HAEMATOPOIETIC STEM CELL TREATMENT

Abstract

The present invention relates to modified haematopoietic stem cells, methods for preparing them, and their use in therapy; as well as methods and reagents for expanding haematopoietic stem cells (HSC) and methods for treating haematological disorders. A particular aspect relates to a method for treating a disease or condition characterised by elevated YTHDF2 expression comprising administering to a patient an effective amount of an YTHDF2 inhibitor. Certain aspects of the invention rely on YTFIDF2 protein level or function and use of YTFIDF2 inhibitors.

Claims

1. A method for treating a disease, disorder or condition characterised by elevated YTHDF2 expression in a patient in need thereof, comprising administering to the patient an effective amount of an YTHDF2 inhibitor.

2. The method according to claim 1, wherein the disease, disorder or condition is characterised by elevated YTHDF2 expression is a haematological cancer.

3. The method according to claim 2, wherein the haematological cancer is selected from the group consisting of: acute lymphocytic leukaemia (AML), chronic lymphocytic leukaemia (CLL), diffuse large B-cell lymphoma (DLBCL), T-cell acute lymphoblastic leukaemia (T-ALL) and B-cell acute lymphoblastic leukaemia (B-ALL).

4. The method according to claim 2 or 3, wherein the cancer is characterised by cancer cells that express increased levels of YTHDF2 mRNA or YTHDF2 protein relative to normal cells.

5. The method according to any of the preceding claims, wherein the patient is a human.

6. The method according to any of the preceding claims, wherein the YTHDF2 inhibitor is a small molecule compound or a nucleic acid based inhibitor molecule.

7. The method according to claim 6, wherein the nucleic acid based inhibitor molecule is selected from the group consisting of a small interfering RNA interference molecule (RNAi), an antisense oligonucleotides (ASO), and a nucleic acid aptamer.

8. The method according to claim 6 or 7, wherein the nucleic acid based inhibitor molecule is a nucleic acid molecule capable of inhibiting mRNA of YTHDF2.

9. The method according to any of claims 1-8, wherein the YTHDF2 inhibitor is specific for YTHDF2.

10. The method according to any of claims 1-9, wherein the amount of the YTHDF2 inhibitor is effective to reduce YTHDF2 protein level, YTHDF2 function, YTHDF2 interactions with the CCR4-NOT complex or YTHDF2 interaction with m.sup.6A mRNA in the cell.

11. The method according to any of claims 1-10, wherein the YTHDF2 inhibitor inhibits YTHDF2 activity in the patient's leukaemic stem cells.

12. The method according to any of the previous claims wherein the YTHDF2 inhibitor is present within a pharmaceutical composition.

13. The method according to any of the previous claims wherein the YTHDF2 inhibitor is administered before, during or after chemotherapy.

14. An in vitro method for determining whether a YTHDF2 inhibitor test compound has potential as an agent to treat a haematological disorder, comprising determining the effect that the test compound has on the amount or function of YTHDF2 protein expressed in a cell that has been contacted with the test compound, wherein if the test compound causes a decrease in YTHDF2 protein level or function in the contacted cell then the test compound has potential as an agent to treat a haematopoietic disorder.

15. The in vitro method according to claim 14, wherein the effect that the test compound has on disrupting the interaction of (i) YTHDF2 with CCR4-NOT complex or (ii) YTHDF2 interaction with m.sup.6A is determined and wherein a test compound which disrupts at least one of said interactions is one that has potential as an agent to treat a haematopoietic disorder.

16. A method for selecting a compound for use in the treatment of a haematological disorder, comprising determining in an in vitro setting whether the test compound is capable of inhibiting YTHDF2 protein, wherein a test compound that inhibits YTHDF2 protein is selected for use in the treatment of a haematological disorder.

17. The method according to claim 16, wherein the selected test compound is then formulated with one or more pharmaceutically acceptable excipients.

18. The method according to claim 16 or 17, wherein the test compound is a small molecule compound, an RNAi or ASO molecule.

19. The method according to any of claims 16-18, wherein the haematological disorder is selected from the group consisting of: acute lymphocytic leukaemia (AML), chronic lymphocytic leukaemia (CLL), diffuse large B-cell lymphoma (DLBCL), T-cell acute lymphoblastic leukaemia (T-ALL) and B-cell acute lymphoblastic leukaemia (B-ALL).

20. A method for selecting a haematological cancer patient for treatment with an YTHDF2 inhibitor comprising, determining whether the patient's cancer cells (i) express elevated YTHDF2 mRNA transcript or protein levels relative to a control or normal cell, wherein if YTHDF2 mRNA transcript or protein levels in the individual's cells are elevated relative to a control or normal cell the individual is selected for treatment with an YTHDF2 inhibitor.

21. The method according to claim 20, wherein the patient's cancer cells are in a biological sample previously taken from the patient.

22. The method according to claim 21, wherein the biological sample comprises biological tissue or fluid.

23. The method according to claim 21 or 22, wherein the biological sample is selected from the group consisting of: cord blood, peripheral blood, bone marrow and tumour tissue.

24. The method according to any of claims 20-23, wherein the amount of YTHDF2 protein in the patient's cancer cells is determined using immunohistochemistry.

25. The method according to any of claims 20-23, wherein the amount of YTHDF2 mRNA transcript is determined.

26. The method according to claim 25, wherein the amount of transcript is determined using quantitative RT-PCR.

27. A haematopoietic stem cell characterised in that the cell has been treated to inactivate YTHDF2 gene or YTHDF2 function.

28. A haematopoietic stem cell according to claim 27, wherein the YTHDF2 gene is stably or transiently inactivated.

29. A method for maintaining long-term self-renewal capacity and multilineage differentiation potential in a HSC comprising treating a HSC to inhibit or inactivate YTHDF2 gene or protein.

30. A method for preparing an inactivated YTHDF2 HSC comprising transfecting a HSC with a nucleic acid molecule capable of inactivating the YTHDF2 gene in the HSC.

31. The method according to any of claims 27 to 30, wherein the YTHDF2 gene is transiently or stably inactivated.

32. The method according to claim 31, wherein the YTHDF2 gene is stably inactivated using Crispr-Cas9 gene editing.

33. The method according to claim 31, wherein the YTHDF2 gene is transiently inactivated using a nucleic acid inhibitor molecule capable of binding to YTHDF2 gene or YTHDF2 mRNA transcript.

34. The method according to any of claims 27 to 30, wherein the YTHDF2 protein function is inactivated using a compound or nucleic acid capable of binding to YTHDF2 protein.

35. A HSC produced by the method according to any of claims 27 to 34.

36. A pharmaceutical composition comprising the HSC of claim 27, 28 or 35.

37. A YTHDF2 inactivated HSC or a population of YTHDF2 inactivated HSCs, or a pharmaceutical composition comprising either, for use in HSCT.

38. A haematopoietic stem cell composition comprising one or more YTHDF2 inactivated stem cells dispersed within a medium.

39. A method for ex vivo expansion of multipotent hematopoietic cells, comprising culturing multipotent hematopoietic cells in a medium comprising an YTHDF2 inhibitor, wherein said inhibitor is present in an amount effective to produce a cell population to expand said multipotent hematopoietic cells.

