New specific erythropoietin protein and use thereof

Abstract

The present invention relates to a new variant of EPO. In this study, the inventors identified a particular EPO differing from kidney-type EPO due to a different glycosylation pattern in patients with different diseases (for example idiopathic erythrocytosis, liver / kidney disorder, tumors...). This new EPO is similar to pre-term newborn EPO produced by the liver that differs from the normal EPO produced from renal interstitial cells due to different post-translational modifications (glycosylations). This new EPO shows an increased biological activity via activation of the EPO receptor (EPOR), that can explain the origin of erythrocytosis. The detection of this specific liver-like EPO could be used as a diagnostic tool to better adapt and follow the treatments of erythrocytosis patients or in other diseases associated with the overproduction of red blood cells. Thus, the invention relates to a specific EPO protein which differs from the EPO with kidney-type glycosylation pattern.

Description

[0001] NEW SPECIFIC ERYTHROPOIETIN PROTEIN AND USE THEREOF

[0002] FIELD OF THE INVENTION:

[0003] The present invention relates to a specific EPO protein which differ from the kidneytype EPO by a different glycosylation pattern.

[0004] BACKGROUND OF THE INVENTION:

[0005] Erythropoietin (EPO), a glycoprotein hormone, is a fundamental regulator of erythropoiesis whose expression is tightly regulated by local changes in oxygen partial pressure. When oxygen levels decrease, cells initiate the stabilization of the Hypoxia Inducible Factor (HIF), which in turn initiates HIF-dependent transcriptional induction of EPO gene expression.1 5During development, EPO is expressed by neural crest cells and foetal hepatocytes. At birth, the expression in the liver decreases, whereas the interstitial cells of the kidney originating from the neural crest cells become the principal site of production.2,6In adults, EPO is produced mainly by the kidney (90%) and to a lesser extent by other organs and in particular the liver.2,7Mutations in genes of the oxygen sensing pathway have been found in patients with hereditary erythrocytosis (HE), a rare hematological disease characterized by excessive red blood cells production8: VHL9, EGLNl / PHD2w,n, EPASl / HIF2A,u, and more recently EPO14,15. While mutations were mainly associated with an increase in circulating EPO above average explaining red blood cells overproduction, the exact molecular mechanism at play in patients carrying EPO mutation is still enigmatic due to of the discrepancy between the high EPO mRNA levels identified the mutant cellular models, serum EPO levels within the normal range, and increased red blood cell mass in patients. Here, the inventors identified three novel mutations in six families located in the non-coding regions of the EPO locus that disrupts the liver-to-kidney regulatory transition of EPO expression. They studied the effects of these mutations using reporter assays and patient-derived iPSC models differentiated into Hepatic-Like Cells (HLC). The profile of patients' circulating EPO was characterized by iso-electric focusing16and compared to profiles from normal healthy adults with kidney -type EPO, patients with secondary erythrocytosis linked to liver disease and premature neonates, whose EPO expression is controlled by the liver. The activity of these different EPOs purified from blood was also evaluated. Altogether, the inventors results showing a distinctive liver-like EPO with atypical glycosylation pattern and increased activity give unprecedent understanding of puzzling observations for some patients suffering from unexplained erythrocytosis.

[0006] SUMMARY OF THE INVENTION:

[0007] In this study, the inventors identified a particular EPO differing from normal circulating kidney -type EPO due to a different glycosylation pattern in patients with different diseases (for example idiopathic erythrocytosis, liver / kidney disorder, tumors...). This new EPO is similar to pre-term newborn EPO produced by the liver that differs from the normal EPO produced from kidney interstitial cells due to different post-translational modifications (glycosylations). This new EPO shows an increased biological activity via activation of the EPO receptor (EPOR), that can explain the origin of erythrocytosis. The detection of this specific liver-like EPO could be used as a diagnostic tool to better adapt and follow the treatments of erythrocytosis patients or in other diseases associated with the overproduction of red blood cells.

[0008] Thus, the present invention relates to a specific EPO protein which differ from the kidney-type EPO by a different glycosylation pattern.

[0009] Particularly, the invention is defined by its claims.

[0010] DETAILED DESCRIPTION OF THE INVENTION:

[0011] The inventors discovered a specific EPO protein found in blood that differ from wildtype EPO of healthy adult by a different glycosylation pattern as exemplified by a more basic IEF profile after analysis in pH 2-6 gradient (shift of main isoforms from pH 4-4.4 to 4.4-5.5): this different profile is linked to a decrease in sialic acid content. In normal condition, after birth, the EPO is mainly produced by the kidney and presented a specific glycosylation pattern, therefore in the text that follows, the normal EPO (or wild type EPO) will be named " kidneytype EPO".

[0012] Thus, a first aspect of the invention relates to a specific EPO protein which differ from the kidney -type EPO by a different glycosylation pattern.

[0013] Particularly, said specific EPO protein comprises at least 3 sialic acids in less than the kidney-type EPO.

[0014] More particularly said specific EPO protein has other differences in the glycan chains. In a particular embodiment said specific EPO comprises 3, 4, 5, 6, 7, 8, 9, 10 or more sialic acids in less than the kidney-type EPO. As used herein, the term “a specific EPO protein” denotes a new variant or form of EPO also named liver-like EPO in the patent application.

