease other than Parkinson's disease is compared with all of the expression data of the controls by applying the same criteria as above for the statistical analysis by the Welch T-test. A list of specifically deregulated genes is obtained in subjects with neurodegenerative diseases or Parkinson's syndrome other than Parkinson's disease (list).

Genes whose expression is significantly deregulated in common at the different stages of Parkinson's disease HY (list_E) are selected through a graphical representation of Venn diagrams by crossing lists: list_A, list_B and list_C. The last step is to exclude from this list (list_E), genes that are common with those that are also significantly deregulated in patients with other neurodegenerative diseases or parkinsonian syndromes (list_D). This gives a set of genes deregulated specifically in patients with Parkinson's disease (liste_F). The word gene is used to refer to the Agilent probes of the chip. However, several probes may be present for the same gene and probes may still correspond to potential genes or even sequences whose role is not yet known. 4. Possible validation of the results obtained by a predictive analysis with PAM software The expression data corresponding to the set of genes obtained as defined in paragraph C 3. are introduced in the software PAM (Prediction Analysis for Microarray) (version 2.1 http: /lwww-stat.stanford.edul tibsIPAM /) which applies the closest close-centered centroid analysis. The optimal number of genes is determined from the cross validation method (10 times) from the training group.

The training group is made up of subjects for whom class membership is known (the class being defined by patient group or control group). Gene expression data from these subjects are provided for analysis. The training group makes it possible to determine all the genes whose expression is most correlated with the diagnosis (patient / control). Then this set of genes is used to predict the diagnosis of subjects in the test group. The test group is a group consisting of subjects for which class membership will be predicted according to the rules learned from the training group. The true belonging of these samples to the class may or may not be known. Thus concerning the training group, a subject P will be properly classified as a patient when his probability will be close to 1 and will be represented by the symbol corresponding to the sick group. A subject T will be correctly classified as a control subject when its probability is close to 1 and will be represented by the symbol corresponding to the control group. For the test group, a subject with a risk of developing Parkinson's disease will be correctly classified as a patient when its probability is close to 1 and will be represented by the symbol corresponding to the sick group. The error rate makes it possible to evaluate the effectiveness of the set of genes used in the method of detecting molecular markers to determine the presence or risk of developing Parkinson's disease. The error rate is the number of misclassified subjects out of the total number of subjects in that class; two classes are possible: parkinsonian patient or control (healthy or suffering from a parkinsonian syndrome or another neurodegenerative pathology that Parkinson's disease).

EXAMPLE 1 Concordance of the Results Obtained by the Use of Microchips with Those Obtained by Other Expression Measurement Techniques Such as Quantitative RT-PCR (Technical Verification Example) A Quantitative RT-PCR Analysis of the expression of 12 genes, some of which are selected from those which are known to be deregulated when the LRRK2 gene is mutated and having a differential expression of at least a factor 2 (Fo / d Change> 2 or <0.5) significant in Welch T test (p <0.05), was performed. The level of expression by quantitative PCR of certain genes classically deregulated in the genetic forms of Parkinson's disease was also studied.