40. The method according to claim 39, wherein after a suitable period of culturing, the population of cells in the culture medium is substantially enriched in a subpopulation of multipotent hematopoietic stem cells as compared to expansion of said multipotent hematopoietic cells in the absence of the inhibitor.

41. The method according to claim 39 or 40, wherein the multipotent hematopoietic cells to be expanded are obtained from cord blood, peripheral blood, or bone marrow.

42. The method according to any of claims 39-41, wherein the multipotent hematopoietic cells to be expanded are haematopoietic stem cells and primitive progenitor cells.

43. The method according to claim 42, wherein the hematopoietic stem cells and/or primitive progenitor cells are obtained by CD34 selection.

44. The method according to any of claims 39-43, wherein the multipotent hematopoietic cells are hematopoietic stem cells that retain the capacity for in vivo hematopoietic reconstitution upon transplantation.

45. The method according to any of claims 39-42, wherein growth of said hematopoietic stem cells in the presence of said YTHDF2 inhibitor results in at least a 3-fold, such as at least 5-fold or 10-fold, increase in CD34+CD90+ cells as compared to growth in the absence of said HDACI and IDM.

46. The method according to any of claims 39-42, wherein said hematopoietic cells are separated from other cells by selecting for cells for expression of at least one marker associated with stem cells or by physical separation means.

47. The method according to claim, wherein the marker is selected from the group consisting of Lineage (Lin), CD34, CD38, CD45RA, CD49f, and CD90.

48. A population of multipotent haematopoietic cells produced by the method according to any of claims to 39-47.

49. A YTHDF2 inactivated HSC or a pharmaceutical composition comprising said YTHDF2 inactivated HSC for use in therapy.

50. A YTHDF2 inactivated HSC or a pharmaceutical composition comprising said YTHDF2 inactivated HSC for use in haematopoietic stem cell transplantation.

51. A YTHDF2 inactivated HSC or a pharmaceutical composition comprising said YTHDF2 inactivated HSC for use in replenishing HSC or haematopoietic cells (HCs) in a patient in need thereof.

52. A YTHDF2 inactivated HSC or a pharmaceutical composition comprising said YTHDF2 inactivated HSC for the use according to any of claims 49 to 51, wherein the patient has undergone chemotherapy that has ablated HSC and/or haematopoietic cells.

53. A YTHDF2 inactivated HSC or a pharmaceutical composition comprising said YTHDF2 inactivated HSC for the use according to claim 52, wherein the patient has received chemotherapy.

54. A YTHDF2 inactivated HSC for the use according to any of claims 49-53, wherein the patient has been diagnosed with haematological cancer or is in need for HSCT.

Description

DESCRIPTION OF THE FIGURES

[0248] FIG. 1. YTHDF2 expression within the haematopoietic system.

[0249] (A) The sequences encoding the eGFP-PreScission-His6-Flag-HA2 epitope tag were inserted after the starting ATG codon in exon 1 of the Ythdf2 locus, and exon 2 was flanked by LoxP sites (see Ivanova et al., Mol Cell. 67(6): 1059-1067 e1054, 2017). The Ythdf2.sup.fl allele codes for eGFP-PreScission-His6-Flag-HA2-YTHDF2 fusion protein (referred to as GFP-YTHDF2 protein). (B) GFP expression in the indicated foetal liver (FL) cell populations from Ythdf2.sup.fl/fl 14.5 days post coitum (dpc) embryos. YTHDF2 is uniformly expressed in FL Lin.sup.−Sca-1.sup.+c-Kit.sup.+ (LSK) cells, LSKCD48.sup.−CD150.sup.+ HSCs, LSKCD48.sup.−CD150.sup.− multipotent progenitors (MPPs), primitive haematopoietic progenitor cells (i.e. LSKCD48.sup.+CD150.sup.− HPC-1 and LSKCD48.sup.+CD150.sup.+ HPC-2 populations), and Lin.sup.−Sca-1.sup.−c-Kit.sup.+ (LK) myeloid progenitors, and its expression is decreased in differentiated Lin.sup.+ cells. The data represent mean fluorescence intensity (MFI)±SEM (n=3-4 mice per genotype). (C) GFP expression in the BM cell populations from 8-12-week-old mice. YTHDF2 is expressed at higher levels in HSC/progenitor cells compared to the mature Lin.sup.+ cell compartment. The data represent MFI±SEM (n=3-4 mice per genotype).

[0250] FIG. 2. Ythdf2 deletion promotes FL HSC and primitive progenitor cell expansion. (A) Deletion of Ythdf2 from the haematopoietic system using Vav-iCre. FL cells from control 14.5 dpc Ythdf2.sup.fl/fl (Ythdf2.sup.CTL) embryos produce normal GFP-YTHDF2 protein. In Ythdf2.sup.fl/fl;Vav-iCre (Ythdf2.sup.cKO) mice, exon 2 is deleted resulting in a frameshift mutation and a complete loss of the GFP-YTHDF2 protein. (B) A representative histogram showing GFP expression in Ythdf2.sup.CTL LSK cells and the lack of GFP expression in Ythdf2.sup.cKO LSK cells. (C) Frequency of GFP-positive cells in 14.5 dpc FLs of Ythdf2.sup.CTL and Ythdf2.sup.cKO embryos. Data are mean±SEM (n=5 per genotype). Haematopoietic cells from Ythdf2.sup.CKO embryos/mice have no GFP-YTHDF2 protein expression. (D-E) Frequencies (D) and total numbers (E) of LSK, HSC, MPP, HPC-1 and HPC-2 cell populations in FLs from 14. 5dpc Ythdf2.sup.CTL and Ythdf2.sup.cKO embryos. Data are mean±SEM (control n=10 per genotype). **, P<0.01; ***, P<0.001 (Mann-Whitney U test). (F) Ythdf2.sup.fl/fl;Mx1-Cre (Ythdf2.sup.iCKO) and control Ythdf2.sup.fl/fl (Ythdf2.sup.CTL) mice were injected with plpC and analyzed 3 months after the last injection. (G) The graph shows the percentage of GFP-positive cells in BM of plpC-treated Ythdf2.sup.iCKO and Ythdf2.sup.CTL mice (n=10-12). (H) Total BM cellularity of plpC-treated Ythdf2.sup.iCKO and Ythdf2.sup.CTL mice. (I) Total cell numbers of BM monocytes, granulocytes and B cells. (J) Total cell numbers of BM LSK and LK cell populations.