[0015] As used herein, the term “erythropoietin” (EPO) or “wild-type EPO” or in the invention “kidney-type EPO” denotes a glycoprotein that is synthesized mainly by the kidney in adults, circulates in the blood stream and stimulates the proliferation and differentiation of erythroid progenitor cells in the bone marrow. The polypeptide backbone of active EPO is built from 165 aminoacids, and has three N-Glycosylation sites at Asn-24, Asn-38 and Asn-83, respectively, and one O-Glycosylation site at Ser-126. Its average carbohydrate content is about 40%. Its formula is C809H1301N2290240S5 and its Uniprot access number is P01588.

[0016] The degree of terminated glycosylation and the constitution of the sugar chains are of great importance for the biological activity of EPO, in particular its liaison to the EPO receptor that promote the erythropoietic effect. Modifications of the normal pattern of EPO glycosylation might result in altered in vivo activity and be linked to red blood cells proliferation disease and other disorders.

[0017] As used herein, the term “sialic acid” denotes a class of alpha-keto acid sugars with a nine-carbon backbone.

[0018] As used herein the term “glycan” denotes a compound consisting of a large number of monosaccharides linked glycosidically.

[0019] A second aspect of the invention relates to a specific EPO protein wherein said specific EPO protein has a basic isoelectric profile.

[0020] Particularly, said specific EPO protein has a basic IEF profile after electrophoresis in a pH 2-6 gel gradient in comparison to the well-characterized kidney-type EPO profile (shift of main isoforms from pH 4-4.4 for kidney -type EPO to 4.4-5.5 for basic EPO).

[0021] More particularly a separation between the neutral zone (where the main isoforms of the kidney-type EPO are located) and a “basic zone” is made by analyzing the IEF profile of recombinant EPO drug (rEPO) epoetin (alpha, beta, gamma) which also exhibit a “basic profile” (isoforms located around pH 4.4 to 5.2 in the pH 2-6 gel gradient). The “basic” zone starts from the most acidic band of rEPO.

[0022] In a particular embodiment, the basic isoelectric profile of the variant EPO is determined after Iso Electric Focusing (IEF) polyacrylamide gel electrophoresis (or IEF-PAGE immunodetection method) followed by transfer on a membrane and immunodetection using an anti-EPO antibody. In a particular embodiment, the variant of the EPO has a “basic” isoelectric profile and the determination of this profile resulted in the obtention of several bands in the basic zone (pH 4.4 and above) in a pH 2-6 gel gradient obtained thanks to the IEF-PAGE immunodetection method.

[0023] According to the invention, the inventors used the IEF-PAGE immunodetection method (starting with gel with gradient pH 2-6) to evaluate the migration pattern of EPO and related to differences in the glycans linked to EPO amino-acid chain. Other techniques may be used to identify / differentiate this basic EPO from kidney-type EPO that allow identification of differences in the glycans linked to EPO amino-acid sequence such as, but not limited to, mass spectrometry (MS), lectins affinity profiling, capillary IEF electrophoresis, high performance liquid chromatography (HPLC).

[0024] Thus, a third aspect of the invention relates to a specific EPO protein which differ from the kidney-type EPO by a different glycosylation pattern and which has a basic isoelectric profile.

[0025] Particularly, said specific EPO protein comprises at least 3 sialic acids in less than the kidney -type EPO and which has a basic isoelectric profile. According to the invention, the terms “specific EPO protein comprises at least 3 sialic acids in less than the kidney -type EPO” denotes that the kidney -type EPO has a difference of at least 3 sialic acids in less (or fewer) compared to the kidney-type EPO.

[0026] A fourth aspect of the invention relates to a specific EPO protein wherein said specific EPO has a gain-of-function compared to the kidney-type EPO. This gain-of-function has been measured by quantifying the level of STAT5 phosphorylation (a major signal transducer located downstream EPO-EPOR signalisation) after the addition of purified EPO on UT7 cells expressing EPOR. EPO variant has been purified from blood of patients carrying EPO mutations and cord blood, they all presented a basic IEF profile.

[0027] Uses and methods related to the basic EPO

[0028] Since, the inventors showed that the variant EPO has an increased biological activity, the identification of such variant could be used for the diagnosis and new treatments of disease where increased EPO activity has deleterious effects. Thus, a fifth aspect of the invention relates to a method for diagnosing of a disease linked with a gain-of-function of the EPO in a patient in need thereof, comprising detecting in a sample obtained from the patient the specific EPO of the invention, wherein the detection of said variant demonstrates that the patient suffers from a disease linked with a gain-of-function of the EPO.

[0029] According to the invention, the term “a disease linked with a gain-of-function of the EPO” denotes hematological diseases characterized by elevated red cell mass of secondary origins, like erythrocytosis (also known as polycythemia), Hepatic-like Erythropoietin Polycythemia (HEP), idiopathic erythrocytosis or idiopathic polycythemia or liver or kidney diseases including but not limited to tumors, liver thrombosis, cirrhosis particularly vascular cirrhosis and red blood cells (RBC) proliferation of secondary origin (tumors, fibromas etc).

[0030] The method of the invention is an ex vivo method.

[0031] As used herein the term sample denotes blood (fresh or dried), plasma, or serum.