This analysis was performed in parallel with the microarray analysis in order to verify the concordance of the results (that is to say the level of expression of the genes) obtained by these two techniques. at. Retro-transcription step (RT) The retro-transcription of the RNA into cDNA is carried out using the QuantTect Reverse Transcription Kit (Qiagen) from 250 ng of RNA according to the manufacturer's protocol. Residual residual DNA is first removed by DNase treatment prior to performing the retro-transcription reaction. The primers used are a mixture of hexamers and oligo-dT allowing the synthesis of cDNA from all the RNA. b. Real-time PCR step Quantitative PCR was performed according to the manufacturer's recommendations using the SYBR Green PCR Master Kit (Applied Biosystems). This step makes it possible to follow in real time the amplification of a target cDNA fragment using the SYBR Green fluorophore. This DNA intercalator becomes fluorescent when bound to double-stranded DNA. The amount of fluorescence is measured at the end of the elongation step of each PCR cycle. This measurement makes it possible to detect which cycle (Ct for thresho / d cycle) the fluorescence exceeds the threshold determined by the user, on the amplification curve. This technique makes it possible to calculate the relative normalized expression of the target gene. Standardization is relative to a reference gene (18S ribosomal RNA, sense primer CCTGGATACCGCAGCTAGGA (SEQ ID NO: 24), anti-sense primer GCGGCGCAATACGAATGCCCC (SEQ ID NO: 25), hybridization temperature 60 ° C). The calculation method is derived from the 2-mct method to account for the fact that the efficiency is not 100%. The normalized expression is thus the ratio: (1 / efficiency of the target gene and of the gene b '/ (1 / efficiency of the referent gene and of the referent gene) The mixture comprises 10 μl of water, 0.75 μl of each of the primers ( 10 pmol / μl) and 12.50 master (2X) for one well One μl of cDNA is added to the reaction mixture Sample measurements are performed in duplicate Real-time PCR is performed on an ABI device Prism 7700 (Applied Biosystems) The statistical analysis of the expression ratios between the sick individuals and the controls is carried out for each gene by a Mann and Whitney test (threshold of significance p <0.05). used to perform the PCR are shown in Table 7 below: Genes (number Identifiers Accession Temperature Size Description Primer Sequences (5 '-> 3') GenBank hybridization product sequence) PCR (° C) ARG1 Arginasel Meaning ACCCATC III CACACCAGC SEQ ID NO: 2 6 107 bp 58 (NM_000045) antisense TCCTGAGAGTAGCCCTG III SEQ ID NO: 27 ERAF Associated factor in sense AATCAGCAGGTCTTCAATG SEQ ID NO: 28 97 53 (NM_016633) Erythroid antisense GCTGCCTGTAATAGTTGATG SEQ ID NO: 29 SNCA Synuclein sense CCAAAACCAAGGAGGGAGTG SEQ ID NO: 97 bp 60 (NM_000345) antisense ACACCCGTCACCACTGCTCC SEQ ID NO: 31 LRRK2 Kinase 2 rich in sense pattern AGGTACACAAAAGCAGAAAG SEQ ID NO: 32 96 57 (NM_198578) repeat leucine anti-sense ACTTCAATGTG IIIII CTAA SEQ ID NO: 33 pb PINK1 Sense-Induced Kinase 1 AGAGGCCAGCAAGAGACC SEQ ID NO: 34 98 bp 58 (NM_032409) PTEN antisense CTAACTTCAGATTCTTCAGGGC SEQ ID NO: 35 DJ1 PARK 7 sense CCATCTGTGCAGGTCCTA SEQ ID NO: 36 99 bp 55 Parkinson disease (NM_007262) (autosomal recessive , antisense CTCCATTCATCATITI "GTCT SEQ ID NO: 37 early-onset) 7 IFNG Gamma interferon sense GGAAAGAGGAGAGTGACAGA SEQ ID NO: 38 124 bp 56 (NM_000619) antisense ATGTCTTCCTTGATGGTCTC SEQ ID NO: 39 DUSP10 Phosphatase 10 sense GGL IIII GAGTTCATTGAG G SEQ ID NO: 40 123 bp 64 (NM_007207) double antisense specificity GTCATCCGAGTGTGCTTCAT SEQ ID NO: 41 LTF Lactotransferrin sense GCTACTCTGGTGCCTTCAA SEQ ID NO: 42 82 bp 58 (NM_002343) antisense AGGTCCTCAAACACTGTGC SEQ ID NO: 43 SELENBP1 Protein 1 binding sequence sense AAGAACGAGGGAGGTACATG SEQ ID NO: 44 112 bp 55 (NM_003944) with antisense selenium AGAGCAGGATGTCGGTGAT SEQ ID NO: 45 ALAS2 Aminolevulinate delta sense CTGCCAGGGTGCGAGATTTA SEQ ID NO: 46 111 bp 55 (NM_000032) synthetase 2 antisense GTGGTCAGGGTCATTGTGCC SEQ ID UGT2B17 UDP glucuronosyl SEQ ID NO: 48 92 bp 58 (NM_001077) tra family, sense CACAAAAGG II CTATGGAG antisense AAGGTCATTCTGGGGTA SEQ ID NO: 49 i-sense AAGG2 t polyp B17 Table 7: primer sequences, amplimer sizes and hybridization temperatures for each gene studied in quantitative PCR. The sequences of the primers were designed using Oligo 5 software.