[0251] FIG. 3. Ythdf2-deficient FL HSCs display enhanced capacity to reconstitute the HSC pool of the recipient mice upon transplantation. (A) 100 HSCs (CD45.2.sup.+LSKCD48.sup.−CD150.sup.+ cells) sorted from FLs of 14.5 dpc embryos were mixed with 200,000 support CD45.1.sup.+ BM cells and injected into lethally irradiated (11 Gy delivered in a split dose) CD45.1.sup.+/CD45.2.sup.+ recipient mice. n=6-9 recipients per genotype. 16 weeks after transplantation, stem/progenitor cells were harvested and re-transplanted into secondary recipient mice (n=8-9 recipients per genotype). (B) Percentage of donor-derived CD45.2.sup.+ cells in the peripheral blood (PB) of the recipient mice. Data are mean±SEM. (C) Percentage of CD45.2.sup.+ cells in the monocyte, granulocyte, B cell and T cell compartments in the PB of primary recipients. Data are mean±SEM. *, P<0.05 (Mann-Whitney U test). (D) Percentage of donor-derived CD45.2.sup.+ cells in the LSK, HSC, MPP, HPC-1, HPC2, LK, Lin.sup.− and Lin.sup.+ cell compartments in the BM of the primary recipient mice 16 weeks after transplantation. Data are mean±SEM. *, P<0.05; **, P<0.01 (Mann-Whitney U test). (E) Percentage (mean±SEM) of donor-derived CD45.2.sup.+ to differentiated cell compartments in the BM of the primary recipient mice 16 weeks after transplantation. (F) Percentage of donor-derived CD45.2.sup.+ cells in the LSK, HSC, MPP, HPC-1, HPC2, LK, Lin.sup.− and Lin.sup.+ cell compartments in the BM of the secondary recipient mice 16 weeks after transplantation. Data are mean±SEM. **, P<0.01 (Mann-Whitney U test). (G) Percentage (mean±SEM) of donor-derived CD45.2.sup.+ to differentiated cell compartments in the BM of the secondary recipient mice 16 weeks after transplantation. Data are mean±SEM. **, P<0.01 (Mann-Whitney U test).

[0252] FIG. 4. Ythdf2 deletion results in adult HSC expansion and enhanced HSC reconstitution potential. (A) Total numbers of T cells, B cells, granulocytes monocytes and erythroid cells in spleens from Ythdf2.sup.CTL and Ythdf2.sup.CKO mice. Data are mean±s.e.m. (n=5-6 mice per genotype). (B) CFU assays performed with BM cells from 8-10-wk-old mice. CFU-Red, CFU-erythroid and/or megakaryocyte; CFU-G, CFU-granulocyte; CFU-M, CFU-monocyte/macrophage; CFU-GM, CFU-granulocyte and monocyte/macrophage; CFU-Mix, at least three of the following: granulocyte, erythroid, monocyte/macrophage, and megakaryocyte. Data are mean±s.e.m. (n=4 per genotype). (C) Total number of BM LSK, HSC, MPP, HPC-1 and HPC-2 cell populations from Ythdf2.sup.CTL and Ythdf2.sup.CKO mice (n=6-7 mice per genotype). Mice were 8-10 weeks old. *, P<0.05; **, P<0.01; ****, P<0.0001 (D) Total number of BM LSK, HSC, MPP, HPC-1 and HPC-2 cell populations from Ythdf2.sup.CTL and Ythdf2.sup.CKO mice (n=6-7 mice per genotype) treated with 5-fluorouracil. Mice (8-10 weeks old) were treated with 3 doses of 5-fluorouracil administered every month and analysed a month after last dose. **, P<0.01. (E) 200 BM HSCs were transplanted into lethally irradiated 8-10-wk-old syngeneic CD45.1.sup.+/CD45.2.sup.+ recipient mice (n=6-9 recipients per genotype) together with 2×10.sup.5 CD45.1.sup.+ competitor BM cells. The graph shows the percentage of CD45.2.sup.+ cells in the LSK, HSC, MPP, HPC1-2, LK, Lin.sup.− and Lin.sup.+ cells compartments in the BM of recipient mice. Data are mean±s.e.m. **, P<0.01. (F) The graph shows the percentage of CD45.2.sup.+ cells in differentiated cell compartments in the BM of recipient mice. Data are mean±s.e.m. **, P<0.01.

[0253] FIG. 5. YTHDF2 gene expression is increased in different AML subtypes and correlates with LSC activity. (A) YTHDF2 gene expression in control (CTL) and different cytogenetic subgroups of human AML blood cell samples. Violin plots show the distribution of log.sub.2 expression values. Horizontal line in the boxplots indicates median. **, P<0.01; ****, P<0.0001. CNG, cytologically normal with good prognosis; CNI, cytologically normal with intermediate prognosis; CAO, cytologically abnormal not otherwise specified. (B) Western blot of YTHDF2 in normal human CD34.sup.+ cells and AML samples (karyotype details are shown in STAR Methods) is shown (left panel). α-Histone 3 (H3) was used as a loading control. Quantification of YTHDF2 normalized to H3 expression is presented (right panel). (C) YTHDF2 gene expression in primitive AML cell compartments with (LSC.sup.+) and without (LSC.sup.−) leukaemic engraftment potential. ****, P<0.0001. (D) YTHDF2 gene expression in CD34.sup.− and CD34.sup.+ AML cell compartments. FIG. 5 C-D were generated by reanalysis of the dataset published by Ng S., et al., Nature 540, 433-437, 2016.

[0254] FIG. 6. Ythdf2 inactivation compromises AML initiation and propagation. (A) Control Ythdf2.sup.fl/fl (Ythdf2.sup.CTL) and Ythdf2.sup.fl/fl;Vav-iCre (Ythdf2.sup.CKO) FL c-Kit.sup.+ cells were co-transduced with Meis1 and Hoxa9 retroviruses and serially re-plated. 100,000 c-Kit.sup.+ pre-leukaemic cells of both genotypes were transplanted into lethally irradiated recipient mice (n=12-14 per genotype) together with 200,000 BM support cells. (B) CFC counts at each re-plating. Data are mean±s.e.m., n=3 per genotype. *, P<0.05; **, P<0.01. (C) Percentage of CD45.2.sup.+ leukaemic cells in the peripheral blood of the recipient mice 20-60 days after transplantation. (n=12-14 per genotype). **, P<0.01. (D) Kaplan-Meier survival curve of the recipient mice transplanted with Meis1/Hoxa9-expressing pre-leukaemic cells. (n=12-14 per genotype), ****, P<0.001 Log-rank (Mantel-Cox test). (E) Percentage of GFP-positive cells in the CD45.2.sup.+ cell population from moribund recipients of Ythdf2.sup.CTL and Ythdf2.sup.CKO cells (n=5-6). (F) Ythdf2.sup.fl/fl (Ythdf2.sup.CTL) and Ythdf2.sup.fl/fl; Mx1-Cre (Ythdf2.sup.iCKO) FL c-Kit.sup.+ cells were co-transduced with Meis1 and Hoxa9 retroviruses, serially re-plated and transplanted into primary recipient mice, which were left to develop AML. GFP.sup.+ c-Kit.sup.+ CD45.2.sup.+ cells sorted from leukaemic primary recipients were re-transplanted into secondary recipient mice (n=14-16 mice per genotype). (G) Percentage of GFP-expressing cells as a measure of YTHDF2 expression in Ythdf2.sup.CTL and Ythdf2.sup.iCKO leukaemic cells prior to, and 3 weeks after transplantation. ***, P<0.001. (H) Kaplan-Meier survival curve of mice transplanted with Ythdf2.sup.CTL and Ythdf2.sup.iCKO leukaemic cells. (n=14-16 mice per genotype). ****, P<0.001 Log-rank (Mantel-Cox test).

