Differential diagnosis of biphasic and monophasic disorders in depressive patients using a-to-i RNA editing znf267 gene

Abstract

The present invention relates to a method for differential diagnosis of bipolar and monophasic disorders in a human patient during a depressive period from a biological sample of said patient, the method comprising at least determining the relative proportion of RNA editing variants of the A to I editing RNA ZNF267 gene. The invention also relates to a method for monitoring the treatment of a depressive patient exhibiting a bipolar disorder or a monophasic disorder. Finally, the present invention provides a kit for differential diagnosis of bipolar disorder and monophasic disorder in a patient during the depressive phase.

Description

[0001] The present invention relates to a method for differential diagnosis of bipolar disorder versus unipolar disorder in a human patient during a depressive episode from a biological sample of said patient, comprising at least determining the relative proportion of RNA editing variants of the A to I edited RNA ZNF267 gene. The present invention also relates to a method for monitoring the treatment of a depressed patient presenting a bipolar disorder or a unipolar disorder. Finally, the present invention provides a kit for differential diagnosis of bipolar disorder versus unipolar disorder in a patient during a depressive episode.

[0002] Depression is one of the most common mental health disorders affecting nearly 10% of men and 20% of women worldwide and is associated with a significantly increased mortality, mainly due to suicidal behavior. The Diagnostic and Statistics Manual uses a combination of five or more different symptoms to characterize a major depressive episode, such as depressed mood, anhedonia, sleep disturbances, fatigue or indecisiveness. Among mood disorders, bipolar disorder (BD) is one of the most common and disabling disorders, affecting 1% of the world population, characterized by manic episodes, hypomania, and alternating or interwoven depressive episodes. In primary care clinics, 21% of patients treated for depression are screened positive for BD, and 2 / 3 of these patients report that they have never been diagnosed as bipolar. Indeed, Judd and colleagues have shown that patients are manic or hypomanic for less than 10% of the time and are asymptomatic for about half of the time, meaning that they are depressed for 40% of the time [1]. Thus, the average interval between BD symptom onset and correct diagnosis is estimated to be about 7 years, delaying appropriate treatment and care management and increasing the risk of suicide. Various clinical interview-based instruments are available and routinely used in practice by psychiatrists to diagnose BD, including the evaluation of manic symptoms by the Young Mania Rating Scale (YMRS), the Altmanself-rating scale (ASRM) or the Mood Disorder Questionnaire (MDQ). There is a lack of biomarkers establishing boundaries between different subtypes of depression, and the main research aim is to identify reliable and clinically useful biomarkers to distinguish BD from unipolar depression.

[0003] Recent studies have shown an association between depression and RNA alterations by extra- transcriptional mechanisms [2], including RNA methylation [3], microRNAs [4] and RNA editing [5-7]. One of the processes that occurs at the RNA level that has been studied the most is the adenosine (A) to inosine (I) conversion mediated by ADARs (adenosine deaminases acting on RNA), which bind double-stranded RNA (dsRNA) stem loops and modify A to I by deamination. Inosine is interpreted by cellular machinery as guanosine due to its similar chemical features. Thus, RNA editing can induce single amino acid substitutions in coding regions, creating new start or stop codons or modifying splice sites. In addition, it can also affect RNA stability by modifying untranslated regions UTRs and forming different microRNA isoforms. Significant differences in RNA editing have been reported in neurological or immune diseases and other pathologies. In the central nervous system (CNS), the permeability of ion channels and the response to excitatory neurotransmitters have been found to be altered by RNA editing. Recently, we identified modifications in 5-HTR2c (5-hydroxytryptamine receptor 2c) [8] and PDE8A (phosphodiesterase 8A) [9] mRNA editing in the prefrontal cortex (Brodmann area 24) of depressed suicide victims. Interestingly, phosphodiesterases, a key modulator of 5-HTR2c downstream signaling, are involved in inflammatory cell activation, memory and cognition. PDE8A mRNA editing has also been demonstrated to have diagnostic value in depression in blood from HCV patients treated with interferon-alpha

[10] , and in blood from depressed patients and suicide attempters

[11] , compared to age- and gender-matched healthy controls.

[0004] Recently, Salvetat et al.

[12] have published a study that focused on transcriptome-wide RNA editing modifications detected by RNA sequencing (RNA-Seq) to identify new genes with differential A-to-I RNA editing in blood samples from depressed patients (n=57, discovery cohort) compared to healthy controls. The diagnostic potential of this set of edited RNAs was validated by ultra-deep next generation sequencing (NGS) on 410 samples (validation cohort) from either healthy controls (n=143) or depressed patients (n=267). The same approach was applied to identify specific RNA editing sites or RNA editing isoforms (referred to as biomarkers) on specific RNA sequences (gene targets) by dichotomizing depressed patients in unipolar (n=160) or bipolar (n=95) patients. Their diagnostic performance in combination was evaluated by machine learning methods to distinguish unipolar from BD patients (see also PCT applications WO2021 / 089865 and WO2021 / 089866, published on 14 May 2021).

[0005] The current COVID-19 pandemic has led to population confinement, the demographic and social impact of which is still being evaluated. Health restrictions have already caused panic in the field of psychiatry, with an explosion of anxiety and depressive disorders. Indeed, recent studies have shown a substantial increase in the prevalence of depression since the COVID pandemic [13, 14]. Moreover, the spread and high mortality rate of COVID-19 can exacerbate the risk of mental health problems and exacerbate current psychiatric symptoms in certain individuals at risk of anxiety, depression, stress and violence.

[0006] The aim of the present invention is to overcome at least one of the drawbacks of the prior art. Furthermore, the aim of the present invention is to provide an improved method for analyzing molecular biomarkers as diagnostic, prognostic and / or predictive aids in the management of patients with depression. Furthermore, the aim is to provide a method with improved sensitivity and robustness, which is able to discriminate between patients with bipolar disorder and unipolar disorder earlier during depression and, optionally, reliably predict treatment resistance or responsiveness to increase the likelihood of successful treatment.

[0007] The need for reliable and accurate differential diagnosis of these pathologies, allowing adequate treatment, has therefore become a key necessity for the coming years.

[0008] Surprisingly, Zinc Finger Protein 267 (ZNF267) has been identified as an A-to-I RNA biomarker, which exhibits two main variant types (A-to-G and T-to-C variants) and presents sites whose editing can be particularly different between unipolar and bipolar disorders (control vs. depression and control vs. bipolar disorder) (see Table 1).

[0009] The present inventors have also demonstrated that using these specific ZNF267 RNA editing sites, isoforms and for the first pattern or their combination, preferably in combination / association with other A-to-I edited RNA genes of interest such as GAB2, IFNAR1, LYN, MDM2, PTPRC, IL17RA and PRKCB genes, have obtained high diagnostic performances (AUC ROC curves, sensitivity and specificity) in an internal validation cohort (test) and a first independent cohort (external validation cohort / reproducible study) for the differential diagnosis of bipolar patients from unipolar patients during the depressive phase.

[0010] Zinc finger protein 267 (ZNF267) is known as a member of the Kruppel-like transcription factor family, which regulates various biological processes such as cell proliferation and differentiation. It was first characterized as a transcriptional repressor of MMP-10, which is upregulated during the activation process of human hepatic stellate cells

[15] . Zinc finger protein 267 is upregulated in hepatocellular carcinoma, where it is induced by reactive oxygen species (ROS). Loss- and gain-of-function analysis revealed that ZNF267 promotes tumor cell proliferation and migration of hepatocellular carcinoma cells in vitro

[16] . Similarly, it has been found that the expression of ZNF267 is significantly upregulated in tissue samples of patients with non-alcoholic fatty liver disease (NAFLD) compared to normal liver tissue. Incubation of primary human hepatocytes with palmitic acid induced a dose-dependent induction of ZNF267 in vitro, which was associated with lipid accumulation and ROS formation

[17] . A similar oncogenic role of ZNF267 in acute lymphoblastic lymphoma (ALL) has been suggested and indicated to be regulated by miRNA 23a / b

[18] . ZNF267 has been found to be hypomethylated in osteoporosis patients, although its role in bone metabolism remains unknown

[19] . ZNF267 has also been reported to be significantly upregulated during amyloid processing and inflammation

[20] . ZNF267 was previously associated with dementia. ZNF267 was selected as one of the ten most predictive genes used in a blood-based transcriptomic panel for the diagnosis of AD

[21] . Recently, ZNF267 has been shown to be an important transcription factor in the pathogenesis of cognitive impairment and neuroinflammation produced by 1,2-diacetylbenzene (DAB) and targeted by curcumin

[22] .