The results are grouped in Figure 1 and show a strong correlation (r2 = 0.94) between these two techniques. This means that increases or decreases in gene expression observed on the chips can be found using another quantization technique with the same level of expression difference. Therefore, these expression variations are not related to artifacts of the microarray technique but are indeed a reflection of variations in gene expression in the mononuclear cells of patients. EXAMPLE 2 Determination of a list of genes whose expression is derailed in patients with a particular familial form of Parkinson's disease and study of the ability to identify subjects suffering from sporadic forms from this list In order to determine whether from a single type of familial form it is possible to predict the risk of developing Parkinson's disease, the method described in paragraphs A to c above was applied separately for each familial form such as identified in Table 4, namely for mutations of PRKN, LRRK2, SCA2 and SNCA genes. The expression of genes in subjects carrying these different mutations is studied from the RNA extracted from the blood mononuclear cells of these patients and analyzed by Agilent microchips as described in paragraphs B and C above. The prediction value (error rate) of the set of genes thus calculated to determine the presence or the risk of developing Parkinson's disease, and in particular a sporadic form of this disease, is analyzed using PAM software as described herein. above in paragraph C 4. a. Study of deregulated genes only in patients with Parkin PRKN gene mutation The expression analysis of the deregulated genes of subjects carrying PRKN gene mutations compared to healthy controls was performed by Volcano Plot filtering by applying a Welch T test with a significance threshold of p = 0.05 and an expression variation threshold of 1.2 after classification of subjects according to their Hoehn and Yahr stage. This analysis is performed using the Genespring GX 7.3 software. 2,775 genes make up the list_B of genes specifically deregulated compared to the controls of PRKN mutation-positive subjects who have developed Parkinson's disease with an HY stage of 2 or 2.5 (subjects P15, P16). 2,211 genes make up the list_C list of specifically deregulated genes compared to the controls of subjects with a PRKN mutation and who developed Parkinson's disease whose HY stage is 3, 4 or 5 (subjects P13, P14).

The genes specifically deregulated and common to the different HY stages are selected from the list lste_E (270 genes). This list of deregulated genes (List_E) was then crossed with the list of deregulated genes in subjects suffering from other neurodegenerative diseases or Parkinson's syndrome (Listel) which contains 5,645 genes). After excluding genes that are common with those that are also significantly deregulated in patients with other neurodegenerative diseases or parkinsonian syndromes, we obtain a set of 233 genes deregulated specifically in subjects carrying the PRKN mutation (liste_F). This set of 233 genes was analyzed by the PAM software - the training group consists of the 4 subjects carrying mutation of the PRKN gene (subjects P13, P14, P15, P16) and controls (subjects T1 to T20), - the test group consists of the 25 other family subjects (subjects P1 to P12 and P17 to P29), sporadic 17 (subjects Si to S17) and 11 subjects suffering from another parkinsonian syndrome (subjects Al to Ail). The results obtained from the training group are shown in the table below and in Figure 2A. Training Group Results (Threshold = 0.01) True \ Predicts Patients Illnesses Ranking Error Rate in% Subjects P Patients 4 0 0 Subjects T Witnesses 0 20 0 With the 233 gene series, all patient carriers mutations in the PRKN gene are represented by a black square in the group of patients with a probability of 1. Therefore, they are well identified as sick patients. Similarly, the samples from healthy individuals are represented by a white circle with a probability greater than 0.5 (here equal to 1) and are therefore well identified in the group of controls (Figure 2A).

On the other hand, in the test group, figure 2D and the table below show that all the patients who develop a sporadic form (subjects S1 to S17) are represented by the control symbol (white circle) and with a probability higher than 0, 5 (here equal to 1) in the sick group (error rate of 100%).