[0255] FIG. 7. (A) Relative levels of YTHDF2 mRNA (normalised to ACTB) in human AML THP-1 cells transduced with lentiviruses expressing scrambled shRNA (CTL) and shRNA targeting YHTDF2 (KD). Data are mean±s.e.m., n=3. *, P<0.05. (B) Proliferation assays with THP-1 cells expressing CTL and KD shRNAs. Mean±s.e.m., n=3. **, P<0.01; ***, P<0.001. (C) Apoptosis assays. The graph depicts the percentage of Annexin V.sup.+DAPI.sup.− cells. Data are mean±s.e.m., n=3. ***, P<0.001. (D) Percentage of CD11b.sup.−CD14.sup.−, CD11b.sup.+CD14.sup.−, CD11b.sup.+CD14.sup.+ and CD11b.sup.−CD14.sup.+ cells in cultures shown in FIG. 7B-C. (E) Proliferation assays with NOMO-1 cells expressing CTL and KD shRNAs. Mean±s.e.m., n=3. **, P<0.01; ***, P<0.001. (F) Apoptosis assays. Data are mean±s.e.m., n=3. *, P<0.05. (G) NSG mice were injected with THP-1 cells transduced with CTL (n=4) or KD (n=4) lentiviruses and analysed one month later. Percentage of human CD45.sup.+CD33.sup.+ cells in the BM, liver, spleen and PB of the recipient mice is shown. (H) Survival curve of mice transplanted with 10,000 (n=6) and 1,000 (n=3) THP-1 cells. (I) Three independent human primary AML samples (AML1-3) were transduced with CTL, KD1 and KD2 lentiviruses. The graph shows AML-CFC frequencies after 7 days of culture (n=3 technical replicates per sample). (J) Representative colony images from FIG. 7I.

[0256] FIG. 8. YTHDF2 loss enhances stability of m.sup.6A-modified mRNA in AML. (A) Transcript expression scatter plot from Ythdf2.sup.CTL and Ythdf2.sup.CKO pre-leukemic cells (n=5). Significantly upregulated or downregulated transcripts are highlighted in red (P<0.05). (B) m.sup.6A peak FDR (−log.sub.10 Q) in Ythdf2.sup.CTL pre-leukemic cells for transcripts grouped according to expression changes between Ythdf2.sup.CTL and Ythdf2.sup.CKO pre-leukemic cells is shown (down, genes significantly downregulated in Ythdf2.sup.CKO (P<0.05); unchanged, genes not significantly changing in Ythdf2.sup.CKO; up, genes significantly upregulated in Ythdf2.sup.CKO (P<0.05). The upper and lower quartiles and the median are shown for each group. (C) Violin plots showing expression change between Ythdf2.sup.CTL and Ythdf2.sup.CKO pre-leukemic cells for not methylated (no m.sup.6A), methylated (m.sup.6A, −log.sub.10 Q≤25) and highly methylated (m.sup.6A high, −log.sub.10 Q>25) transcripts. The upper and lower quartiles and the median are indicated for each group. (D) Cumulative distributions of transcripts' expression change in Ythdf2.sup.CTL and Ythdf2.sup.CKO pre-leukemic cells for not methylated, methylated and highly methylated transcripts as in C. (E) Mode decay curves for Ythdf2.sup.CTL (black) and Ythdf2.sup.CKO (red) pre-leukemic cell transcriptomes are shown. The shaded areas indicate the first and third quantile decay curves range for each genotype. Transcripts' half-life modes for each genotype are indicated with horizontal dotted lines and are also shown at the panel top. (F) Cumulative distributions of transcripts' half-life in Ythdf2.sup.CTL (left panel) and Ythdf2.sup.CKO (right panel) pre-leukemic cells are shown for not methylated, methylated and highly methylated transcripts as in C. The half-life change significance between the methylated and not methylated transcripts are indicated. (G) Cumulative distributions of relative stability change between Ythdf2.sup.CTL and Ythdf2.sup.CKO pre-leukemic cells are shown for not methylated, methylated and highly methylated transcripts as in C. The relative stability change significances between the methylated and not methylated transcripts are indicated. (H) CPDB analysis of genes significantly upregulated in Ythdf2.sup.CKO pre-leukemic cells (P<0.05) with high m.sup.6A levels (−log.sub.10 Q>25) in mouse pre-leukemic cells and also methylated in human AML cell lines. (I) GSEA using LSC signature gene set for genes that negatively correlate with YTHDF2 expression in human AML samples. (J) m.sup.6A IP read coverage (blue) from Ythdf2.sup.CTL pre-leukemic cells along the Trnfrs1b genomic locus (upper panel) and m.sup.6A IP read coverage from NOMO-1, and MA9.3ITD cells along TNFRSF1B genomic locus (lower panels) are shown. Input coverage is shown in green. (K) Tnfrsf1b enrichment in YTHDF2 immuno-precipitates from Ythdf2.sup.CTL pre-leukemic cells is shown. Tnfrsf1b background levels were determined using Ythdf2.sup.CKO pre-leukemic cells (n=3). (L) Decay curves for Trnfrs1b in Ythdf2.sup.CTL (top panel) and Ythdf2.sup.CKO (bottom panel) pre-leukemic cells transcriptomes are shown. The centre value and the error bars at each time point indicate the conversion rate mean and standard deviation, respectively. The conversion rates for each biological replicate are indicated with dots. The Trnfrs1b half-life is also shown. (M) FACS plots showing the expression of TNFR2 on the cell surface of Ythdf2.sup.CTL and Ythdf2.sup.CKO pre-leukemic cells. The inner graph displays the quantification of TNFR2 expression (n=4). (N) Percentage of Annexin V.sup.+DAPI.sup.− pre-leukemic cells treated with TNF-α at 0 and 6-hour timepoints (n=3). *, P<0.05; **, P<0.01

EXAMPLES

Example 1. The Expression of YTHDF2 at Different Levels of Haematopoietic Hierarchy

[0257] To determine YTHDF2 expression in the haematopoietic system, we employed mice in which eGFP-PreScission-His6-Flag-HA2 tag is inserted after the start codon of YTHDF2 (creating a functional fusion protein), and additionally exon 2 of Ythdf2 is flanked by LoxP sites (FIG. 1A) (Ivanova et al., Mol Cell. 67(6): 1059-1067 e1054, 2017). All the details regarding these mice are described in Ivanova et al. (ibid). In these mice, eGFP fluorescence reports endogenous YTHDF2 protein levels, which can be detects by flow cytometry.

[0258] We produced adult mice (by inter-crossing the above mice) and 14.5 days postcoitum (dpc) embryos and obtained bone marrow cells (BM) and foetal liver (FL; the major site of definitive haematopoiesis during development) cells using standard protocols as previously described (Mortensen et al., J Exp Med. 208(3): 455-467, 2011; Vukovic et al., J Exp Med. 212(13):2223-2234, 2015; Vukovic et al., Blood. 127(23): 2841-2846, 2016; Guitart et al., J Exp Med. 214(3): 719-735, 2017).