[0011] It is an object of the present invention to provide a method for diagnosing Alzheimer’s disease.

[0012] These objects have been solved by the aspects of the present invention as specified below.

[0013] According to a first aspect, the present invention relates to a method for analyzing the expression of biomarker RNA molecules, comprising the steps of:

[0014] A- isolating RNA from a blood sample obtained from a subject and determining the expression of at least one biomarker RNA molecule in the isolated RNA and providing an expression profile based on the results, wherein the biomarker is the relative proportion of RNA editing at at least a given editing site and / or isoform and / or pattern or a combination of given sites and / or isoforms and / or patterns, which are capable of editing on the ZNF267 gene;

[0015] / or pattern or a combination of given sites and / or isoforms and / or patterns, which are capable of editing on the ZNF267 gene;

[0016] B- determining the expression of at least another biomarker RNA molecule in the isolated RNA based on the same isolated RNA and providing an expression profile based on the results, wherein the biomarker is the relative proportion of RNA editing at at least a given editing site and / or isoform and / or pattern or a combination of given sites and / or isoforms and / or patterns, which is capable of editing on another A-to-l edited RNA gene;

[0017] and

[0018] C- performing a combined analysis of the results using the expression profile determined in step A and step B.

[0019] In a preferred embodiment of the present method, in step B, the another A-to-l edited RNA gene is selected from the group of A-to-l edited RNA genes consisting of GAB2, IFNAR1, LYN, MDM2, PTPRC, IL17RA and PRKCB.

[0020] In another preferred embodiment of the present method, in step A, the blood sample is obtained from a patient during a depressive episode, preferably during a moderate or severe depressive episode.

[0021] In another preferred embodiment of the present application, the combined analysis of the results using the expression profile determined in step A and step B of the method of the present application provides valuable information allowing a differential diagnosis of bipolar disorder versus unipolar disorder in a patient during a depressive episode, preferably during a moderate or severe depressive episode of the patient.

[0022] As demonstrated in the examples, the combined analysis of the A-to-l edited RNA gene biomarker expression profile enhances the prognostic and predictive value of the obtained results and can provide valuable diagnostic, prognostic and / or predictive information, in particular for a differential diagnosis of bipolar disorder versus unipolar disorder in a patient during a depressive episode, preferably during a moderate or severe depressive episode of the patient.

[0023] Thus, the in vitro method of the present application can be used as an improved diagnostic, prognostic and / or predictive aid in the management of patients with depression. It can not only be used to support diagnosis, prognosis, but also to select the most suitable treatment for a patient with depression.

[0024] According to a second aspect, the present application relates to a method for in vitro differential diagnosis of bipolar disorder versus unipolar disorder during a depressive phase, preferably during a moderate or severe depressive phase, in a patient, the method comprising determining the relative proportion of RNA editing at at least a given editing site and / or isoform and / or pattern or combination of given sites and / or isoforms and / or patterns that can be edited on the ZNF267 gene in a sample of the patient.

[0025] By definition, the relative proportion of RNA editing at a given editing "site" represents the sum of editing modifications measured at this unique genomic coordinate. In contrast, an edited mRNA isoform is a unique molecule that can or not contain multiple editing modifications on the same transcript. For example, for a given transcript, the edited mRNA isoform BC contains A to I modifications on site B and site C within the same transcript. An editing pattern is a combination of editing events that occur at the sites of interest in a transcript. By definition, a transcript has as many editing patterns as possible combinations given a list of sites of interest. For example, for a given transcript with sites of interest ABC, the patterns analyzed are [A, B, C, AB, BC, AC, ABC].

[0026] The term "biomarker" refers to an editing site or isoform or pattern of RNA containing one or more different edited positions, in particular in the comparison of bipolar disorder versus unipolar disorder.

[0027] In a preferred embodiment, the method comprises at least determining at least the relative proportion of RNA editing at a given pattern or combination of patterns that can be edited on the ZNF267 gene in a sample of the patient.

[0028] In a preferred embodiment, the method according to the present application further comprises determining the relative proportion of RNA editing at at least a given editing site and / or isoform and / or pattern or combination of given sites and / or isoforms and / or patterns that can be edited on another A to I edited RNA gene selected from the group comprising GAB2, IFNAR1, LYN, MDM2, PTPRC, IL17RA and PRKCB A to I edited RNA genes, more preferably on at least 2, 3, 4, 5 or 6 of these genes.

[0029] In a further preferred embodiment, the method comprises at least determining in a patient sample the relative proportion of RNA editing of a given pattern or combination of patterns that can be edited on a GAB2, IFNAR1, LYN, MDM2, PTPRC, IL17RA or PRKCB A to I editing RNA gene.

[0030] More preferably, in the method according to the present invention, determination of the relative proportion of RNA editing at a given editing site and / or isoform and / or pattern or a combination of given sites and / or isoforms and / or patterns on the A to I ZNF267 gene and on an A to I editing RNA gene selected from the group comprising GAB2, IFNAR1, LYN, MDM2, PTPRC, IL17RA and PRKCB A to I editing RNA genes is performed in the same patient sample.

[0031] In a preferred embodiment, the method according to the present invention further comprises determining the relative proportion of RNA editing at a given editing site and / or isoform and / or pattern or a combination of given sites and / or isoforms and / or patterns that can be edited on the GAB2, IFNAR1, LYN, MDM2, PTPRC, IL17RA and PRKCB A to I editing RNA genes.

[0032] In a preferred embodiment, PCR products of the gene region of interest of the selected A to I editing RNA biomarkers are sequenced by NGS.

[0033] In a preferred embodiment, the primer pair used to obtain the PCR product of the ZNF267 gene of interest is:

[0034] ZNF267 MPx_F2 GGCTGAGGTGGTCTTGGATG (SEQ ID No.1) ZNF267 MPx_R1 CCTCGCCTTCCACTGTGATT (SEQ ID No. 2) (see Table 2).

[0035] In a preferred embodiment, the primer pairs used to obtain PCR products of the GAB2, IFNAR1, LYN, MDM2, PTPRC, IL17RA, and PRKCB genes of interest are listed in Table 7.

[0036] In a preferred embodiment, the method can also be used for in vitro diagnosis of bipolar patients in moderately or severely depressed patients.

[0037] In another preferred embodiment, the method can be used for in vitro diagnosis of unipolar patients in moderately or severely depressed patients.

[0038] In another preferred embodiment, the method is a method for in vitro diagnosing a bipolar patient from a unipolar patient.

[0039] The standard symptoms of bipolar disorder, depressive disorder, moderate or severe depression and MDD are well known to the person skilled in the art and are for example described in the DSM-5 TM book of the American Psychiatric Association (2013, pages DSM-V p123-154);

[0040] The diagnoses included in this section are bipolar I disorder, bipolar II disorder, cyclothymic disorder, substance / medication-induced bipolar and related disorder, bipolar and related disorder due to another medical condition, other specified bipolar and related disorder, and unspecified bipolar and related disorder. However, the vast majority of individuals whose symptoms meet the criteria for a full-blown symptomatic episode of mania also experience episodes of major depression during their lifetime. Bipolar II disorder (which requires a lifetime experience of at least one major depressive episode and at least one hypomanic episode) is no longer considered to be a "milder" condition than bipolar I disorder, primarily because of the amount of time spent in depression by individuals with this condition, and because the emotional instability experienced by individuals with bipolar II disorder is often accompanied by severe impairment in work and social functioning. The diagnosis of cyclothymic disorder is for adults who experience periods of hypomania and depression for at least 2 years (for children, an entire year), but who never meet the criteria for a full-blown episode of mania, hypomania, or major depression.

[0041] The standard symptoms of unipolar depression are also well known to the person skilled in the art and are for example described in the DSM-5 TM book of the American Psychiatric Association (2013, pages 155-188). Depressive disorders include disruptive mood dysregulation disorder, major depressive disorder (including major depressive episodes), persistent depressive disorder (dysthymia), premenstrual dysphoric disorder, substance / medication-induced depressive disorder, depressive disorder due to another medical condition, other specified depressive disorder, and unspecified depressive disorder.