To be correctly classified, they would have had to be represented by the sick symbol (black square) and classified in the sick group with a probability greater than 0.5. Results of the test group (threshold = 0.01) True \ Predicted Ill Witnesses Error rate Figures in% Subjects A Witnesses 0 11 0 2B Subjects P Patients 3 22 88.00 2C Subjects S Patients 0 17 100.00 2D Only P4, P5 and P26 patient profiles are recognized as belonging to Parkinson's patients (table above). They are represented by a black square with a probability greater than 0.5 in the group of patients (Figure 2C). The misclassification rate for this group is 88%. Further analyzes were performed by decreasing the number of genes included in the list _F in PAM, but the different sets of genes obtained during these analyzes also make it impossible to distinguish the sporadic cases from the control subjects. These results therefore show that all 233 genes identified solely from the information of patients carrying mutations of the PRKN gene do not correctly classify sporadic subjects or other familial cases. b. Study of Deregulated Genes Only in Carriers with an SCA2 Gene Mutation That Result in Parkinson's Disease Expression analysis of deregulated genes in subjects with SCA2 gene mutation compared with healthy controls was performed by filtering in Volcano Plot by applying a Welch T test with a significance threshold of p = 0.05 and an expression variation threshold of 1.2 after classification of subjects according to their Hoehn and Yahr stage. 4,757 genes make up the list_A list of specifically deregulated genes compared to controls in subjects with an SCA2 mutation and who developed Parkinson's disease with an HY stage of 0, 1 or 1.5 (subjects P18, P22). 4,288 genes make up the list_B of genes specifically deregulated compared to the controls of subjects with an SCA2 mutation and who developed Parkinson's disease whose HY stage is 2 or 2.5 (subjects P17, P19, P20). 6,446 genes make up the list_C list of genes specifically deregulated compared to controls in subjects with an SCA2 mutation and who developed Parkinson's disease whose HY stage is 3, 4 or 5 (subjects P21). The genes specifically deregulated and common to the different HY stages are selected from the list_E list (217 genes). This list of deregulated genes (List_E) was then crossed with the list of deregulated genes in subjects suffering from other neurodegenerative diseases or Parkinson's syndrome (Listel) which contains 5,645 genes). After excluding genes that are common with those that are also significantly deregulated in patients with other neurodegenerative diseases or parkinsonian syndromes, we obtain a set of 176 genes deregulated specifically in subjects carrying the SCA2 mutation (Liste_F). This set of genes was analyzed using PAM software. The training group consists of 6 subjects carrying mutations of the SCA2 gene (subjects P17 to P22) and controls (subjects Ti to T20). The test group consists of 23 other family subjects (subjects P1 to P16 and P23 to P29), 17 sporadic subjects (subjects S17 to S17) and subjects with another Parkinson syndrome (subjects Al to A11): Results obtained from the training group are shown in the table below and in Figure 3A: Training Group Results (Threshold-0.01) True \ Predicts Patients Illnesses Ranking Error Rate in% Subjects P Patients 3 3 50 Subjects T Witnesses 0 20 0 With this set of genes, 3 subjects with SCA2 mutation out of 6 are identified as sick patients and all controls are identified as controls.

On the other hand, in the test group, the table below and Figure 3C show that only 2 patients are recognized in the class of patients (P16 and P26). Three subjects (Al, A5, A10) with another neurodegenerative disease are poorly classified (Figure 3B) and none of the sporadic subjects have been identified as ill (Figure 3D). To be correctly classified, they would have had to be represented by the sick symbol (black square) and classified in the sick group with a probability greater than 0.5. Results of the test group (threshold = 0.01) True \ Predicted Patients Witnesses Error rate Figures in% Subjects A Witnesses 3 8 27.27 3B Subjects P Patients 2 21 91.30 3C Subjects S Patients 0 17 100 3D Additional analyzes were performed by decreasing the number of genes included in the list _F in PAM, but the different sets of genes obtained during these analyzes also make it impossible to distinguish the sporadic cases from the controls. These results thus show that the series of 176 genes identified solely from the information of patients carrying mutations of the SCA2 gene does not correctly classify sporadic subjects or other familial cases. c. Analysis of Deregulated Genes Only in Cases Carrying a mutation in the LRRK2 Gene The expression analysis of the deregulated genes of the subjects carrying the mutation of the LRRK2 gene compared to the healthy controls was carried out by a Volcano Plot filtering by applying a Welch T test with a threshold of significance of p = 0.05 and a threshold of expression variation of 1.2 after classification of subjects according to their stage Hoehn and Yahr. 9 005 genes make up the list A list of specifically deregulated genes compared to the controls of the subjects carrying a mutation LRRK2 and having developed a Parkinson's disease whose HY stage is 0, 1 or 1.5 (subjects P3, P5). 5 783 genes make up the list_B list of specifically deregulated genes compared to the controls of subjects with an LRRK2 mutation and who developed Parkinson's disease whose HY stage is 2 or 2.5 (subjects P4, P7, P8, Pli ). Subject P12 was not considered in this analysis for the selection of deregulated genes because P12 had both Parkinson's disease and manic-depressive syndrome, which could bias specificity towards Parkinson's disease selected genes. 6,665 genes make up the list_C of genes specifically deregulated compared to the controls of subjects carrying a mutation LRRK2 and having developed a Parkinson's disease whose HY stage is 3, 4 or 5 (subjects P2, P9, P10). The P6 sample showed significant contamination with hemoglobin, which could bias gene selection; it was therefore not taken into account in this analysis. The genes specifically deregulated and common to the different HY stages are selected from the list_E list (759 genes).