[0259] Using staining procedures (Mortensen et al., J Exp Med. 208(3): 455-467, 2011, Vukovic et al., Blood. 127(23): 2841-2846, 2016) with defined antibodies recognising cell surface antigens (Lineage markers, Sca-1, c-Kit.sup.+, CD150 and CD48), we performed FACS acquisitions of BM and FL Lin.sup.−Sca-1.sup.+c-Kit.sup.+ (LSK) cells, LSKCD48.sup.−CD150.sup.+ HSCs, LSKCD48.sup.−CD150.sup.− multipotent progenitors (MPPs), primitive haematopoietic progenitor cells (i.e. LSKCD48.sup.+CD150.sup.− HPC-1 and LSKCD48.sup.+CD150.sup.+ HPC-2 populations), and Lin.sup.−Sca-1.sup.−c-Kit.sup.+ (LK) myeloid progenitors, and Lin.sup.+ cells, and determined GFP expression in these cells using flow cytometry (FACS).

[0260] The data for each population were calculated as mean fluorescence intensity of GFP expression in each population using FlowJo software (FlowJo Llc).

[0261] All hematopoietic cells in FL and adult BM expressed GFP-YTHDF2 (FIGS. 1B and 1C). Notably, YTHDF2 was highly expressed in Lin.sup.−Sca-1.sup.+c-Kit.sup.+ (LSK) cells, LSKCD48.sup.−CD150.sup.+HSCs, LSKCD48.sup.−CD150.sup.− multipotent progenitors (MPPs), primitive hematopoietic progenitor cells (i.e. LSKCD48.sup.+CD150.sup.− HPC-1 and LSKCD48.sup.+CD150.sup.+ HPC-2 populations), and Lin.sup.−Sca-1.sup.−c-Kit.sup.+ (LK) myeloid progenitors, and its expression was decreased in differentiated Lin.sup.+ cells (FIGS. 1B and 1C).

[0262] These results indicated that YTHDF2 protein is highly expressed in stem and progenitor cells and its expression is decreased in more mature haematopoietic cells.

Example 2. The Impact of Ythdf2 Deletion on Numbers of Foetal Liver (FL) and Adult Stem and/or Primitive Progenitor Cells

[0263] To reveal the role of YTHDF2 in HSC biology and multilineage haematopoiesis, we conditionally deleted Ythdf2 specifically from the haematopoietic system shortly after the emergence of definitive HSCs using Vav-iCre (de Boer et al., Eur J Immunol. 33(2): 314-325, 2003). To achieve this, we mated adult Ythdf2.sup.fl/fl mice as described in Ivanova et al., (ibid) with adult Vav-iCre mice as described in de Boer et al., (ibid).

[0264] We next performed timed matings (as described in Bamforth et al., Nat Genet. 29(4): 469-474, 2001, Kranc et al., Mol Cell Biol. 23(21): 7658-7666, 2003, Guitart et al., J Exp Med. 214(3): 719-735, 2017)) to generate Ythdf2.sup.fl/fl;Vav-iCre (referred to as Ythdf2.sup.cKO or Ythdf2.sup.CKO) and control (referred to as Ythdf2.sup.CTL) embryos at 14.5 days dpc. FACS analyses of FL cells indicated that Ythdf2.sup.cKO cells from FLs did not express the YTHDF2 protein indicating that Ythdf2.sup.cKO cells have lost YTHDF2 expression (FIG. 2A-C).

[0265] We next determined the frequencies and absolute numbers of HSCs and primitive progenitors in FLs using FACS analyses as previously described (Guitart et al., J Exp Med. 214(3): 719-735, 2017). We found that FLs of Ythdf2.sup.cKO embryos displayed significantly increased frequencies and absolute numbers of HSCs and primitive progenitors compared to control FLs (FIG. 2D-E). Therefore, loss of YTHDF2 expression results in enhanced expansion of HSCs and primitive cells within FL.

[0266] To rule out compensatory mechanisms resulting from Vav-iCre-mediated-recombination during embryonic hematopoiesis, we employed Mx1-Cre, which upon plpC injection, acutely deletes Ythdf2 in Ythdf2.sup.iCKO adult mice (FIG. 2F). Acute Ythdf2 deletion (FIG. 2G) had no impact on mouse survival (data not shown) or multilineage hematopoiesis (FIG. 2H-l)) and resulted in increased numbers of BM LSK cells but not LK myeloid progenitor cells (FIG. 2J). Thus, Ythdf2 hematopoietic-specific ablation during development or acute deletion in the adult mouse leads to an expansion of the primitive cell compartment at the top of the hematopoietic differentiation hierarchy and does not derail normal hematopoiesis.

Example 3. Transplantation of Ythdf2-Deficient HSCs into Primary Recipient Mice

[0267] We employed long-term competitive transplantation assays (FIG. 3A) to determine the reconstitution and self-renewal capacity of FL Ythdf2-deficient and control HSCs. To achieve this, we used standard transplantation protocols as previously described (Guitart et al., J Exp Med. 214(3): 719-735, 2017).

[0268] Lethal irradiation of CD45.1.sup.+/CD45.2.sup.+ recipient mice was achieved using a split dose of 11 Gy (two doses of 5.5 Gy administered at least 4 h apart) at a mean rate of 0.58 Gy/min using a Cesium 137 irradiator (GammaCell 40; Best Theratronics).

[0269] For primary transplantations, 100 HSCs (LSKCD48.sup.−CD150.sup.+CD45.2.sup.+) sorted from FLs of 14.5-dpc embryos or 200,000 unfractionated FL cells were mixed with 200,000 support CD45.1.sup.+ BM cells and injected into lethally irradiated (11 Gy delivered in a split dose) CD45.1.sup.+/CD45.2.sup.+ recipient mice.

[0270] Peripheral blood analyses (described previously (Guitart et al., J Exp Med. 214(3): 719-735, 2017)) indicated that Ythdf2-deficient HSCs gave equal overall long-term reconstitution compared to control HSCs for at least 16 weeks after transplantation (FIGS. 3B and 3C). However, while Ythdf2-deficient HSCs had slightly enhanced myeloid lineage reconstitution capacity compared to control HSCs, they had normal B-cell and slightly compromised T-cell reconstitution potentials (FIGS. 3B and 3C).

[0271] Notably, FACS analyses of BM of recipient mice 16 weeks after transplantation (which employed previously described approaches (Mortensen et al., J Exp Med. 208(3): 455-467, 2011, Guitart et al., Blood. 122(10): 1741-1745, 2013, Vukovic et al., Blood. 127(23): 2841-2846, 2016, Guitart et al., J Exp Med. 214(3): 719-735, 2017)) demonstrated that HSCs lacking Ythdf2 displayed enhanced contribution to the HSC/primitive progenitor cell compartments of the recipient mice (FIG. 3D) and reconstituted normal BM haematopoiesis (FIG. 3E).

Example 4. Transplantation of Ythdf2-Deficient FL HSCs into Secondary Recipient Mice

[0272] To test the self-renewal capacity of Ythdf2-deficient HSCs, we culled the primary recipients, isolated BM cells and sorted stem and progenitor cells from the primary recipient mice. 2,000 CD45.2.sup.+ LSK cells sorted from BM of primary recipients were mixed with 200,000 support CD45.1.sup.+ wild-type BM cells and re-transplanted into secondary recipients (Mortensen et al., J Exp Med. 208(3): 455-467, 2011, Guitart et al., Blood. 122(10): 1741-1745, 2013, Vukovic et al., Blood. 127(23): 2841-2846, 2016, Guitart et al., J Exp Med. 214(3): 719-735, 2017).