[0042] In another aspect of the application, there is provided a method for in vitro differential diagnosis of bipolar disorder from unipolar disorder in a patient during a depressive episode according to the application, said method comprising the steps of:

[0043] a) determining in a sample of the patient to be tested the presence of at least a given edit site and / or isoform and / or

[0044] or the relative proportion of RNA editing at a given editing site and / or isoform and / or pattern or combination of editing sites and / or isoforms and / or patterns that are capable of editing on the ZNF267 mRNA gene and optionally at least one other of the GAB2, IFNAR1, LYN, MDM2, PTPRC, IL17RA and PRKCB A to I mRNA genes, preferably on 2, 3, 4, 5, 6 or 7 of these genes;

[0045] b) determining a value resulting from the ratio obtained in step a);

[0046] c) classifying the patient as bipolar or unipolar after comparing the resulting value obtained in step b) with a control value obtained for bipolar patients and / or unipolar patients, wherein the control value is determined in a manner comparable to the manner in which the resulting value obtained in step b) is determined.

[0047] In a preferred embodiment, in step c) of the method according to the application, the resulting value obtained for the patient to be tested is compared to a threshold value (such as but not limited to a cut-off value or a probability) associated with the disorder to be identified, and the patient is classified as bipolar or unipolar patient according to the selected threshold value if the resulting value is equal to, or greater than, or less than the threshold value.

[0048] According to a preferred embodiment, the sample of the patient is a biological sample, preferably blood, serum, urine, saliva, tears or plasma, more preferably a blood sample, serum, even more preferably the sample of the patient is a blood sample, even more preferably a whole blood sample.

[0049] According to another preferred embodiment of the application, the method is performed in vitro or ex vivo.

[0050] For example, but not limited to, in the method of differential diagnosis according to the application, the threshold value, cut-off value or probability can be, but is not limited to:

[0051] - Z-score calculated for unipolar or bipolar patients during the depressive phase of the learning cohort

[0052] In the present specification, the terms gene or target have the same meaning and can be used interchangeably.

[0053] In a preferred embodiment, in the in vitro method for differential diagnosis of bipolar patients from unipolar patients during the depressive phase according to the application, a combination of at least two biomarkers is selected, whether or not the combination is statistically significant between patients with bipolar or unipolar disorders, for example p-value < 10 -4.

[0054] It is also preferred the in vitro method for differential diagnosis between bipolar and unipolar patients during depressive phase according to the present application, wherein in step b) the generated values are calculated by implementing an algorithm of a multivariate method comprising:

[0055] - a ROC procedure, in particular for identifying linear combinations, which maximizes the AUC (Area Under the Curve) ROC, and wherein the equation of the respective combination is provided and can be used as a new virtual marker Z, as follows:

[0056] Z = a.(Biomarker 1) + b.(Biomarker 2) +... i.(Biomarker i) +...

[0057] n.(Biomarker n)

[0058] where i is the coefficient of calculation and (Biomarker i) is the level of the biomarker considered (i.e. the level of the RNA editing site or isoform of a given target / biomarker); and / or

[0059] - a Random Forest (RF) method applied to assess RNA editing sites and / or isoforms combinations, in particular for ranking the importance of RNA editing sites and / or isoforms, and for combining the best RNA editing sites and / or isoforms, and / or optionally

[0060] - a multivariate analysis applied to assess RNA editing sites and / or isoforms combinations for diagnosis, selected from the group consisting of, for example:

[0061] - a logistic regression model applied to univariate and multivariate analysis to estimate the relative risk of a patient at different levels of RNA editing sites or isoforms values;

[0062] - a CART (Classification And Regression Tree) method applied to assess RNA editing sites and / or isoforms combinations; and / or

[0063] - a Support Vector Machine (SVM) method;

[0064] - an Artificial Neural Network (ANN) method;

[0065] - a Bayesian network method;

[0066] - a WKNN (Weighted k-Nearest Neighbors) method;

[0067] - Partial Least Square-Discriminant Analysis (PLS-DA);

[0068] - Linear and Quadratic Discriminant Analysis (LDA / QDA), and

[0069] - Extratree algorithm (8),

[0070] - any other mathematical method combining biomarkers.

[0071] Preferably, the resulting values are calculated by the Extratree algorithm.

[0072] It is preferred that the additional biomarker selected (in addition to the ZNF267 biomarker) is one, wherein the resulting value in step b) is statistically / significantly different from the control result value obtained, p < 0.05.

[0073] It is also preferred that the method of the application, wherein for the biomarker or combination of biomarkers selected in step a) the following criteria must be fulfilled:

[0074] - coverage > 30;

[0075] - AUC > 0.6, more preferably AUC > 0.8, even more preferably AUC > 0.890;

[0076] - 0.95 > fold change > 1.05, and

[0077] - p < 0.05.

[0078] It is also preferred that the method of the application, wherein the algorithm or equation allowing the calculation of the result value or depression score Z is selected from the Z equations listed in the examples, preferably the equation implements the following combination of 8 biomarkers: ZNF267, GAB2, IFNAR1, LYN, MDM2, PTPRC, IL17RA and PRKCB A to I edited RNA biomarkers.

[0079] It is also preferred that the method of the application, wherein in step a) the relative proportion of RNA editing of a given edit and / or isoform is measured in the biological sample by NGS.

[0080] It is also preferred that the method of the application, wherein in step a) the amplicon / nucleic acid sequence used for the detection of the RNA editing site and / or isoform of the RNA transcript of the biomarker is obtained or obtainable with the primer sets listed in Table 2 and Table 7 for each selected biomarker.

[0081] According to another aspect, there is provided a method for determining the effectiveness of a treatment in a subject or for predicting or monitoring the response to a treatment in a patient, said method comprising:

[0082] I - prior to and / or during and / or after the treatment of the patient, the expression of the biomarker RNA molecules is analyzed according to the method of the application;

[0083] and

[0084] II - a combined analysis of the results is performed using the expression profiles.

[0085] In another aspect, a method for monitoring the treatment of a bipolar or unipolar disorder in a human patient during a depressive phase, said method comprising:

[0086] A) prior to the beginning of the treatment expected to be monitored, a differential diagnosis of bipolar and unipolar disorder is performed on the human patient by the method according to the application;

[0087] B) after a period of time during which the patient has received the expected treatment for the diagnosed bipolar or unipolar disorder, steps (a) to (c) of the method are repeated in order to obtain a post-treatment result value;

[0088] C) the post-treatment result value from step (c) is compared to the result value obtained before treatment.

[0089] In another aspect, the application relates to a method for determining from a biological sample whether a patient during a depressive phase will be a responder to a bipolar or unipolar disorder treatment allowing the patient to be in an affective normal state (normal emotional state), said method comprising the following steps:

[0090] A) prior to the beginning of the treatment expected to be monitored, a differential diagnosis of bipolar and unipolar disorder is performed on the human patient by the method according to the application;

[0091] B) after a period of time during which the patient has received the expected treatment for the diagnosed bipolar or unipolar disorder, steps (a) to (c) of the method of the application are repeated in order to obtain a post-treatment result value;

[0092] C) the post-treatment result value from step (c) is compared to:

[0093] - the result value obtained before treatment, and

[0094] - a result value obtained on a control bipolar or unipolar disorder patient in an affective normal state, and

[0095] If the post-treatment result value obtained in step B) is closer to the emotional normal state result value than the result value obtained before treatment, the patient is classified as a responder to the treatment.

[0096] In another aspect, the present application relates to a kit for differential diagnosis of bipolar disorder from unipolar disorder in a human patient during a moderate or severe depressive episode, said kit comprising:

[0097] 1 ) - optionally, instructions for use of the method for differential diagnosis of bipolar disorder from unipolar disorder in a human patient during a depressive episode, in order to obtain generated values, the analysis of said generated values determining whether the depressive patient is presenting bipolar disorder or unipolar disorder;

[0098] and

[0099] 2) a) a pair of primers having the sequences SEQ ID No. 1 and 2, and

[0100] b) at least 1 pair, preferably 2 pairs, 3 pairs, 4 pairs, 5 pairs, 6 pairs or 7 pairs of primers selected from the group of pairs of primers having the sequences SEQ ID No. 3 and 4, 5 and 6, 7 and 8, 9 and 10, 1 1 and 12, 13 and 14 and 15 and 16, more preferably 7 pairs of primers having the sequences SEQ ID No. 3 and 4, 5 and 6, 7 and 8, 9 and 10, 1 1 and 12, 13 and 14 and 15 and 16.