[0032] As used herein, the term “patient” denotes a mammal, such as a rodent, a feline, a canine, and a primate. Particularly, the patient according to the invention is a human and more particularly, the patient is a human with a disease (erythrocytosis / liver / kidney / tumor) that could benefit from been linked to the specific EPO of the invention offering new understanding of the disease and therapeutic options.

[0033] To detect the variant of EPO according to the invention, a method like IEF-PAGE immunodetection can be used.

[0034] In a sixth aspect, the invention also relates to a method for treating a disease linked with a gain-of-function of the EPO in a patient diagnosed with such disease according to the method of the invention comprising the administration to said patient of an appropriate treatment to reduce the basic EPO concentration in blood (for example liver targeted gene silencing or trapping of circulating basic EPO). Efficiency of the treatment can also be followed using the same ex-vivo method used for diagnosis.

[0035] The invention will be further illustrated by the following figures and examples. However, these examples and figures should not be interpreted in any way as limiting the scope of the present invention.

[0036] FIGURES: Figure 1: IEF (isoelectric focusing electrophoresis) of EPO, immunopurified from plasmas or serum of adults, performed on a IEF gel with a 2-6 pH gradient. Human EPO profile is constituted of several isoforms (8 to 16 listed on the right part of the figure) located from the acidic to the basic region of the gel. The most intense bands are normally present in the “neutral area” defined by exclusion of the acidic and basic regions corresponding to standard EPO drugs (Aranesp (darbepoietin) specific to the acidic area and Neorecormon (recombinant epoietin), specific to the basic area). R= ratio of intensity of the second most intense band in the basic area over the most intense band in the neutral area. R>1.3 define a basic profile.

[0037] Figure 2: Schematic representation of EPO post-transcriptional modifications containing 3 N-glycosylation and one O-glycosylation sites. EPO is a glycoprotein with high proportion of carbohydrates and many negative “acidic” charges are linked to presence of sialic acids at the end of the glycan chains.

[0038] Figure 3: Sialidase (removal of all sialic acid which are terminal end of glycan chains) and PNGase (clivage of all N-glycan chains) digestion were performed on purified EPO and EPO profile was analyzed using IEF with an extended pH gradient of 2 to 10 to better evaluate the impact of acid charges removal. These Glycosidase digestion experiments demonstrated that this particular “basic” profile was likely due to glycosylation specificities as PNGase action produced similar IEF profile for the Liver-like EPO and normal wild-type EPO (WT) (as well as recombinant human EPO rHuEPO produced in CHO cells) while removal of sialic acids led to more similar wild-type and Liver-like profile even if an intense ultra-basic band present in rHuEPO but almost absent in wild-type / kidney-type EPO indicated some other difference in post-translational modifications.

[0039] Figure 4: IEF profiles of EPO isolated from patients with erythrocytosis associated with EPO mutations or associated with hepatic pathologies.

[0040] Figure 5: IEF profiles of EPO isolated from preterm newborns. Monitoring of IEF profiles of EPO isolated from preterm newborns at different developmental ages, w: weeks (calculated in weeks, after the first day of mother's amenorrhea), d: days.

[0041] EXAMPLE:

[0042] Material & Methods

[0043] Patients' samples.

[0044] All study participants signed written informed consents. Blood samples were collected for diagnostic and research purposes after the approval of the local ethics committee. Samples from newborns and children were analyzed according to ClinicalTrials.gov NCT03957863. Umbilical cord bloods were provided by the Biological Resources Center of the Saint Louis Hospital, APHP (Paris, France). After blood centrifugation, plasmas were kept frozen at -20°C until analysis.

[0045] Patients' DNA Sequencing.

[0046] All study participants signed written informed consents. Blood samples were collected for research purposes after receiving the approval from the different local ethics committees. DNA was extracted from EDTA whole blood using the QuickGene610L® and molecular screening was performed by next generation sequencing (NGS). Molecular screening was performed by high-throughput sequencing using different technologies, depending on the sequencing center: KAPA HyperPlus" (Roche) associated with IDT probes for the capture provided by Sophia Genetics. The NGS panel included the following genes: VHL, EGLN1, EGLN2, EGLN3, HIF1A, EPAS1, EPOR, SH2B3, JAK2, BPGM, and EPO. The NGS panel covers all exons, intron / exon junctions (at least 25 base pairs of the intronic sequences), and partial sequences of the 5' and 3'UTR (size depending on the gene). Detailed sequences and bed files available upon request.

[0047] In silico analysis of EPO mutations.

[0048] Links to the websites used for genetic variants analysis are provided below:

[0049] MobiDetails: https: / / mobidetails.iurc.montp.inserm.fr / MD / genes

[0050] Variants frequencies were determined using the values reported by gnomAD on the Mobidetails website. This site was also used for splicing analysis using SPiP-Splicing Pipeline Prediction.

[0051] Differentiation of induced Pluripotent Stem Cells (iPSC) into hepatocyte-like cells (HLCs).