[0273] FACS analyses of BM of the recipient mice (16 weeks after transplantation) revealed that Ythdf2-deficient HSCs had increased ability to reconstitute all haematopoietic compartments of the secondary recipient mice, compared to control HSCs (FIG. 3F-G), Therefore, Ythdf2 deletion promotes HSC and primitive progenitor cell expansion and enhances their reconstitution capacity.

Example 5. Analyses of Adult Mice Lacking Ythdf2 Specifically from the Haematopoietic System

[0274] We next investigated the impact of Ythdf2 deletion on adult haematopoiesis and HSC functions. We mated Ythdf2.sup.fl/fl mice to Ythdf2.sup.fl/fl;Vav-iCre mice to generate Ythdf2.sup.CKO and control mice. Ythdf2.sup.CKO mice were born at normal Mendelian ratios, matured to adulthood without any obvious defects (as defined by their survival, appearance and daily monitoring of their behaviour). Furthermore, FACS analyses (using protocols we previously described (Kranc et al., Cell Stem Cell. 5(6): 659-665, 2009, Vukovic et al., Blood. 127(23): 2841-2846, 2016, Guitart et al., J Exp Med. 214(3): 719-735, 2017) revealed that they had normal multilineage haematopoiesis (FIG. 4A).

[0275] Colony-forming cell (CFC) assays in GM 3434 medium (from Stemcell Technologies), as described previously (Kranc et al., Cell Stem Cell. 5(6): 659-665, 2009) indicated normal differentiation potential of BM cells lacking Ythdf2 (FIG. 4B). Thus, despite the modest changes in the peripheral blood upon Ythdf2 deletion, we conclude that YTHDF2 is not critical for steady-state hematopoiesis.

[0276] Notably, FACS analyses (using our standard approaches (Kranc et al., Cell Stem Cell. 5(6): 659-665, 2009, Guitart et al., Blood. 122(10): 1741-1745, 2013, Vukovic et al., Blood. 127(23): 2841-2846, 2016, Guitart et al., J Exp Med. 214(3): 719-735, 2017)) of adult Ythdf2.sup.CKO mice indicated that they displayed expansion of BM HSCs and primitive (HPC-1 and HPC-2) progenitor cells compared to Ythdf2.sup.CTL mice (FIG. 4C).

Example 6. Analyses of HSCs and Progenitor Cells in Adult Ythdf2-Deficient Mice Treated with 5-Fluorouracil

[0277] We next examined the consequences of Ythdf2 deletion on HSC responses to haematopoietic injury, by subjecting Ythdf2.sup.CKO and Ythdf2.sup.CTL mice to serial injections of the myelosuppressive agent 5-fluorouracil (5-FU), which depletes cycling haematopoietic cells and forces HSCs to proliferate and self-renew.

[0278] 8-10 weeks old Ythdf2.sup.CKO and Ythdf2.sup.CTL mice received 3 sequential doses of 5-FU (150 mg/kg; 25-30 days apart) and were analysed 30 days after the last 5-FU administration. BM analyses (carried out as described previously (Guitart et al., Blood. 122(10): 1741-1745, 2013, Vukovic et al., Blood. 127(23): 2841-2846, 2016)) indicated that 5-FU-treated Ythdf2.sup.CKO mice displayed strikingly increased numbers of HSCs and primitive progenitors compared to control mice (FIG. 4D).

Example 7. Analyses of Reconstitution Capacity of Adult Ythdf2-Deficient HSCs

[0279] To reveal the repopulation capacity of Ythdf2-deficient HSCs, we transplanted 200 CD45.2.sup.+ HSCs sorted from BM of Ythdf2.sup.CKO and Ythdf2.sup.CTL mice (as we described previously (Guitart et al., Blood. 122(10): 1741-1745, 2013, Vukovic et al., Blood. 127(23): 2841-2846, 2016, Guitart et al., J Exp Med. 214(3): 719-735, 2017)) into lethally irradiated syngeneic CD45.1.sup.+/CD45.2.sup.+ recipients (together with competitor BM cells obtained from CD45.1.sup.+ mice), using our standard transplantation protocols as descried in (Guitart et al., Blood. 122(10): 1741-1745, 2013, Vukovic et al., Blood. 127(23): 2841-2846, 2016, Guitart et al., J Exp Med. 214(3): 719-735, 2017). FACS analyse revealed that Ythdf2-deficient HSCs displayed significantly increased capacity to contribute to the HSC, MPP, HPC-1 and HPC-2 compartments (FIG. 4E) and differentiated cell compartments (FIG. 4F) of the recipient mice compared to control HSCs, indicating that Ythdf2-deficient HSCs outcompeted control HSCs and displayed enhanced reconstitution capacity. Therefore, targeting Ythdf2 promotes HSC expansion, and enhances their reconstitution and regenerative capacity.

[0280] These data taken together suggest that YTHDF2 can be a therapeutic target for ex vivo HSC expansion or enhancing regenerative capacity of HSCs upon chemotherapy or transplantation.

Example 8. The Role of YTHDF2 in AML

[0281] To determine the expression of YTHDF2 in different AML samples, we used the following publicly available datasets:

TABLE-US-00002 Dataset # samples Reference GSE52891 18 (Bachas et al., PLoS One. 10(4): e0121730, 2015) GSE61804 160 (Metzelder et al., Leukaemia. 29(7): 1470-1477, 2015) GSE12417 73 (Metzeler et al., Blood. 112(10): 4193-4201, 2008) GSE13159 576 (Haferlach et al., J Clin Oncol. 28(15): 2529-2537, 2010) GSE15061 505 (Mills et al., Blood. 114(5): 1063- 1072, 2009) GSE15434 54 (Klein et al., BMC Bioinformatics. 10(422, 2009) GSE16015 65 (Haferlach et al., Blood. 114(14): 3024-3032, 2009) GSE19577 40 (Pigazzi et al., Leukaemia. 25(3): 560-563, 2011) GSE22845 45 (Taskesen et al., Blood. 117(8): 2469-2475, 2011) GSE10358 99 (Tomasson et al., Blood. 111(9): 4797-4808, 2008)

[0282] Affymetrix data were downloaded as raw CEL files from the GEO database. Only bone marrow samples, with a total of 1732 samples (11 datasets), were used for analysis. The Simpleaffy package from Bioconductor was used to extract quality measurement of microarrays. RNA degradation was assessed through GAPDH and ACTB housekeeping genes. Samples with NUSE<1.05 and relative log expression (RLE)<0.15 were excluded from further analysis. Finally, the retained samples were assessed for their homogeneity using the Bioconductor array QualityMetrics package. Low quality RNA and outlier samples were excluded from further analysis. High quality samples retained after quality control were background corrected and normalized using RMAexpress software (http://rmaexpress.bmbolstad.com/). Pairwise comparisons between each karyotype and control were performed using student t-test. Adjustment for multiple comparisons was done using Benjamini and Hochberg. ANOVA test was used for multiclass comparisons. Figures were created using R/Bioconductor package “ggplot2”.

[0283] The analyses of these global gene expression datasets of different AML samples revealed that the expression of YTHDF2 was significantly higher across the AML samples with diverse cytogenetic abnormalities compared to non-leukaemic controls (FIG. 5A). This was corroborated by western blotting, which demonstrated that YTHDF2 is highly expressed in patient-derived AML samples (FIG. 5B).