This list of deregulated genes (Liste_E) was then crossed with the list of deregulated genes in subjects with other neurodegenerative diseases or Parkinson's syndrome (List_D). After excluding genes that are common with those that are also significantly deregulated in patients with other neurodegenerative diseases or parkinsonian syndromes, we obtain a set of 438 genes deregulated specifically in subjects carrying the mutation LRRK2 (Liste_F). This set of 438 genes is analyzed with PAM software. The training group consists of 9 subjects carrying mutations of the LRRK2 gene. (subjects P2 to P5 and P7 to P11) and controls (subjects Ti to T20).

The test group consisted of the 18 other family subjects (Pl and P13 to P29 subjects) as well as the 2 other subjects carrying an LRRK2 mutation (subjects P6 and P12), 17 sporadic subjects (subjects S1 to S17) and 11 subjects with another Parkinson syndrome (subjects Al to A11). The results obtained from the training group are shown in the following table and in Figure 4A: Training Group Results (Threshold = 0.01) True \ Predictor Bearer Healthy Subject Error Rate Mutation (Controls) ranking in% Subjects P Patients 9 0 0 Subjects T Witnesses _ 8 12 40 With this set of genes, all subjects with LRRK2 mutation are well identified as sick patients and 12 out of 20 controls are identified as controls in the group (Figure 4A), a classification error rate of 40%. On the other hand, in the test group, the table below shows that only 17 patients are recognized in the class of patients Pi, P6, P12, P14, P15, P26, P27, P28, P29 (FIG. 4C) and S3, S5. Si, S8, S9, S10, S14, and S17 (Figure 4D). All subjects with another neurodegenerative disease are well classified (Figure 4B). Test Group Results (Threshold = 0.01) True \ Predicted Patients Illnesses Error Rate Sequence Figures in Subjects A Witnesses 0 11 0 4B Subjects P Patients 9 11 55.00 4C Subjects S Patients 8 9 52.94 4D Des Further analyzes were carried out by decreasing the number of genes included in the F list in PAM, but the different sets of genes obtained during these analyzes also make it impossible to distinguish the sporadic cases from the control subjects. These results thus show that the set of 438 genes obtained solely from the results of dysregulated expression of genes in patients carrying a mutation LRRK2 does not make it possible to classify more than 8 sporadic cases out of 17, a rate of classification error of 52.94%.