[0101] In another aspect, the present application relates to a kit for differential diagnosis of bipolar disorder from unipolar disorder in a human patient during a moderate or severe depressive episode, said kit comprising:

[0102] 1 ) - optionally, instructions for use of the method for differential diagnosis of bipolar disorder from unipolar disorder in a human patient during a depressive episode, in order to obtain generated values, the analysis of said generated values determining whether the depressive patient is presenting bipolar disorder or unipolar disorder;

[0103] and

[0104] 2) a) a pair of primers, which pair of primers allows obtaining an amplicon comprising or equivalent to the amplicon that can be obtained by the pair of primers having the sequences SEQ ID No. 1 and 2; and

[0105] b) at least 1 pair, preferably 2 pairs, 3 pairs, 4 pairs, 5 pairs, 6 pairs or 7 pairs of primers, which allow obtaining an amplicon comprising or equivalent to the amplicon that can be obtained by the pair of primers selected from the group of pairs of primers having the sequences SEQ ID No. 3 and 4,

[0106] the amplicons obtained with the pairs of primers of the group of pairs of primers 5 and 6, 7 and 8, 9 and 10, 11 and 12, 13 and 14 and 15 and 16, preferably the 7 pairs of primers which allow obtaining the 7 amplicons comprising or equivalent to the sequences SEQ ID No. 3 and 4, 5 and 6,

[0107] the 7 amplicons obtained with the 7 pairs of primers 7 and 8, 9 and 10, 11 and 12, 13 and 14 and 15 and 16.

[0108] In another preferred embodiment, the present application relates to a kit for differential diagnosis of bipolar disorder from unipolar disorder in a human patient during a moderate or severe depressive episode, said kit comprising:

[0109] 1) optionally, instructions for use of the method for differential diagnosis of bipolar disorder from unipolar disorder in a human patient during a depressive episode, in order to obtain the generated values, the analysis of said generated values determining whether the depressive patient is presenting bipolar disorder or unipolar disorder; and

[0110] 2) a) a pair of primers which allows obtaining an amplicon comprising at least all the given editing sites and / or isoforms editable on the ZNF267 mRNA gene, each primer of the pair having less than 40, preferably 35, 30 or 25 nucleotides in length and comprising at least 10, preferably 12 or 15 consecutive nucleotides of the primers having the sequences SEQ ID No. 1 and 2; and

[0111] b) at least 1, preferably 2, 3, 4, 5, 6 or 7 pairs of primers which allow obtaining an amplicon comprising at least all the given editing sites and / or isoforms editable on the mRNA genes selected from the group consisting of IFNAR1, LYN, PRKCB, MDM2,

[0112] GAB2, IL17RA and PTPRC mRNA genes, and wherein for each selected mRNA gene, the respective pair of primers has less than 40, preferably 35, 30 or 25 nucleotides in length and comprises at least 10, preferably 12 or 15 consecutive nucleotides of the pairs of primers having the sequences SEQ ID No. 3 and 4, 5 and 6, 7 and 8, 9 and 10, 11 and 12, 13 and 14 and 15 and 16,

[0113] Preferably 7 pairs of primers are allowed to obtain 7 amplicons comprising at least for each amplicon all given editing sites and / or isoforms that can be edited on the corresponding mRNA gene, and wherein the primers of the 7 pairs of primers have a length of less than 40, preferably 35, 30 or 25 nucleotides and comprise at least 10, preferably 12 or 15 consecutive nucleotides of the respective pair of primers having the sequences SEQ ID No. 3 and 4, 5 and 6, 7 and 8, 9 and 10, 11 and 12, 13 and 14 and 15 and 16.

[0114] The following examples and drawings describe certain embodiments and are included to illustrate a number of aspects of the application, and are not intended to limit the scope of the application. One skilled in the art will readily recognize from the following discussion and examples that alternative embodiments of the methods and techniques illustrated herein can be employed without departing from the scope of the application.

[0115] Other features and advantages of the present application will appear in the rest of the description, examples, drawings and tables, the legends of which are given hereafter. BRIEF DESCRIPTION OF DRAWINGS

[0116] Figure 1 : STARD flowchart for Montpellier participants (study 1 and study 2)

[0117] Figure 2 : STARD flowchart for Les Toises participants (study 3)

[0118] Figure 3A and Figure 3B : Overview of our A-I RNA editing analysis. A) Overview of the workflow of the RNA-Seq and editing panel analysis; B) Overview of the bioinformatics workflow of the targeted next generation sequencing

[0119] Figure 4: Receiver operating characteristic curves and diagnostic performance of bipolar depression in the two studies for model #154

[0120] Figure 5: Receiver operating characteristic curves and diagnostic performance of bipolar depression in the two studies for model #284

[0121] The probability of a correct response for the test set using the case-specific training random forest model (extraTree model #284) is plotted. The test dataset of study 2 contains 70 patients from the unipolar group and 28 patients from the bipolar group. The replication dataset of study 3 contains 67 patients from the unipolar group and 28 patients from the bipolar group. The number of targets used: 8 (GAB2, IFNAR1, LYN, MDM2, PTPRC, IL17RA, ZNF267 and PRKCB) and the number of RNA editing variants used: 22 (extraTree model #284).

[0122] Figure 6 : artificial intelligence program

[0123] Example 1 : Cohorts

[0124] 1.1. Subjects and clinical evaluation

[0125] Study 1 (Discovery, CHU Montpellier - France)

[0126] Depressed patients (DEP) were recruited from outpatients of the Emergency Psychiatry and Post-Acute Care Department (CHRU of Montpellier) from September 2016 to January 2019, according to the principles of the Helsinki Declaration of 1975 and its subsequent updates. The study was approved by the French local ethics committee (CPP Sud-Méditerranée IV of Montpellier, CPP No. A01978-41) and registered with the reference identifier NCT02855918.

[0127] First discovery cohort (n=57) between patients with depression (n=26) and controls (n=31) for RNA-Seq experiment and biomarker discovery. Among depressed patients, 12 were bipolar.

[0128] Study 2 (Validation, CHU Montpellier - France)

[0129] A second validation cohort of 245 Caucasian depressed patients (155 unipolar and 90 bipolar) was recruited from outpatients of the Emergency Psychiatry and Post-Acute Care Department (CHRU of Montpellier, Clinicaltrials.gov identifier: NCT02855918). The study was approved by the French local ethics committee (CPP Sud-Méditerranée IV of Montpellier, CPP No. A01978-41). All participants aged between 18 and 65 years old understood and signed a written informed consent before entering the study. Diagnosis of unipolar depression and bipolar disorder was made by a psychiatrist. During a standardized interview, the psychiatrist administered the French version of the Montgomery-Asberg Depression Rating Scale (MADRS) and the Hamilton Depression Rating Scale (HAMD) to assess the severity of depression. The Mini International Neuropsychiatric Interview (MINI) was used to assess the presence of bipolar disorder. The Mini Neuropsychiatric Interview (MINI) was used to assess the presence of bipolar disorder. Depression was rated on the Montgomery-Asberg Depression Rating Scale (MADRS) and the clinician-rated 30-item Inventory of Depressive Symptomatology (IDS-C30). Severity levels of depression (i.e. mild, moderate and severe) were defined by 7 < MADRS < 19 and / or 12 < IDS-C-30 < 23, 20 < MADRS < 34 and / or 24 < IDS-C-30 < 36, MADRS > 35 and / or IDS-C-30 > 37, respectively. Manic symptoms were assessed by the Young Mania Rating Scale (YMRS). The presence of bipolar disorder was assessed by the clinician's expertise. All patients received treatment classified in 5 categories: anxiolytics, hypnotics and sedatives, antidepressants, antipsychotics and antiepileptics. The complete flowchart of patients through the study is shown in Figure 1 : STARD flowchart of Montpellier participants (study 1 and study 2). Figure 1

[0130] Study 3 (Replication, Les Toises center - Switzerland)

[0131] A second cohort of 94 Caucasian depressed patients (67 unipolar and 28 bipolar) was recruited at the Les Toises Psychiatry and Psychotherapy Center in Lausanne, Switzerland.