[0284] For western blotting shown in FIG. 5B, the following samples were used: 70 (karyotype 46,XY,del(7)(q22q32)[20]), 104 (karyotype 46,XX,t(6;9;11)(p2?1;p22;q23)[6]/45,idem,der(15)t(15;17)(p11.2;q11.2),−17[4] [variant of t(9;11)]0, 108 (karyotype 46,XX,t(6;11)(q27;q23)[10]), 149 (karyotype 46,XX,t(15;17)(q22;q11.2)[7]/46,sl,−6,add(16)(q12),+mar[3]/46,XX[3]), 163 (karyotype 45,X,−Y,t(8;21)(q22;q22)[8]/46,XY[2]), 191 (karyotype 46,XX [20]), 205 (karyotype 44,XX,add(3)(p25),−5,−7[12]), 419 (karyotype 46,XX,t(1;22)(p21;p11.2),ins(10;11)(p12;q23q1?4)[10] nb variant of t(10;11) MLL-MLLT10 fusion), 539 (karyotype 46,XY [20]), 685 (karyotype 46,XX,t(6;9)).

[0285] Using the analyses described above, we next compared YTHDF2 expression in publicly available datasets (Ng et al., Nature. 540(7633): 433-437, 2016) from AML primitive cell compartments which gave rise to leukaemia upon xenotransplantation and the equivalent AML cell compartments which failed to initiate leukaemia. These in silico analyses showed that LSC activity correlated with increased YTHDF2 expression (FIG. 5C). Thus, YTHDF2 expression is increased in LSCs compared to cells which do not possess the LSC activity.

[0286] Furthermore, given that the majority of CD34.sup.+ and a minority of CD34.sup.− fractions have LSC activity (Eppert et al., Nat Med 17:1086-1093, 2011; Ng et al., Nature. 540(7633): 433-437, 2016; Sarry et al., J Clin Invest 121:384-395, 2011), we also compared YTHDF2 expression between these fractions and found that YTHDF2 was expressed at higher levels in CD34.sup.+ fractions of AML samples (FIG. 5D).

[0287] These data indicate that YTHDF2 expression level can be a biomarker for diseases that can be treated with a YTHDF2 inhibitor.

Example 9. The Role of YTHDF2 in Leukaemic Transformation In Vitro

[0288] To investigate the functional requirement for YTHDF2 in leukemogenesis we employed conditional genetics and a mouse model of AML in which Meis1 and Hoxa9, oncogenes frequently overexpressed in human AML subtypes (Lawrence et al., Leukaemia. 13(12): 1993-1999, 1999, Drabkin et al., Leukaemia. 16(2): 186-195, 2002), drive the development and maintenance of LSCs. We have previously described this model and the protocol in detail (Vukovic et al., J Exp Med. 212(13):2223-2234, 2015, Guitart et al., J Exp Med. 214(3): 719-735, 2017).

[0289] In this model (FIG. 6A), HSPCs (c-Kit.sup.+ cells) were sorted from FLs of 14.5-dpc embryos after c-Kit (CD117) enrichment using magnetic-activated cell-sorting columns (Miltenyi Biotec). 10,000 cells were simultaneously transduced with mouse stem cell virus (MSCV)-Meis1a-puro and MSCV-Hoxa9-neo retroviruses and subsequently subjected to three rounds of CFC assays in MethoCult (M3231) supplemented with 20 ng/ml stem cell factor, 10 ng/ml IL-3, 10 ng/ml IL-6, and 10 ng/ml granulocyte/macrophage stem cell factor.

[0290] Colonies were counted 6-7 d after plating, and 2,500 cells were replated. Serial replating generates pre-leukaemic cells, which upon transplantation to recipient mice give rise to self-renewing LSCs causing AML (Kroon et al., EMBO J. 17(13): 3714-3725, 1998, Wang et al., Science. 327(5973): 1650-1653, 2010, Velasco-Hernandez et al., Blood. 2014, Vukovic et al., J Exp Med. 212(13):2223-2234, 2015, Guitart et al., J Exp Med. 214(3): 719-735, 2017).

[0291] We mated Ythdf2.sup.fl/fl mice (as described by Ivanova et al., ibid; in which GFP is inserted after the start codon of Ythdf2 and exon 2 of Ythdf2 is flanked by LoxP sites) with the Vav-iCre deleter (de Boer et al., Eur J Immunol. 33(2): 314-325, 2003) which recombines specifically within the haematopoietic system. Timed matings of Ythdf2.sup.fl/fl with Ythdf2.sup.fl/fl;Vav-iCre Ythdf2.sup.CKO mice gave rise to Ythdf2.sup.fl/fl;Vav-iCre (Ythdf2.sup.CKO) and control Ythdf2.sup.fl/fl or (Ythdf2.sup.CTL) embryos. We observed at weaning normal Mendelian distribution of Ythdf2.sup.CTL and Ythdf2.sup.CKO mice (i.e. Ythdf2.sup.fl/fl X Ythdf2.sup.fl/fl;Vav-iCre matings resulted in 65 Ythdf2.sup.CTL and 47 Ythdf2.sup.CKO mice at weaning, P=0.28) and found no increased mortality of Ythdf2.sup.CKO mice thereafter.

[0292] Fetal liver (FL) HSPC were prepared from these embryos as previously described (Guitart et al., J Exp Med. 214(3): 719-735, 2017). We subsequently transduced FL HSPCs with Meis1/Hoxa9 retroviruses (FIG. 6A), as per protocol described above. While Ythdf2.sup.CKO cells produced significantly lower numbers of colonies upon serial re-plating (FIG. 6B).

[0293] Thus, we found that Ythdf2.sup.CKO cells produced significantly lower numbers of colonies upon serial re-plating (FIG. 6B). These results indicated that Ythdf2.sup.CKO have compromised ability to undergo leukaemic transformation.

Example 10. The Role of YTHDF2 in AML Initiation

[0294] To test the ability of Ythdf2.sup.CKO and Ythdf2.sup.CTL pre-leukaemic cells (i.e. cells produced in 3 rounds of serial re-plating, as previously described (Vukovic et al., J Exp Med. 212(13):2223-2234, 2015; Guitart et al., J Exp Med. 214(3): 719-735, 2017), to generate AML upon transplantation (FIG. 6A), we transplanted them into recipient mice as we described previously (Vukovic et al., 2015, ibid; Guitart et al., 2017 ibid). The monitoring and analyses of these mice (according to standard protocols described in (Vukovic et al., 2015, ibid; Guitart et al., 2017 ibid), demonstrated that Ythdf2-deficient pre-leukaemic cells generated AML with substantially longer disease latency compared to control cells (FIG. 6C-D). The loss of YTHDF2 expression was confirmed in Ythdf2.sup.CKO cells isolated from moribund recipient mice based of the lack of the GFP expression (FIG. 6E), excluding the possibility that the disease with the delayed latency is caused by the accumulation of cells which escaped Ythdf2 deletion. Therefore, Ythdf2 is required for LSC development and disease initiation in AML driven by Meis1/Hoxa9.