Similar results were obtained when subjects P6 and P12 were excluded from the test group during PAM analysis (in addition to being excluded in the determination of lists B_list and C_list). In conclusion, the three analyzes described above in points a., B. and c. of Example 2 show that it is not possible to predict subjects who have Parkinson's disease or who may develop a sporadic form of Parkinson's disease using information from only one type of deleterious mutation . EXAMPLE 3: Determination of a Set of Aenes the Expression of which is Deregulated in Patients with Different Familial Forms of Parkinson's Disease and Study of the Ability to Identify Subjects Affected, Including Sporadic Forms, from That Disease together In this example, gene expression data obtained from a set of 4 types of familial forms of Parkinson's disease were taken into account to determine if the set of genes thus obtained can predict the disease. Parkinson's disease, and in particular sporadic forms of the disease. Thus, differential gene expression data obtained from all patients with mutations of the SNCA, LRRK2, SCA2 and PRKN genes responsible for these 4 familial forms of Parkinson's disease have been taken into account in the study. analysis by the GeneSpring software, except for the subjects P6 and P12 (see Example 2 above). The other Parkinsonian patients (patients P23, P24, P25, P26, P27, P28, P29) and sporadic were not included in this first analysis with GeneSpring, but were included in the analysis with PAM. They will be used to verify the predictive value of the genes selected in this analysis. The expression analysis of the genes deregulated of the subjects carrying SCA2, PRKN, SNCA and LRRK2 gene mutations compared to the healthy controls was carried out by a Volcano Plot filtering by applying a Welch T test with a threshold of significance of p = 0.05 and an expression variation threshold of 1.2 after classification of subjects according to their stage Hoehn and Yahr. 4,415 genes make up the list_A list of specifically deregulated genes compared to the controls of subjects carrying a SCA2, PRKN, SNCA and LRRK2 mutation and having developed a Parkinson's disease whose HY stage is 0, 1 or 1,5 ( subjects P1, P3, P5, P18, P22). 2 297 genes make up the list_B of genes that are specifically deregulated compared to controls with SCA2, PRKN, SNCA and LRRK2 mutation and have developed Parkinson's disease with an HY stage of 2 or 2.5 (P4 subjects , P7, P8, P11, P15, P16, P17, P19, P20). As for example 2 c., Subject P12 was not considered for this analysis. 3,292 genes make up the list_C list of genes specifically deregulated compared to controls with SCA2, PRKN, SNCA and LRRK2 mutation and who developed Parkinson's disease with an HY stage of 3, 4 or 5 (P2 subjects , P9, P10, P13, P14, P21). As for example 2 c., Subject P6 was not considered for this analysis. The genes that are specifically deregulated and common to the different HY 10 stages are selected from the list_E list (38 genes). This list of deregulated genes (Liste_E) was then crossed with the list of genes deregulated in subjects with other neurodegenerative diseases or Parkinson's syndrome (List_D composed of 5,645 genes). After excluding genes that are common with those that are also significantly deregulated in patients with other neurodegenerative diseases or parkinsonian syndromes, a set of 23 genes is obtained (List_F). This set of 23 genes grouped in Table 8 below corresponds to specifically deregulated genes in subjects carrying the SCA2, PRKN, SNCA and LRRK2 mutation.