[0052] IPSCs have been generated by the the iPSC core facility of Nantes University. Patient's PBMCs were reprogrammed by Sendai viruses expressing 0CT4, S0X2, KLF4 and c-MYC (CytoTune™-IPS 2.0 Sendai Reprogramming kit, Life Technologies). IPSCs from passage 21-25 were differentiated into HLCs as previously described18,19. IPSCs were plated in tissueculture dishes previously coated with Matrigel (Corning) at 0.05 mg / ml for 1 hour and cultured in StemMACS iPS-Brew. Once cells reached 80-85% confluence, differentiation was performed with RPMI 1640 (Life Technologies) and B27 (Life Technologies) containing Activin A (AA; Miltenyi), fibroblast growth factor 2 (FGF2; Miltenyi), bone morphogenetic protein 4 (BMP4; Miltenyi) and hepatocyte growth factors (HGFs; Miltenyi) in the following order: AA / BMP4 / FGF2 (2 days); AA (3 days), BMP4 / FGF2 (5 days) and HGF (5 days). Cells were then incubated for 5 days with Hepatocyte Culture Medium (Lonza) supplemented with Oncostatin M (Miltenyi). After 22 days of differentiation, cells were cultured in normoxia or at 1% O2 for 24 hours. For each iPSC cell line, two clones were differentiated two or three times as replicates. Each differentiation of a mutated iPSC was performed in parallel with at least one wild-type control on the same experiment.

[0053] After differentiation of hiPCS, hepatic markers were examined by Real-time RT-PCR with Taqman gene expression assays (Applied Biosystems) using a QuantStudio 5 Real-Time PCR System (Applied Biosystems). Relative quantification of the selected genes: ALB (albumin, Hs00910225_ml), HNF4 (hepatocyte nuclear factor 4, Hs00230853_ml), AFP (alpha fetoprotein, Hs00173490_ml), F0XA2 (forkhead box A2, Hs00232764_ml) was performed against a standard curve and the values were normalized against the input determined for the reference genes, RPLPO (ribosomal protein lateral stalk subunit P0, Hs99999902_ml), ACTB (actin beta, Hs99999903_ml) or RPL13A (ribosomal protein L13a, Hs04194366_gl). TaqMan Gene Expression Assay IDs (Applied Biosystems) are given in parentheses after the gene names.

[0054] Erythropoietin gene expression analysis.

[0055] RNA samples were isolated using the Nucleospin RNA Kit (Macherey Nagel). Reverse transcription of 1 pg RNA into cDNA was performed using Maxima First Strand cDNA Synthesis Kit (Applied Biosystems) as recommended by the supplier. Quantitative polymerase chain reaction (qPCR) studies were performed in triplicate using the Taqman® Univeral PCR Master Mix (Applied Biosystems) and QuantStudio 5 Real-Time PCR System (Applied Biosystems). Relative quantification of EPO (Erythropoietin, Hs00171267_ml and Hs01071097_ml) was obtained by normalisation to the mean of the reference genes, RPLPO (ribosomal protein lateral stalk subunit P0, Hs99999902_ml), ACTB (actin beta, Hs99999903_ml), and RPL13A (ribosomal protein L13a, Hs04194366_gl), using the 2'AACtmethod. TaqMan Gene Expression Assay IDs (Applied Biosystems) are given in parentheses after the gene names.

[0056] IsoElectric Focusing (IEF) analysis.

[0057] The profiles of the EPO present in the patients plasmas were assessed using the IEF-PAGE method usually applied to doping control samples as previously described20,21. Briefly, after specific immunopurification of EPO from plasma using magnetic beads coated with an anti -EPO antibody (clone 9C21D11), the purified EPO migrated on a 2-6 pH gradient IEF gel . The separated proteins were transferred on a PVDF membrane, which was then incubated with another specific anti-EPO antibody (clone AE7A5) to get the EPO profile. A discontinuous buffer transfer system was used for higher transfer efficiency (starting from the cathode: six filter papers soaked with Towbin buffer, 1.3 mM SDS; gel equilibrated with three consecutive baths of Towbin buffer, 1.3 mM SDS; membrane conditioned in 25 mM Tris, pH 10.4, 10% ethanol; four filter papers soaked in 25 mM Tris, pH 10.4, 10% ethanol; two filter papers soaked in 300 mM Tris, pH 10.4, 10% ethanol)21. Human EPO profile is constituted of several isoforms (usually 8 to 16) located from the acidic to the basic region of the gel. The most intense bands are normally present in the “neutral area” defined by exclusion of the acidic and basic regions corresponding to standard EPO drugs (Aranesp (darbepoietin) specific to the acidic area and Neorecormon (recombinant epoietin), specific to the basic area). For glycosylation studies, IEF was performed on 2-10 pH gradient, the rest of the procedure was unchanged.

[0058] The classification of EPO isoforms is based on the quantitative integration of each bands composing the IEF-PAGE pattern. Bands are defined (for the anti-doping purpose ) in two areas: -one basic including all bands of rEPO drug epoetin used as standard (the bands are numbered from 1 (most acidic in this area) to 7 (most basic) and - one endogenous area, named "neutral" including the other bands and usually the most intense bands of a wild-type human blood EPO (WT) with kidney -type pattern.