[0132] All participants aged 18 years or older understood and signed a written informed consent prior to entering the study. Diagnosis of unipolar depression and bipolar disorder was made by a psychiatrist. Severity levels of depression (i.e. mild, moderate and severe) were defined by 7 < MADRS < 19, 20 < MADRS < 34, MADRS > 35, respectively. Manic symptoms were assessed by the Young Mania Rating Scale (YMRS). The presence of bipolar disorder was assessed by the clinician's expertise. All patients received treatment classified in 5 categories: anxiolytics, hypnotics and sedatives, antidepressants, antipsychotics and antiepileptics. The complete flowchart of patients through the study is shown in Figure 2 : STARD flowchart of Les Toises participants (study 3). Figure 2

[0133] 1.2. Inclusion criteria

[0134] Study 1 and Study 2

[0135] All subjects must meet the following inclusion criteria:

[0136] - Subjects aged 18 to 65 years

[0137] - Subjects who signed an informed consent

[0138] - Capable of understanding the nature, purpose and methods of the study ​​

[0139] - Able to understand and perform clinical and neuropsychological assessments.

[0140] - Subjects with a primary psychiatric diagnosis of major depressive episode according to DSM-5 criteria (presence of psychiatric comorbidities is not an exclusion criterion).

[0141] - Non-pregnant women.

[0142] Study 3

[0143] All subjects must meet the following inclusion criteria:

[0144] - Subjects aged 18 years or older

[0145] - Subjects who signed an informed consent form

[0146] - Able to understand the nature, purpose and methods of the study

[0147] - Able to understand and perform clinical and neuropsychological assessments.

[0148] - Subjects using harmful substances will not be considered.

[0149] - Subjects with a primary psychiatric diagnosis of major depressive episode according to DSM-5 criteria (presence of psychiatric comorbidities is not an exclusion criterion).

[0150] We chose to include only patients with moderate and severe severity of depression (MADRS > 20 and / or IDSC-30 > 24) in both studies, in order to avoid “borderline” patients, e.g. patients with low scores on depression scales or patients who are defined as depressed by one scale and as healthy by another.

[0151] 1.3. Biological samples

[0152] Patients included in both studies received blood samples, including standard processing (complete blood count) and 2 whole blood samples in PAXgene tubes™ to allow extraction of total RNA from blood cells and measurement of RNA editing of targets.

[0153] Example 2: Methods

[0154] The description of the process analysis is described in more detail below and in the publication Salvetat et al., Translational Psychiatry (2022) 12: 182

[12] .

[0155] 2.1. RNA extraction and qualification from whole blood

[0156] According to the manufacturer's protocol, TM Samples were recovered from blood RNA tubes, randomly distributed among different groups of extracts, and separated using a MagNA Pure 96 instrument (Roche) (LifeScience). Total RNA concentration and quality were determined using a Qubit Fluorometer (Invitrogen) and a LabChip (Perkin-Elmer, HT RNA Reagent Kit), respectively.

[0157] 2.2. Library preparation for RNA sequencing in the discovery cohort

[0158] For RNA-Seq libraries, we used a TruSeq Stranded Total RNA library kit (Illumina) specifically customized for blood samples according to the manufacturer's instructions. Briefly, 300 ng of total RNA was depleted in rRNA and globin mRNA (Ribo-Zero globin), purified, and fragmented into an average of 250 bp fragments. The first strand of cDNA was synthesized using Superscript II reverse transcriptase (Thermo Fisher Scientific, random primers), and the second strand of cDNA was synthesized using DNA polymerase I and RNase H (UTP incorporation). DNA fragments were selectively enriched to obtain libraries, which were then normalized, denatured (0.1 M NaOH), and sequenced on an Illumina NextSeq 500 (high yield, 2 × 75 bp read length). Approximately 70 million reads / sample were obtained.

[0159] 2.3. Targeted next generation sequencing library preparation and sequencing

[0160] On a Peqstar 96x thermal cycler, the region of interest of each gene was amplified using Q5 hot start high fidelity enzyme (New England Biolabs) with validated primers according to the manufacturer’s guidelines. Quality control was ensured: amplicon purity was determined with a nucleic acid analyser (LabChip Gx, Perkin Elmer) and PCR products were quantified with a fluorescence-based Qubit method (ThermoFisher Scientific). PCR products were purified with magnetic beads (High Prep PCR MAGbio system, Mokascience) and then quantified (Qubit system) to calculate the purification yield. After indexing the samples (Q5 hot start high fidelity PCR enzyme, Nextera XT indexing kit by Illumina), they were pooled and purified (Magbio PCR cleanup system). The resulting library was denatured (0.1 M NaOH), spiked with PhiX control V3 (Illumina), loaded on a sequencing cartridge (Illumina MiSeq Reagent Kit V3 or Illumina NextSeq 500 / 550 Mid-Output) according to Illumina’s guidelines and sequenced using ultra-deep sequencing at standard concentration (1 x 150 bp read length). Around 1 million reads / sample were obtained.

[0161] 2.4. Bioinformatic analysis of A to I RNA editing panel data

[0162] Paired-end reads generated by the lllumina NextSeq 500 were demultiplexed using bcl2fastq (version 2.17.1.14, lllumina). Sequencing quality was performed using FastQC (version 0.11.7) and MultiQC software. Identification and quantification of A-to-l editing events were performed using RNAEditor (version 1.0) with default parameters. RNAEditor accepts FASTQ files as input and implements a fully automated workflow that includes a mapping step using BWA with the human genome version (GRCh38 release-83), followed by a PCR duplicates removal step, a local realignment and base quality score recalibration step using the Genome Analysis Toolkit (GATK4). To detect RNA editing events, a Unified Genotyper of GATK4 was used, followed by several purification steps to reduce the number of false positives (exclusion of known SNPs, elimination of variants in splice junctions, removal of variants in homopolymers), and finally an RNA editing event annotation step. We used high-confidence editing filtering criteria for further analysis (base quality > 25, mapping quality > 20, mean / median coverage > 30x, minimum editing reads > 2, editing degree variation > 10%, removal of edits with editing degree of 100%, and Wilcoxon test p-value < 0.05). Functional annotation of editing events was then performed with ANNOVAR, RepeatMasker (http: / / repeatmasker.org) and REDIportal. Alu editing index (AEI) was performed according to the method described by Bazak et al. Figure 3A A flowchart of our A-to-l RNA editing panel analysis is shown in Figure 1.

[0163] 2.5. Bioinformatic analysis of targeted sequencing data

[0164] Sequencing data were downloaded from the Illumina NextSeq 500. Sequencing quality was performed using FastQC software (version 0.11.7, https: / / github.com / s-andrews / FastQC / ). For further analysis, a minimum sequencing depth of 10,000 reads per sample was considered. A pre-processing step consisting of removal of adaptor sequences and filtering of sequences according to length and quality score was performed. Short reads (<100 nt) and reads with an average QC <20 were removed. To improve sequence alignment quality, the flexible read trimming and filtering tool for Illumina NGS data (fastp version) was used. After performing the pre-processing step, an additional quality control was performed on each cleaned fastq file before further analysis. Alignment of processed reads was performed using bowtie2 (version 2.2.9) with end-to-end sensitive mode. Alignment was performed against the reference human genome sequence GRCh38. Non- uniquely aligned reads, unaligned or containing insertions / deletions (INDELs) were removed from downstream analysis by the SAMtools software (version 1.7). SAMtools mpileup was used for SVN calling. The finding of edited positions in the alignment was performed by using an in-house script to count the number of different nucleotides in each genomic position. For each position, the script calculates the percentage of reads with a “G” [number of “G” reads / (number of “G” reads + number of “A” reads) * 100]. Genomic positions with a percentage in “G” reads >0.1 are “A to I edited sites” by definition. The last step is to calculate the percentage of all possible isoforms and / or patterns for each transcript. By definition, the relative proportion of RNA editing at a given edited “site” represents the sum of editing modifications measured at that unique genomic coordinate. In contrast, an edited mRNA isoform is a unique molecule that can contain or not multiple editing modifications on the same transcript. For example, for a given transcript, the edited mRNA isoform BC contains an A to I modification on site B and site C within the same transcript. An editing pattern is a combination of editing events that occur at the sites of interest in a transcript. By definition, a transcript has as many editing patterns as possible combinations for a given list of sites of interest. For example, for a given transcript with sites of interest ABC, the patterns analyzed are [A, B, C, AB, BC, AC, ABC]. A relative proportion of at least 0.1% was set as a threshold in order to be included in the analysis. A flowchart of our A to I RNA targeted next generation sequencing pipeline is shown in Figure 3B Figure 3A and Figure 3B ​Summary of our A-I RNA editing analysis. A) Summary of the workflow of RNA-Seq and editing panel analysis; B) Summary of the bioinformatics workflow of targeted next generation sequencing