No. Chip Code Gene Name Code Genbank 1 A_232216712 Transient Potential Receptor of NMV017662 Cation Channel, Subfamily M, Member 6 (TRPM6) 2 A_24_P245838 Mannosyl (beta-1,4-) - glycoprotein beta-1,4-AK125361 N-acetylglucosaminyltransferase (MGAT3) 3 A_232369316 Vang-like 1 (VANGL1) NM_138959 4 A_2321904 Transmembrane domain 4, sub-family A, NM 000139 member 2 (MS4A2) A_23_P349321 Xylosyltransferase type I (XYLT1) NM_022166 A_23_P117662 Histidine decarboxylase (HDC) NM_002112 7 A_232133916 Component C2 of the complement (C2) NM_000063 8 A_322162187 Component C2 of the complement (C2) NM_000063 9 A_2328961 Interleukin 7 (IL7 NM_000880 10 A 32 P175313 EST PM1-NN1207-071200-003-a12 NM1207 BF960555 11 A_3226433 EST yr57d04, rl Complementary DNA of clone H64096 IMAGE: 209383 soares fetal liver spleen, similar to the protein of the human heat shock family of 71KD 12 A_322162183 Component C2 of complement (C2) NM_000063 13 A_322104572 EST UI-E-DW1-ahe-d-15- 0-UI.s 1 clone of BM670971 the complementary DNA UI-E-DW1-ahe-d-15-0-UI 3 '14 A_32_13234485 IS UI-1BB1p-auq-g-04-0-UI.sl BQ024427 NCI_CGAP_PI6 Complementary DNA of the clone IU 1-BB1p-auq-g-04-0-UI 3'15 A_23260047 Protein 1 containing two-spiral NM_006283 acid structures (TACC1) 16 A_32245875 EST ws88f04.xl Homo sapiens, AW009873 Complementary DNA of clone IMAGE: 2505055 3 ', similar to the related sequence of phospholipase C TR: Q62084 Q62084 17 A_23_P352435 Signaling G protein regulator NM_002926 12 (RGS12) 18 A_232387649 DNAJA4 homologous to DnaJ (Hsp40), sub-AF116663 family A, member 4 (DNAJA4) or PRO1472 19 A_23_P200252 Open reading frame 57, chromosome 1 NM_032324 (C1orf57) A_23_P66174 S1-13 binding domain of protein kinase NM_00102440 1 (SBK1) 21 A_23_P33927 FKSG43 AF334945 A_23_P333640 Papiline (PAPLN) sulphated glycoprotein NM_173462 similar to proteoglycan 23 A_242323682 region chromosome of lateral sclerosis NM_0209 Amyotrophic 2 (juvenile) For the PAM analysis, the samples were divided as follows: the training group corresponds to the healthy controls (subjects Ti to T20) and the 20 subjects carrying a mutation LRRK2, PRKN, SNCA and SCA2 (subjects P1, P2, P3, P4, P5, P7, P8, P9, P10, P11, P13, P14, P15, P16, P17, P18, P19, P20, P21 and P22). - the test group includes 9 other family cases (ie family cases other than those carrying a PRKN, LRRK2, SCA2 or SNCA mutation (subjects P), namely subjects, P23, P24, P25 , P26, P27, P28, P29 and 2 LRRK2 mutation-positive individuals (P6 and P12), the 17 sporadic cases (S1 to S17 subjects), and 11 subjects with other Parkinson's syndrome (Ai subjects). A11) The results of the training group analysis are summarized in the following table and in Figure 5A: Training group results (threshold = 0.01) True ^ Predicts Patients Indicators Error rate of ranking in% Subjects P Patients 20 0 0 Subjects T Witnesses 2 18 10 This set of 23 genes classifies patients and controls with an error rate of 0% and 10% respectively on the training group. of the test group are shown in the following table and in Figures 5B, 5C and 5D Results of the test group (threshold = 0.01) True \ P Patients Rate of error Figures in% Subjects A Witnesses 4 7 36,36 5B Subjects P Patients 6 3 33,33 5C Subjects S Patients 15 2 11,76 5D These results show that 5 out of 26 Parkinson's patients in the group test are misclassified. These are 2 sporadic cases (Si and S13, Figure 5D) and 3 other familial cases (P24, P25, P12 Figure 5C) that were classified as controls represented by a white circle instead of patients represented by a square black in the group of patients. Pauci-symptomatic family members (P28, P29) were well classified as patients at risk of developing Parkinson's disease (as well as the subject carrying a LRRK2 (P6) mutation with hemoglobin contamination). Subjects with another neurodegenerative pathology A5, A6, A8 and All are misclassified as represented by a black square instead of a white circle in the control group (Figure 5B). Similar results were obtained when subjects P6 and P12 were excluded from the test group during PAM analysis (in addition to being excluded in the determination of lists B_list and C_list). In conclusion, these results show that the set of genes deregulated in familial forms of Parkinson's disease determined according to the method of the invention not only makes it possible to determine whether subjects are suffering from Parkinson's disease with a level of low error, even for paucisymptomatic cases, but also to identify sporadic cases of the disease with a very low error rate (in this example of the order of 11.76%). Moreover, these results demonstrate that it is essential to take into consideration the deregulated genes common to several family forms incriminating different genes to obtain a synergistic effect of the information obtained and allow the identification of these sporadic cases.

Claims (13)