[0059] 2nd most intense band in basic area Ratio "r" is calculated based on this formula: r

[0060] most intense band in endogenous area

[0061] - Profile with r<l .3 was classified “WT”

[0062] - Profile with r>l .3 with the three most intense bands in the basic area in position 1 ,2 ,3 or 2 ,3 ,4 or 3, 4, 5 was classified “basic”

[0063] Desialylation and N Deglycosylation of EPO,

[0064] Purified EPO (about 100 mIU) in citrate buffer pH 5.6 was incubated with 2.5U of sialidase (Arthrobacter ureafaciens, Neuraminidase, ROCHE) at 37°C for 4h30 for desialylation and incubated with 2.5U of sialidase (Arthrobacter ureafaciens) and 1U of PNGase F (Flavobacterium meningosepticum, N-Glycosidase F, ROCHE) at 37°C for 4h30 for N deglycosylation. Then, the digested EPO was submitted to IEF 2-10 pH gradient and specific immunodetection as described above.

[0065] Purification of EPO,

[0066] High volumes of plasma (100-200mL) obtained from phlebotomy (patients with EPO mutations) or blood donation (healthy controls) or cord blood were immunopurified using NHS Mag magnetic beads coated with anti-EPO20. EPO was concentrated in a volume below 200 pL and buffer was exchanged to PBS. All concentrated EPOs were frozen until analysis. An aliquote was taken to determine EPO concentration using Human Erythropoietin Quantikine IVD ELISA Kit; R&D Systems.

[0067] AlphaLISA® SureFire® Ultra™ p-STAT5 Assay.

[0068] UT7 cells were cultured in alpha-MEM (gibco, 22571020) supplemented with 10% FBS, 1% L-Gln, and 1% PS in the presence of 2U / mL EPO. Cells were subcultured and maintained at a density of 300,000-500,000 cells / mL. UT7 cells were starved and EPO-deprived overnight at 37 °C, by incubation in IMDM Iscove’s medium (Gibco, 12440053) in the presence of 0.4% BSA at a density of IxlO6cells / mL. A serial dilution of commercial rHuEPO (EPREX®) in PBS / BSA 0.1% was prepared in duplicates for each concentration in a 96-well plate. 40 pL of EPO starved UT7 cells (300, 000 cells per well) were added to a flatbottom 96-well plate containing either 40 pL of rHuEPO or 40 pL of pure plasma or purified EPO to be tested (in duplicates). Following a 15 min incubation at 37°C in a humidified atmosphere, 20 pL of lysis buffer (5X) was added to each well, and cells were lysed on an orbital plate agitator (450 rpm) for 15 min at room temperature. Cell lysates were then stored at -80°C. Phosphorylation of endogenous STAT5 was measured using the AlphaLISA® SureFire® Ultra™ p-STAT5 (Tyr694 / 699) Assay Kit (ALSU-PST5) according to the manufacturer’s instructions. Briefly, 10 pL of each lysate were transferred into a 384-well Optiplate™. For each point, 5 pL of Acceptor Mix was added. Plate was then sealed with Topseal-A adhesive film, and incubated in the dark for 1 hour at room temperature. 5 pL of Donor Mix was added to each well under subdued light and the plate was again sealed with Topseal-A adhesive film and covered by a foil, followed by 1-hour incubation at room temperature in the dark. Finally, the plate was read with an ENSPIRE plate reader (Alpha Technology-compatible), using standard AlphaLISA settings.

[0069] Statistical Analysis.

[0070] All values in the figures are presented as mean ± standard error of the mean (SEM). Differences in means between multiple groups were analyzed with one-way ANOVA, or alternatively, when not applicable, with Kruskal-Wallis followed by Tukey or Dunn posthoc tests. In Figure 4, global mutant, concentration and their interaction effects were analyzed with a two-way ANOVA. Per concentration mutant effects were tested using Tukey posthoc multiple tests. All statistics were performed using GraphPad Prism software 9.5.1. Values of p<0.05 were considered statistically significant.

[0071] Results:

[0072] Discovery of the variant of the EPO and applications. Identification of new mutations in non-coding regions of the EPO gene.

[0073] Patients with erythrocytosis were selected after exclusion of classical causes of erythrocytosis (polycythaemia vera or secondary erythrocytosis associated with certain renal, cardiac, or pulmonary disorders) according to the French flow chart11. Genetic screening was performed on genomic DNA using a sequencing panel specifically designed for erythrocytosis and including genes involved in the hypoxia pathway (VHL, PHD2, HIF2A, EPO). Three different genetic variants were identified in non-coding sequences of the EPO gene in six families: the C.-252OT variant located in the promoter, the c,14-28T>C and c,14-26A>G variants located in intron 1 in the proximity of exon 2 (data not shown). History of erythrocytosis was described in a total of thirty patients. Studied patients presented erythrocytosis associated with moderate to severe elevation of hematological parameters. Genetic sequencing identified germline heterozygous mutation, except for one patient with severe disease who was homozygous for the C.-252OT variant (data not shown). All patients displayed EPO in serum within the normal ranges (5 - 25 mUI / mE). Finally, the variants were absent from the Genome Aggregation Databases (gnomAD v4), they segregated with the disease within the families, and were detected in different families, providing strong arguments for their involvement in the pathology. In order to identify the molecular mechanisms at the origin of developed erythrocytoses, the impact of these novel mutations on EPO function was investigated.

[0074] Functional studies ofEPO mutations.