[0165] 2.6. Statistical analysis of data

[0166] All statistics and numbers were computed with the "R / Bioconductor" statistical open source software. BMK (i.e. RNA editing site and editing isoform / pattern of the targeted gene) values are usually expressed as mean ± standard error of the mean (SEM). To ensure normally distributed data, each biomarker data was transformed using recipe package v0.2.0 (with step_nzv, step_BoxCox and step_normalize parameters) in each dataset (study 2 and study 3). Depending on normality and sample variance distribution in each cohort, the most appropriate test between Mann-Whitney rank sum test, Student's t-test or Welch's t-test was used for difference analysis. Highly correlated significant variables were removed using the findCorrelation function from the caret package. Significant variables with a coefficient of variation (CV) from the reproducibility and replicability analysis above 30% have also been removed.

[0167] All selected RNA editing variants were combined with age, gender, psychiatric treatment and addiction using the extraTree algorithm (one of the latest machine learning algorithms for classification). The extraTree method derived from random forest

[23] requires the use of a training set for building the model and a test set for validating the model. We shared our dataset: about 60% of the dataset was used for the learning phase and 40% for the test phase. This sharing was random and respects the initial proportions of the various regulations in each sample. We used a grid learning method for each individual tree, where we stated certain maximum parameter sets (ntree = 1000, mtry = (1, 20), splitrule = "extratrees", and min.node.size = (1, 15)). The sub-trees were trained with a classic cross-validation. RF results are shown on the test dataset (N = 98 samples) from study 2 and the replication dataset of study 3 from an unseen algorithm. Implementation was done using the R ranger package (version 0.13.1) and the R caret package (version 6.0-90).

[0168] Example 3: Results

[0169] 3.1 ZNF267 in the discovery cohort (Study 1, editing panel analysis)

[0170] Using our RNA editing panel pipeline, we identified 40,398 A-to-I edited positions in at least one sample with high confidence (see Methods and Figure 3A ). Two major variant types were identified (A-to-G and T-to-C variants), which represent 72.6% of the RNA variants. Most A-to-I edited sites present a moderate editing degree, with 10-30% editing degree representing the largest proportion. This RNA editing is mainly present in introns and 3' untranslated regions (UTRs) as well as Alu repeat regions, and has a uniform redistribution throughout the genome. We then performed a differential analysis to identify sites whose editing can be particularly different between unipolar and bipolar disorders, between controls and depression, and between controls and bipolar disorder.

[0171] To select genes of interest, we used stringent statistical criteria (coverage > 50, p-value < 0.05, AUC > 0.750, 0.833 > fold change > 1.2). ZNF267 is a unique biomarker selected for this diagnostic performance (Table 1).

[0172] 3.2 ZNF267 in the validation cohort (Study 2, targeted sequencing analysis)

[0173] Analysis of ZN267 RNA edited amplicon on clinical samples (Study 2) based on ultra-deep targeted sequencing on our RNA editing measurement platform. Indeed, NGS technology offers a unique opportunity to study deep RNA editing events. Our platform measures adenosine-to-inosine (A-to-I) modifications. The biological process is standard, with an RNA extraction step, followed by a 2-step PCR sequencing library preparation. For the bioinformatics pipeline, we used our A-I RNA editing targeted NGS pipeline. The bioinformatics pipeline is fully automated and parallelized Figure 3B

[0174] 3.2.1 ZNF267 primers used to detect RNA editing sites The region of interest was amplified using validated primers and the PCR products were purified and sequenced by NGS (see Table 2).

[0175]

[0176] 3.2.2 Quality control of sequencing data All quality standards and requirements were met with the application of our algorithm:

[0177] - The coverage of each biomarker after filtering must be greater than 10,000 reads.

[0178] - The percentage of correctly sequenced sequences after filtering (Q25) is greater than 80%.

[0179]

[0180] 3.2.3 Differential biomarkers of ZNF267 in cohort 2 UN vs BP ​

[0181] In the present study, the term "biomarker" refers to an editing site or isoform or pattern of RNA that contains one or more positions that are differentially edited in UN vs. BD comparison.

[0182] Differential analysis of all detected RNA editing biomarkers of ZNF267 amplicon was performed on the large cohort of 245 participants. 58 biomarkers were found to be statistically significant with adjusted p-value < 0.05 between unipolar depression and bipolar depression in the validation cohort (Table 3).

[0183] 3.2.4 mROC combination of ZNF267 biomarkers in cohort 2 UN vs BP

[0184] To evaluate the diagnostic performance of ZNF267 significant biomarkers in the 2nd cohort, we used the mROC method, a dedicated program to identify linear combinations, which maximizes the AUC (Area Under the Curve) ROC. The results of the top 5 best combinations obtained are shown in Table 4.

[0185] 3.3 ZNF267 in relation to other biomarkers of interest (Study 2 and Study 3, targeted sequencing analysis)

[0186] In addition to ZNF267 biomarkers, we have selected 7 targets of interest for the diagnosis of unipolar depression vs. bipolar depression based on our previous results [salvetat et al. transl psych 2020 and WO2021089866A1]. These targets are IFNAR1, GAB2, IL17RA, LYN, MDM2, PRKCB and PTPRC. The same A-IRNA editing targeted NGS pipeline was applied as for ZNF267.

[0187] To combine all A to I RNA editing biomarkers, we used machine learning methods (extraTrees, R package caret and ranger) to develop a clinical decision algorithm.

[0188] The best algorithm performances using 8 targets are represented in Table 5.

[0189] Using the same approach, the best algorithm performances using 7 targets without ZNF267 are presented in Table 6. As we can see, the best performances are obtained with ZNF267 (see Table 5).

[0190] For Example 1, the extratree algorithm model #154 gave an AUC ROC curve of 0.896 (CI 95%: [0.835-0.956]), a sensitivity of 82.1%, and a specificity of 80.0% (see Figure 4), allowing to clearly distinguish unipolar and bipolar patients on the study 2 test dataset (internal validation). The same model (#154) gave an AUC ROC curve of 0.887 (CI 95%: [0.811-0.952]) with a sensitivity of 85.7% and a specificity of 81.8% on the study 3 replication dataset (external validation) (see Figure 4), allowing to confirm the diagnostic performance of study 2 on an independent cohort (Figure 4: Receiver Operating Characteristic Curve and diagnostic performance of Bipolar Depression in two studies for model #154).

[0191] The probability of a correct response for the test set using the case-specific trained random forest model (extraTree model #154) was plotted. The test dataset of study 2 (internal validation) contained 70 patients from the unipolar group and 28 patients from the bipolar group. The replication dataset of study 3 (external validation) contained 67 patients from the unipolar group and 28 patients from the bipolar group. The number of targets used: 8 (GAB2, IFNAR1, LYN, MDM2, PTPRC, IL17RA, ZNF267, and PRKCB), and the number of RNA edited variants used: 22 (extraTree model #154).

[0192] For Example 2, the extratree algorithm model #284 gave an AUC ROC curve of 0.901 (CI 95%: [0.840-0.959]), a sensitivity of 80.6%, and a specificity of 85.1% (see Figure 5), allowing to clearly distinguish unipolar and bipolar patients on the study 2 test (internal validation) dataset. The same model (#284) gave an AUC ROC curve of 0.906 (CI 95%: [0.832-0.964]) with a sensitivity of 85.7% and a specificity of 80.3% on the study 3 replication (external validation) dataset, allowing to confirm the diagnostic performance of study 2 on an independent cohort.