REVENDICATIONS1. A method for detecting in a biological sample to be tested the differential expression of a set of molecular markers associated with the presence or risk of developing Parkinson's disease, characterized in that it comprises the following steps: a) in vitro extraction molecular markers from the biological sample to be tested a2) in vitro extraction of molecular markers from biological samples of subjects suffering from various familial forms of Parkinson's disease, subjects suffering from other neurodegenerative diseases or parkinsonian syndromes and healthy subjects, b) determine the set of molecular markers whose expression is significantly deregulated in common in subjects with different familial forms regardless of the stage of the disease but whose expression is not significantly deregulated in subjects with other neurodegenerative diseases or Parkinsonian syndromes, compared to that measured in healthy subjects, c) verify that the expression of all these molecular markers is well deregulated in the sample to be tested. The method of claim 1 for detecting the presence or risk of developing a sporadic form of Parkinson's disease. The method of claim 1 or claim 2 for detecting the presence of an asymptomatic form of Parkinson's disease. 4. Method according to any one of the preceding claims, characterized in that the subjects with different familial forms are patients carrying mutations in SNCA, PRKN, SCA2 or LRRK2 genes. 5. Method according to any one of the preceding claims, characterized in that the subjects suffering from other neurodegenerative diseases or parkinsonian syndromes are patients with SCA, MSA or DLB. 6. Method according to any one of the preceding claims, characterized in that the biological sample to be analyzed consists of RNA extracted from blood mononuclear cells. 7. Method according to any one of the preceding claims, characterized in that the deregulated expression of the molecular markers is determined by a quantitative measurement of the expression of genes or proteins selected from microchips and RT-PCR. 8. Method according to any one of the preceding claims, characterized in that the molecular markers are RNA or proteins. 9. Method according to any one of the preceding claims, characterized in that the set of molecular markers which is determined according to step b) is the list of the following genes: Gene name Code Genbank 1 transient potential receptor of the channel to cation, under NM_017662 M family, member 6 (TRPM6) 2 Mannosyl (beta-1,4-) - glycoprotein beta-1,4-N- AK125361 acetylglucosaminyltransferase (MGAT3) 3 Vang-like 1 (VANGLI) NM_138959 4 Transmembrane domain 4 subfamily A, member NM 000139 2 (MS4A2) Xylosyltransferase type I (XYLT1) NM 022166 6 Histidine decarboxylase (HDC) NM_002112 7 Component C2 of complement (C2) NM_000063 8 Interleukin 7 (IL7) NM 000880 9 EST PM1-NN1207 -071200-003-a12 NM1207 BF960555 EST yr57d04.rl Complementary DNA of clone 10 IMAGE: 209383 soares fetal liver spleen, similar to the H64096 protein of the human heat shock family of 71KD 11 EST UI-E-DWl-ahe-d- 15-0-UI.s1 clone of BM670971 complementary DNA UI-E-DWl-ahe-d-1 Compound DNA of the clone UI-1-BB1p-aq-g-04-O-3'-13I protein 1 containing bi-spiral acid structures NM_006283 (TACCI) 14 EST ws88f04.xl Homo sapiens, complementary DNA of AW009873 clone IMAGE: 2505055 3 ', similar to the sequence of phospholipase C TR: Q62084 Q62084 Signaling protein G regulator 12 (RGS12) NM_002926 16 DNAJA4 homologous to Dna] (Hsp40), subfamily A, AF116663 member 4 (DNAJA4) or PR01472 17 Open reading frame 57, chromosome 1 (Clorf57) NM_032324 18 SH3 binding domain of protein kinase 1 (SBK1) NM_00102440 19 FKSG43 AF334945 Papiline (PAPLN) sulfated glycoprotein similar to NM_173462 proteoglycan 21 chromosomal region of lateral sclerosis NM_020919 amyotrophic 2 (juvenile) 10. Use of specific reagents of a set of molecular markers identified according to step b) of the method according to any one of the preceding claims to determine the presence or risk of developing Parkinson's disease. 11. Use according to claim 10, characterized in that said specific reagents are specific for the set of genes as defined in claim 9. 12. Use according to claim 11, characterized in that said specific reagents are the sequences SEQ ID NO: 1 to SEQ ID NO: 23. 13. Kit for carrying out the method according to any one of claims 1 to 9 comprising at least the following elements: - specific reagents of the set of molecular markers as identified in step b) of said method, - optionally the reagents and / or devices necessary and adapted to carry out the extraction of the molecular markers to be analyzed and / or the relative detection and / or quantification of these markers with respect to the quantity of the molecular markers of the same type, - possibly a sample containing molecular markers of the type chosen from a patient with Parkinson's disease in a proven way (positive control). 16. Kit according to claim 13, characterized in that said specific reagents are specific for the genes as defined in claim 9. 17. Kit according to claim 14, characterized in that the molecular markers are RNAs. 18. Kit according to claim 13 or claim 14, characterized in that said gene-specific reagents are specific primers or probes. 19. Kit according to claim 16, characterized in that said gene-specific probes are the sequences SEQ ID NO: 1 to SEQ ID NO: 23. 20. Kit according to any one of claims 13 to 17, characterized in that the positive control is a sample from a subject with a pathogenic mutation responsible for Parkinson's disease at an illness stage HH greater than or equal to 3.19. Kit according to any one of Claims 13 to 18, characterized in that the negative control is a sample derived from a subject that is free from any personal or family history of Parkinson's disease, without any other apparent disease or taking medical treatment. .