[0075] Luciferase reporter constructs containing the proximal region of the EPO promoter in presence or absence of the distal 5’- and / or 3’-HREs (Hypoxia Response Element) were used to measure the impact of the mutation on EPO transcription (data not shown). Luciferase assays in HEK293 cells demonstrated significantly increased reporter gene expression with the c.-252OT mutated promoter (minimal and extended version) under hypoxic conditions for all constructs in the presence of the 5’-HRE and / or 3’-HRE (data not shown).

[0076] Regarding the functional impact of the mutations located in intron 1, we generated a luciferase reporter construct driven by the EPO regulatory elements including the intronic region (data not shown). Luciferase reporter assays demonstrated that the intronic region contains potential inhibitory regulatory elements that reduce the expression of the reporter gene under hypoxic conditions (data not shown). Based on these results, we next evaluated the impact of c,14-26A>G and c,14-28T>C mutations in presence of co-expressed GATA factors, known to be involved in E O regulation. Transient transfection experiments revealed that introduction of the mutations within these reporter constructs further counteract this decreased EPO expression, by altering the inhibitory effect either via GATA2 (c,14-28T>C), GATA3 or GATA4 (both mutations) under hypoxic conditions (data not shown). Both mutations also resulted in significantly increased HfF-2a-mediated induction oiEPO reporter gene expression (data not shown), demonstrating that these two mutations affect the regulation of EPO transcription.

[0077] Finally, we investigated the effects of all three mutations on endogenous EPO mRNA expression levels in patient-derived iPSC. As HIF-2a and GATA factors playing major role in regulation of EPO in the liver, we differentiated iPSC into Hepatic-Like Cells (HLC). Our results showed a pronounced increase in EPO expression in all mutant cells under hypoxic conditions (3.4 to 15.5 fold increase, ***p<0.001, data not shown), consistent with our RNAsequencing results (data not shown), ultimately demonstrating an impact of the mutations on EPO transcription.

[0078] Collectively, these results strongly suggest that the identified mutations in the EPO locus are causal for development of erythrocytosis due to the upregulation oi EPO expression.

[0079] Analyses of the patients' circulating EPO.

[0080] Given the complexity of the involved mechanisms (mutations associated with elevated EPO mRNA expression in patient's derived cells but circulating EPO levels in patients' serum within the normal range, data not shown), a full biochemical characterization of EPO was next performed in collaboration with the French Anti-Doping Laboratory (LADF), which developed a high-end method to characterise EPO by isoelectric focusing (IEF) in a pH gradient (pH 2 to 6 separated in three areas: basic / neutral / acidic). This methodallows separation of proteins by their charge (migration in the gel until each protein reach its isoelectric point) followed by immunodetection of EPO using a specific antibody after transfer of the proteins on a membrane. This method was developed to detect doping with EPO drug21. This analysis allows a semi-quantitative identification of the heterogeneity of circulating EPO with varying degrees of glycosylations. In healthy individuals, the isoelectric profile of circulating EPO is characterised by a distribution of total EPO up to sixteen isoforms due to different glycosylated forms, with the main isoforms distributed in the neutral or slightly basic pH range of the gel (Figure 1).

[0081] Samples from the six families reported here were examined, in addition to plasma obtained from patients with erythrocytosis associated to two EPO mutations previously described (c.32delG15and c.-136G>A14). Interestingly, all patients' samples exhibited a modified IEF profile, which resulted in a change of EPO isoforms distribution with the main EPO isoforms present in the basic pH range of the gel. As EPO is a glycoprotein with high proportion of carbohydrates, many negative “acidic” charges are linked to presence of sialic acids at the end of the glycan chains (Figure 2). To evaluate if the basic isoforms were linked to differences in the glycosylation pattern of EPO, sialidase and glycosidase digestion experiments were performed, and EPO profile was analyzed using IEF with an extended pH gradient of 2 to 10 to better evaluate the impact of acid charges removal. Our results showed that in presence of sialidase only, the IEF pattern was similar to the wild-type one except for one intense highly basic band, which was similar to the one present in recombinant EPO produced in mammalian cells. After treatment with N-glycosidase the IEF pattern was similar between WT, recombinant EPO and mutant EPO in agreement with the absence of mutation inside the amino-acid coding sequence. These results indicate that the particular “basic” profile of our patients, compared to WT EPO, is due to some glycosylation modifications suggesting a lower content in sialic acids notably (Figure 3).

[0082] Following these results, we next examined if the same IEF profile could be observed in patients with erythrocytosis associated with VHL mutations and high circulating EPO levels, which could potentially lead to glycosylation defects due to EPO overproduction. Surprisingly, their EPO profile was close to the one from healthy adults, thus ruling out the assumption that an uncontrolled excessive synthesis of EPO causes misglycosylation (data not shown). We next analysed other cases of patients with acquired erythrocytosis associated with hepatic pathologies22,23.

[0083] Interestingly, the erythrocytosis disappeared and the IEF profile recovered a regular kidney profile after the placement of a TIPS following hepatic thrombosis diagnosis, then reducing portal pressure and improving circulation, blood flow and liver function as evidenced by better levels of hepatic parameters (data not shown). Our results demonstrate that the liver produces a more «basic» isoform of EPO, also present in a case of hepatopathies (data not shown) which is associated with specific post-translational glycosylation modifications.