[0193] Figure 5: Receiver Operating Characteristic Curve and diagnostic performance of Bipolar Depression in two studies for model #284.

[0194] The probabilities of correct responses of the test set using the case-specific training random forest model (extraTree model #284) were plotted. The test dataset of study 2 (internal validation) contained 70 patients from the unipolar group and 28 patients from the bipolar group. The replication dataset of study 3 (external validation) contained 67 patients from the unipolar group and 28 patients from the bipolar group. The number of targets used: 8 (GAB2, IFNAR1, LYN, MDM2, PTPRC, IL17RA, ZNF267, and PRKCB), and the number of RNA editing variants used: 22 (extraTree model #284).

[0195] Bibliography :

[0196] 1. Judd LL, Akiskal HS, Schettler PJ, Endicott J, Maser J, Solomon DA, et al. The longterm natural history of the weekly symptomatic status of bipolar I disorder. Arch Gen Psychiatry. 2002;59:530-7.

[0197] 2. Jung Y, Goldman D. Role of RNA modifications in brain and behavior. Genes Brain Behav. 2018;17:e12444.

[0198] 3. Shelton RC. The molecular neurobiology of depression. Psychiatr Clin North Am. 2007;30:1-11.

[0199] 4. Lee SY, Lu RB, Wang LJ, Chang CH, Lu T, Wang TY, et al. Serum miRNA as a possible biomarker in the diagnosis of bipolar II disorder. Sci Rep. 2020;10:1131.

[0200] 5. Giacopuzzi E, Gennarelli M, Sacco C, Filippini A, Mingardi J, Magri C, et al. Genome-wide analysis of consistently RNA edited sites in human blood reveals interactions with mRNA processing genes and suggests correlations with cell types and biological variables. BMC Genomics. 2018; 19:963.

[0201] 6. Yang W, Wang Q, Kanes SJ, Murray JM, Nishikura K. Altered RNA editing of serotonin 5-HT2C receptor induced by interferon: implications for depression associated with cytokine therapy. Brain Res Mol Brain Res. 2004; 124:70-8.

[0202] 7. Lyddon R, Dwork AJ, Keddache M, Siever LJ, Dracheva S. Serotonin 2c receptor RNA editing in major depression and suicide. World J Biol Psychiatry. 2013; 14:590-601.

[0203] 8. Weissmann D, van der Laan S, Underwood MD, Salvetat N, Cavarec L, Vincent L, et al. Region-specific alterations of A-to-I RNA editing of serotonin 2c receptor in the cortex of suicides with major depression. Transl Psychiatry. 2016; 6:e878.

[0204] 9. Chimienti F, Cavarec L, Vincent L, Salvetat N, Arango V, Underwood MD, et al. Brain region-specific alterations of RNA editing in PDE8A mRNA in suicide decedents. Transl Psychiatry. 2019;9:91.

[0205] 10. Salvetat N, Van der Laan S, Vire B, Chimienti F, Cleophax S, Bronowicki JP, et al. RNA editing blood biomarkers for predicting mood alterations in HCV patients. J Neurovirol. 2019;25:825-36.

[0206] 11. Salvetat N, Chimienti F, Cayzac C, Dubuc B, Checa-Robles F, Dupre P, et al. Phosphodiesterase 8A to discriminate in blood samples depressed patients and suicide attempters from healthy controls based on A-to-I RNA editing modifications. Transl Psychiatry. 2021;11:255.

[0207] 12. Salvetat N, Checa-Robles FJ, Patel V, Cayzac C, Dubuc B, Chimienti F, Abraham JD, Dupre P, Vetter D, Meruez S, Lang JP, Kupfer DJ, Courtet P, Weissmann D. A game changer for bipolar disorder diagnosis using RNA editing-based biomarkers. Transl Psychiatry. 2022 May 4;12(1):182.

[0208] 13. Bueno-Notivol J, Gracia-Garcia P, Olaya B, Lasheras I, Lopez-Anton R, Santabarbara J. Prevalence of depression during the COVID-19 outbreak: a meta-analysis of community-based studies. Int J Clin Health Psychol. 2021;21:100196.

[0209] 14. Wu T, Jia X, Shi H, Niu J, Yin X, Xie J, et al. Prevalence of mental health problems during the COVID-19 pandemic: A systematic review and meta-analysis. J Affect Disord. 2021;281:91-98.

[0210] 15. Schnabl B, Hu K, Muhlbauer M, Hellerbrand C, Stefanovic B, Brenner DA, et al. (2005): Zinc finger protein 267 is up-regulated during the activation process of human hepatic stellate cells and functions as a negative transcriptional regulator of MMP-10. Biochemical and biophysical research communications. 335:87-96.

[0211] 16. Schnabl B, Valletta D, Kirovski G, Hellerbrand C (2011): Zinc finger protein 267 is up-regulated in hepatocellular carcinoma and promotes tumor cell proliferation and migration. Experimental and molecular pathology. 91 : 695-701.

[0212] 17. Schnabl B, Czech B, Valletta D, Weiss TS, Kirovski G, Hellerbrand C (2011): Increased expression of zinc finger protein 267 in non-alcoholic fatty liver disease. International journal of clinical and experimental pathology. 4: 661-666.

[0213] 18. Lu L, Chen XM, Tao HM, et al. (2015) Regulation of the expression of zinc finger protein genes by microRNAs enriched within acute lymphoblastic leukemia-derived microvesicles. Genet Mol Res. 14(4): 11884-1189

[0214] 19. Cheishvili D, Parashar S, Mahmood N, Arakelian A, Kremer R, Goltzman D, et al. (2018): Identification of an Epigenetic Signature of Osteoporosis in Blood DNA of Postmenopausal Women. Journal of bone and mineral research: the official journal of the American Society for Bone and Mineral Research. 33: 1980-1989.

[0215] 20. Patel S, Howard D, Chowdhury N, et al. (2021) Characterization of human genes modulated by por-phyromonas gingivalis highlights the ribosome, hypothalamus, and cholinergic neurons. Front Immunol. 12: 646259.

[0216] 21. Fehlbaum-Beurdeley P, Sol O, Desire L, Touchon J, Dantoine T, Vercelletto M, et al. (2012) Validation of Aclarus Dx™, A Blood-Based Transcriptomic Signature for the Diagnosis of Alzheimer’s Disease. J Alzheimers Dis. 32: 169-81.

[0217] 22. Nguyen HD, Jo WH, Hoang NHM, Kim MS. Curcumin-Attenuated TREM-1 / DAP12 / NLRP3 / Caspase-1 / IL1B, TLR4 / NF-κB Pathways, and Tau Hyperphosphorylation Induced by 1,2-Diacetyl Benzene: An in Vitro and in Silico Study. Neurotox Res. 2022 Oct;40(5):1272-1291.

[0218] 23. Pierre Geurts Damien Ernst Louis Wehenkel. Extremely randomized trees. Mach Learn (2006) 63:3-42 DOI 10.1007 / s10994-006-6226-1

[0219] Table 1 : Differential A to I editing sites of ZNF267 gene between healthy controls and depressed patients, healthy controls and depressed bipolar patients, and depressed unipolar patients and depressed bipolar patients Table 2: ZNF267 validation primers used to detect RNA editing sites .

[0220]

[0221] Table 3: Differential A to I editing biomarkers of ZNF267 amplicons between depressed unipolar and depressed bipolar in the internal validation cohort (Study 2)

[0222]

[0223] Table 4: Diagnostic performance of ZNF267 editing biomarkers combination between depressed unipolar and depressed bipolar in the internal validation cohort (Study 2) Table 5: Top 5 best extratree models using ZNF267 to diagnose depressed unipolar vs depressed bipolar

[0224]

[0225]

[0226]

[0227]

[0228] Table 6: Example of extratree models using 6 targets of interest without ZNF267 to diagnose depressed unipolar vs depressed bipolar Table 7: Primers used to detect MDM2, PRKCB, IFNAR1, LYN, GAB2, IL17RA and PTPRC RNA editing sites

[0229]

[0230] ​

[0231]

[0232]

[0233] ​ ​

[0234]

[0235] Target list: GAB2, IFNAR1, LYN, MDM2, PTPRC, and PRKCB

[0236] ​ ​

[0237]

[0238]

Claims

1. An in vitro method for analyzing the expression of biomarker RNA molecules, the method comprising the following steps: A- isolating RNA from a blood sample obtained from the subject and determining the expression of at least one biomarker RNA molecule in the isolated RNA and providing an expression profile based on the results, wherein the biomarker is a relative proportion of RNA editing at at least a given editing site and / or isoform and / or pattern or a combination of given sites and / or isoforms and / or patterns that is capable of editing on the ZNF267 gene; B- based on the same isolated RNA, determining the expression of at least another biomarker RNA molecule in the isolated RNA and providing an expression profile based on the results, wherein the biomarker is the relative proportion of RNA editing at at least a given editing site and / or isoform and / or pattern, or a combination of given sites and / or isoforms and / or patterns, that is capable of editing on another A to I editing RNA gene; as well as C- Perform a combined analysis of the results using the expression profiles determined in step A and step B.