[0084] Given that in mammals, EPO is produced by liver during foetal life24, we next analysed EPO from pre-term newborns. Strikingly, the EPO IEF pattern in newborns displayed the same basic profile as observed in our patients (Figure 4). Hence, patients with erythrocytosis associated with EPO non-coding mutations produce EPO that is qualitatively similar to the one expressed by the foetal liver during development. Interestingly, by examining the EPO profile of a cohort of newborns from severe preterm to normal term, pre-term children shows isoforms only in the basic area and the most intense bands are those numbered 3, 4 and 5, as shown in Figure 5. These bands should correspond to isoforms produced mainly in the liver. This EPO IEF profile shifts to an adult one, by apparition of more acidic isoforms, probably linked to renal EPO expression after birth or about 41 weeks of maternal amenorrhea. This shift of profile from the basic one to the normal neutral one if followed by the shift in the most intense bands: chronologically 3, 4 and 5, then 2, 3, and 4 and finally 1, 2 and 3. Healthy adult profile is composed of one of the three more intense bands in the neutral area as shown in Figure 1. This shift is therefore related to the total developmental time of the infant (counted in weeks of maternal amenorrhea) and not to the number of days in the open air after birth (Figure 5). The same basic EPO forms could also be identified in blood from umbilical cord obtained from neonates (all at least 37 weeks of gestation) (data not shown), a profile in accordance with their developmental age (data not shown).

[0085] Our definition of the basic EPO pattern, like the ones identified in the familial erythrocytosis patients and pre-term children, is that the three most intense bands of the profile shall be present in the basic area comprised between band 1 and 6 of the rEPO standard.

[0086] Analysis of different plasma from haematological diseases, notably erythrocytosis / polycythemia, liver disease, kidney disorders, tumors have highlighted similar basic pattern, named “liver like isoforms”, where bands numbered 1,2,3 or 2,3,4 or 3,4,5 are the most intense bands as shown in Figure 1. Additionally, “basic” IEF profiles are highly reproducible but heterogenous showing at least 3-4 subtypes reflecting possibly the type or degree of the disease but also the type of pathology (data not shown).

[0087] Thus, this method allows to identify and differentiate the “basic” EPO from the normal EPO and could be used to diagnostic some disease with a shift from the kidney-type EPO to the “basic” EPO.

[0088] Functional study of the liver-like EPO.

[0089] To study the properties of this particular basic EPO, we attempted to produce it in cell lines, however, we were unable to produce a correctly glycosylated EPO that resembles the normal circulating EPO in the blood and could serve as a control (data not shown). Activity assays of pure plasma performed on the human erythroleukemia cell line (EPO-dependent UT7) were inconclusive (data not shown). Consequently, we purified and concentrated circulating EPO from blood donations of healthy carriers, from phlebotomy samples of patients with erythrocytosis carrying an EPO mutation and from cord blood, to assess the activity of the basic profile EPO. The activity of the purified EPO was measured by assaying the EPO-EPOR signaling pathway on UT7 (quantification of STAT5 phosphorylation using the AlphaScreen technology). We tested different EPO concentrations corresponding to the linear activity range of commercial EPO (data not shown). Our results indicated that the patients' EPO had consistently higher activity than normal EPO at each of the tested doses, with a statistical significance (p<0.05) for 1.2 U / mL (data not shown). Additionally, we purified EPO from cord blood of termed newborns (at least 37 weeks old), who still expressed high level of foetal liver EPO (data not shown). Despite a mixed IEF profile, these EPO still presented a significant gain-of-function compared to the activity of the adult control EPO.

[0090] REFERENCES:

[0091] Throughout this application, various references describe the state of the art to which this invention pertains. The disclosures of these references are hereby incorporated by reference into the present disclosure.

[0092] 1. Suzuki, N., Hirano, I., Pan, X., Minegishi, N. & Yamamoto, M. Erythropoietin production in neuroepithelial and neural crest cells during primitive erythropoiesis. Nat Commun 4, 2902 (2013).

[0093] 2. Zanjani, E. D., Ascensao, J. L., McGlave, P. B., Banisadre, M. & Ash, R. C. Studies on the liver to kidney switch of erythropoietin production. J Clin Invest 67, 1183-1188 (1981).

[0094] 3. Ilaria M.C. Orlando et al. Distal and proximal hypoxia response elements cooperate to regulate organ-specific erythropoietin gene expression. Haematol ogica 105, 2774-2784 (2019).

[0095] 4. Watts, D. et al. Hypoxia Pathway Proteins are Master Regulators of Erythropoiesis. International Journal of Molecular Sciences 21, 8131 (2020).

[0096] 5. Kapitsinou, P. P. et al. Hepatic HIF-2 regulates erythropoietic responses to hypoxia in renal anemia. Blood 116, 3039-48 (2010).

[0097] 6. Suzuki, N. Erythropoietin gene expression: developmental-stage specificity, cell-type specificity, and hypoxia inducibility. Tohoku J Exp Med 235, 233-240 (2015).

[0098] 7. Dame, C. et al. Erythropoietin mRNA expression in human fetal and neonatal tissue. Blood 92, 3218-25 (1998).

[0099] 8. Bento, C. et al. Genetic basis of congenital erythrocytosis: mutation update and online databases. Hum. Mutat. 35, 15-26 (2014).

[0100] 9. Ang, S. O. et al. Disruption of oxygen homeostasis underlies congenital Chuvash polycythemia. Nat Genet 32, 614-21 (2002).

[0101] 10. Percy, M. J. et al. A family with erythrocytosis establishes a role for prolyl hydroxylase domain protein 2 in oxygen homeostasis. Proc Natl Acad Sci U S A 103, 654-9 (2006). 11. Delamare M, Le Roy A, Pacault M, Schmitt L, Garrec C, Maaziz N, Myllykoski M, Rimbert A, Karaghiannis V, Aral B, Catherwood M, Airaud F, Mansour-Hendili L, Hoogewijs D, Peroni E, Idriss S, Lesieur V, Caillaud A, Si-Tayeb K, Chariau C, Gaignerie A, Rab M, Haferlach T, Meggendorfer M, Bezieau S, Benetti A, Casadevall N, Hirsch P, Rose C, Wemeau M, Galacteros F, Cassinat B, Bellosillo B, Bento C, van Wijk R, Petrides P, Randi ML, McMullin MF, Koivunen P, ECYT-3 consortium, Girodon F and Gardie B. Delamare. Characterization of genetic variants in the EGLN1 / PHD2 gene identified in a European collection of patients with erythrocytosis. Haematologica, in press.

[0102] 12. Percy, M. J. et al. A gain-of-function mutation in the HIF2A gene in familial erythrocytosis. N Engl J Med 358, 162-8 (2008).

[0103] 13. Karaghiannis, V. et al. Comprehensive in silico and functional studies for classification of EPAS1 / HIF2A genetic variants identified in patients with erythrocytosis. Haematologica (2023) doi: 10.3324 / haematol.2022.281698.

[0104] 14. Taylor, J. C. et al. Factors influencing success of clinical genome sequencing across abroad spectrum of disorders. Nat. Genet. 47, 717-726 (2015).

[0105] 15. Zmajkovic, J. et al. A Gain-of-Function Mutation in EPO in Familial Erythrocytosis. N. Engl. J. Med. 378, 924-930 (2018).

[0106] 16. Lasne, F., Martin, L. & Martin, J. A. A fast preparative method for detection of recombinant erythropoietin in blood samples. Drug Test Anal 2, 494-495 (2010).

[0107] 17. Si-Tayeb, K. et al. Urine-sample-derived human induced pluripotent stem cells as a model to study PCSK9-mediated autosomal dominant hypercholesterolemia. Dis Model Meeh 9, 81-90 (2016).

[0108] 18. DeLaF orest, A. et al. HNF4A is essential for specification of hepatic progenitors from human pluripotent stem cells. Development 138, 4143-4153 (2011).

[0109] 19. Si-Tayeb, K. et al. Highly efficient generation of human hepatocyte-like cells from induced pluripotent stem cells. Hepatology 51, 297-305 (2010).

[0110] 20. Martin, L. et al. New insights for identification of doping with recombinant human erythropoietin micro-doses after high hydration. Drug Test Anal 8, 1119-1130 (2016).

[0111] 21. Martin, L. et al. Improved detection methods significantly increase the detection window for EPO microdoses. Drug Test Anal 13, 101-112 (2021).

[0112] 22. McFadzean, A. J., Todd, D. & Tso, S. C. Erythrocytosis associated with hepatocellular carcinoma. Blood 29, 808-811 (1967). 23. Sakisaka, S. et al. Erythropoietin production in hepatocellular carcinoma cells associated with polycythemia: immunohistochemical evidence. Hepatology 18, 1357-1362 (1993).

[0113] 24. Dame, C. et al. Hepatic erythropoietin gene regulation by GATA-4. J Biol Chem 279, 2955-2961 (2004).

Claims

CLAIMS:

1. A specific EPO protein wherein said specific EPO protein has a basic isoelectric profile.

2. The specific EPO protein according to the claim 1 wherein said EPO has a basic IEF profile after electrophoresis in a pH 2-6 gel gradient in comparison to the kidney-type EPO profile.

3. A specific EPO protein which differs from the kidney -type EPO by a different glycosylation pattern.

4. The specific EPO protein according to the claim 3 wherein said EPO comprises at least 3 sialic acids in less than the kidney-type EPO.

5. The specific EPO protein according to the claim 4 wherein said EPO comprises 3, 4, 5, 6, 7, 8, 9 or more sialic acids in less than the kidney-type EPO.

6. A specific EPO protein which differs from the kidney-type EPO by a different glycosylation pattern and which has a basic isoelectric profile.

7. The specific EPO protein according to the claim 6 wherein said EPO comprises at least 4 sialic acids in less than the kidney-type EPO and which has a basic isoelectric profile.

8. A method for diagnosing of a disease linked with a gain-of-function of the EPO in a patient in need thereof, comprising detecting in a sample obtained from the patient the specific EPO according to the claims 1 to 7, wherein the detection of said variant demonstrates that the patient suffers from a disease linked with a gain-of-function of the EPO.

9. The method according to the claim 8 wherein the disease linked with a gain-of-function of the EPO is an erythrocytosis, an Hepatic-like Erythropoietin Polycythemia (HEP), an idiopathic erythrocytosis or idiopathic polycythemia or liver or kidney diseases including but not limited to tumors, liver thrombosis, cirrhosis particularly vascular cirrhosis and red blood cells (RBC) proliferation of secondary origin.

Images