2. The method of claim 1, wherein in step B, the another A to I editing RNA gene is selected from the group of A to I editing RNA genes consisting of GAB2, IFNAR1, LYN, MDM2, PTPRC, IL17RA and PRKCB A to I editing RNA genes.

3. The method according to claim 1 or 2, wherein in step A, the blood sample is obtained from the patient during a period of depression, preferably during a period of moderate or severe depression.

4. A method for the in vitro differential diagnosis of bipolar disorder from unipolar disorder in a patient during a period of moderate or severe depression, the method comprising determining the relative proportion of RNA editing in a sample from the patient at at least a given editing site and / or isoform and / or pattern or a combination of given sites and / or isoforms and / or patterns, wherein the given editing site and / or isoform and / or pattern or the combination of given sites and / or isoforms and / or patterns is capable of editing on the ZNF267 gene.

5. The method of claim 4, further comprising determining the relative proportion of RNA editing at at least a given editing site and / or isoform and / or pattern or a combination of given sites and / or isoforms and / or patterns, preferably in the same sample, wherein the given editing site and / or isoform and / or pattern or a combination of given sites and / or isoforms and / or patterns is capable of editing on another A to I editing RNA gene selected from the group comprising GAB2, IFNAR1, LYN, MDM2, PTPRC, IL17RA and PRKCB A to I editing RNA biomarkers, more preferably on at least 2, 3, 4, 5 or 6 of these genes.

6. The method of claim 4 or 5, further comprising determining the relative proportion of RNA editing at at least a given editing site and / or isoform and / or pattern or a combination of given sites and / or isoforms and / or patterns that can edit on GAB2, IFNAR1, LYN, MDM2, PTPRC, IL17RA and PRKCB A to I editing RNA genes.

7. The method according to one of claims 1 to 6, wherein the PCR products of the gene region of interest of the selected A to I edited RNA gene are sequenced by NGS.

8. The method according to claim 1 , wherein: a) The primer pairs used to obtain the PCR product of the ZNF267 gene of interest are: ZNF267 MPx_F2 GGCTGAGGTGGTCTTGGATG(SEQ ID No.1) ZNF267 MPx_R1 CCTCGCCTTCCACTGTGATT (SEQ ID No. 2); and b) The primer pairs used to obtain PCR products of the IFNAR1, LYN, PRKCB, MDM2, GAB2, IL17RA and PTPRC genes of interest are listed in Table 7 (SEQ ID No. 3 to 16), respectively.

9. Method for the in vitro differential diagnosis of bipolar disorder from unipolar disorder in a patient during a moderate or severe depressive phase according to one of claims 4 to 8, comprising the following steps: a) determining the relative proportion of RNA editing at at least a given editing site and / or isoform and / or pattern or a combination of given sites and / or isoforms and / or patterns in a sample from said patient to be tested, said given editing site and / or isoform and / or pattern or a combination of given sites and / or isoforms and / or patterns being capable of editing on the ZNF267 mRNA gene and optionally being capable of editing on at least one further one of the GAB2, IFNAR1, LYN, MDM2, PTPRC, IL17RA and PRKCB A to I editing RNA genes, preferably on 2, 3, 4, 5, 6 or 7 of these genes; b) determining a value resulting from the ratio obtained in step a); c) classifying said patient as biphasic or monophasic after comparing said generated value obtained in step b) with a control value obtained for biphasic patients and / or monophasic patients, wherein said control value is determined in a manner comparable to said generated value obtained in step b).

10. The method according to claim 9, wherein in step c), the generated value obtained for the patient to be tested is compared with a threshold value (such as but not limited to a cut-off value or probability) associated with the disorder desired to be identified, and if the generated value is equal to, or greater than, or less than the threshold value, the patient is classified as a bipolar patient or a monopolar patient according to the selected threshold value.

11. The method according to one of claims 1 to 10, wherein the sample of the patient is a biological sample, preferably a blood, serum, urine, saliva, tear or plasma sample, more preferably a blood sample, serum, even more preferably the sample of the patient is a blood sample, even more preferably a whole blood sample.

12. A method for determining the effectiveness of a treatment in a subject or for predicting or monitoring a treatment response in a patient, the method comprising: I- analyzing the expression of biomarker RNA molecules according to the method of claims 1 to 3 before and / or during and / or after treatment of said patient; as well as II- Perform a combined analysis of the results using the expression profiles determined in step C for each analysis.

13. A method for monitoring treatment of bipolar disorder or unipolar disorder in a human patient during a moderate or severe depressive period based on a biological sample of the patient, the method comprising: A) prior to initiating the treatment desired to be monitored, differentially diagnosing bipolar disorder from unipolar disorder in the human by the method according to any one of claims 9 and 10; B) repeating steps (a) through c) of the method after a period of time during which the patient receives the desired treatment for the diagnosed bipolar disorder or unipolar disorder to obtain a post-treatment outcome value; C) comparing said post-treatment result value from step (c) with the result value obtained before treatment.

14. A method for determining, based on a biological sample, whether a patient during a depressive phase will be a responder to a bipolar disorder treatment or a unipolar disorder treatment that allows the patient to be in a euthymic state (normal mood state), the method comprising the steps of: A) prior to initiating the treatment desired to be monitored, differentially diagnosing bipolar disorder from unipolar disorder in the human by the method according to any one of claims 9 to 10; B) repeating steps (a) to c) of the method of the present invention after a period of time during which the patient receives the desired treatment for the diagnosed bipolar disorder or unipolar disorder to obtain a post-treatment outcome value; C) comparing said post-treatment outcome value from step (c) with: - the result value obtained before treatment, and - the outcome value is known or obtained for a control group of bipolar or unipolar patients in a euthymic state, and If the post-treatment outcome value obtained in step B) is closer to the affective normal state outcome value than the outcome value obtained before treatment, the patient is classified as a responder to the treatment.

15. A kit for differential diagnosis of bipolar disorder from unipolar disorder in a human patient during a moderate or severe depressive phase, the kit comprising: 1) - Optionally, instructions for using a method for differential diagnosis of bipolar disorder from unipolar disorder in a human patient during a depressive phase, so as to obtain a generated value, analysis of which determines whether the depressed patient presents with bipolar disorder or unipolar disorder; as well as 2) a) a primer pair having the sequences SEQ ID No. 1 and 2, and b) at least one pair, preferably two, three, four, five, six or seven pairs of primers, selected from the group of primer pairs having sequences of SEQ ID Nos. 3 to 16, more preferably seven pairs of primers having sequences of SEQ ID Nos. 3 to 16.

16. A kit for differential diagnosis of bipolar disorder from unipolar disorder in a human patient during a moderate or severe depressive phase, the kit comprising: 1) - Optionally, instructions for using a method for differential diagnosis of bipolar disorder from unipolar disorder in a human patient during a depressive phase, so as to obtain a generated value, analysis of which determines whether the depressed patient presents with bipolar disorder or unipolar disorder; as well as 2) a) a pair of primers allowing to obtain an amplicon comprising or identical to the amplicon obtainable by the primer pair having the sequences SEQ ID No. 1 and 2; and b) at least one, preferably two, three, four, five, six or seven pairs of primers, which allow obtaining amplicons comprising or identical to the amplicons obtainable by a pair of primers selected from the group of primer pairs having sequences SEQ ID Nos. 3 to 16, preferably seven pairs of primers, which allow obtaining seven amplicons comprising or identical to the seven amplicons obtainable by seven pairs of primers having sequences SEQ ID Nos. 3 to 16.

Images