Example 11. The Role of YTHDF2 in AML Propagation

[0295] We next asked whether acute deletion of Ythdf2 from established LSCs using Mx1-Cre impacts on LSC maintenance and leukaemia propagation. To achieve this, we crossed adult Ythdf2.sup.fl/fl mice (described in Ivanova et al., ibid) with Mx1-Cre mice (described in (Kuhn et al., Science. 269(5229): 1427-1429, 1995)) in which Cre expression is induced by interferon. By performing timed matings as described (Guitart et al., J Exp Med. 214(3): 719-735, 2017), we generated 14.5 dpc Ythdf2.sup.fl/fl; Mx1-Cre (Ythdf2.sup.iKO) and control Ythdf2.sup.fl/fl (Ythdf2.sup.CTL) embryos and extracted FL HSPCs from them using c-Kit enrichment (by employing magnetic-activated cell-sorting columns (Miltenyi Biotec)).

[0296] Using the Meis1/Hoxa9 mouse AML model described above (Examples 9 and 10), we next transduced HSPCs with Meis1/Hoxa9 retroviruses and transplanted them into lethally irradiated primary recipient mice (FIG. 6F). Upon leukaemia development (determined by the presence of leukaemic cells in peripheral blood using FACS) the mice were culled and BM cell suspensions were generated as we previously described (Kranc et al., Cell Stem Cell. 5(6): 659-665, 2009; Guitart et al., Blood. 122(10): 1741-1745, 2013; Vukovic et al., J Exp Med. 212(13):2223-2234, 2015; Vukovic et al., Blood. 127(23): 2841-2846, 2016; Guitart et al., J Exp Med. 214(3): 719-735, 2017).

[0297] c-Kit.sup.+ cells (a population enriched for LSCs) were isolated from BM using magnetic-activated cell-sorting columns (Miltenyi Biotec). Given the leakiness of Mx1-Cre upon transplantation (i.e. Cre can be induced even without interferon administration) (Velasco-Hernandez et al., Stem Cell Reports. 7(1): 11-18, 2016), the population was further sorted for GFP.sup.+ cells to enrich for those expressing GFP-YTHDF2 protein (FIG. 6G, before secondary transplantation). GFP.sup.+ cell sorting was performed on a FACSAria Fusion cell sorter (BD). 100,00 of c-Kit.sup.+ cells GFP.sup.+ were re-transplanted into secondary recipients using protocols described previously (Vukovic et al., J Exp Med. 212(13):2223-2234, 2015; Guitart et al., J Exp Med. 214(3): 719-735, 2017).

[0298] While Ythdf2.sup.CTL c-Kit.sup.+ GFP.sup.+ cells caused aggressive AML upon transplantation to secondary recipients (FIG. 6H), Ythdf2.sup.iKO c-Kit.sup.+GFP.sup.+ cells lost YTHDF2 expression (FIG. 6G, after secondary transplantation) due to spontaneous Mx1-Cre activation upon LSC transplantation (even without the administration of polyinosinic-polycytidylic acid; plpC) (Velasco-Hernandez et al., Stem Cell Reports. 7(1): 11-18, 2016), and failed to efficiently propagate the disease (FIG. 6H). Thus, YTHDF2 is required for disease propagation in Meis1/Hoxa9 model of AML.

Example 12. The Role of YTHDF2 in Human AML Cell Line with MLL Translocation

[0299] To investigate the requirement for YTHDF2 in human established leukaemic cells, we knocked down the expression of YTHDF2 in human AML THP-1 cells harbouring MLL-AF9 translocation using shRNA approaches as we previously described (Karvela et al., Autophagy. 12(6): 936-948, 2016, Sinclair et al., Blood. 128(3): 371-383, 2016, Guitart et al., J Exp Med. 214(3): 719-735, 2017). The sequences of shRNA targeting human YTHDF2 were:

TABLE-US-00003 (KD1) (SEQ ID NO: 3)  CCGGTACTGATTAAGTCAGGATTAACTCGAGTTAATCCTGACTTAATCAG TATTTTTG (KD2) (SEQ ID NO: 4)  CCGGCGGTCCATTAATAACTATAACCTCGAGGTTATAGTTATTAATGGAC CGTTTTTG

[0300] THP-1 cells were cultured at 400,000 cells/ml in RPMI-1640 GlutaMAX containing 10% FBS, 25 mM HEPES, 100 U/ml penicillin, and 100 μg/ml streptomycin. Lentiviruses were prepared and THP-1 cells were transduced as we previously described (Kranc et al., Cell Stem Cell. 5(6): 659-665, 2009; Vukovic et al., J Exp Med. 212(13):2223-2234, 2015). Transduced THP-1 cells were grown in the presence of 5 μg/ml puromycin. Knockdown efficiency (e.g. YTHDF2 gene expression) was determined using RT-quantitative PCR as we previously described (Kranc et al., Cell Stem Cell. 5(6): 659-665, 2009, Sinclair et al., Blood. 128(3): 371-383, 2016, Guitart et al., J Exp Med. 214(3): 719-735, 2017).

[0301] Efficient YTHDF2 knockdown was validated by Q-PCR and western blotting (FIG. 7A), as previously described (Guitart et al., J Exp Med. 214(3): 719-735, 2017). We next performed cell proliferation assays by measuring cell numbers by flow cytometry using the Accuri C6 Flow Cytometer (BD) as described (Vukovic et al., J Exp Med. 212(13):2223-2234, 2015). Apoptosis of THP-1 cells with YTHDF2 knockdown was determined by staining with Annexin V-FITC and DAPI followed by FACS analysis. Percentage of Annexin V.sup.+DAPI.sup.− cells (% of apoptotic cells) was calculated using FlowJo software. Notably, we found that YTHDF2 knockdown (FIG. 7A) inhibited THP-1 cell growth and increased their apoptosis (FIG. 7B-C) but had no impact on their myeloid differentiation status (FIG. 7D). This finding was corroborated in NOMO-1 AML cells harbouring MLL-AF9 translocation (FIG. 7E-F). Furthermore, THP-1 cells with YTHDF2 knockdown had decreased capacity to engraft AML compared to those with control shRNA (FIG. 7G) and displayed impaired ability to cause fatal AML upon xenotransplantation (FIG. 7H). Finally, we performed knockdown experiments in independent human primary AML samples and found that YTHDF2 depletion significantly decreased the clonogenic potential of AML cells in CFC assays (FIG. 7I-J). In summary, YTHDF2 is necessary for human AML cell survival and leukaemic cell engraftment.

[0302] For CFC assays shown in FIG. 7I-J, the following samples were used: 160 (AML1) (karyotype 46,XX,t(9;11)(p22;q23),der(21;22)(q10;q10),+der(21;22)[cp10]; MLL-MLLT3 rearrangement; clonal evolution with add(Xp); add(4q); add(7q); +21 at relapse), 292 (AML2) (karyotype 46,XX,t(15;17); PML-RARA rearrangement [no cyto report available]), 251 (AML3) (karyotype 46,XY,t(6;9)(p22;q34)[9]/46,XY,der(6)t(6;9),der(9)t(6;9)del(9)(q21q34)[2]).

[0303] Therefore, YTHDF2 is required for human AML cell survival.

[0304] Taken together, these data indicate that the m.sup.6A reader YTHDF2 is required for development and maintenance of LSCs in a mouse model of AML and is necessary for survival of human established AML cells. Our data indicate the m.sup.6A reader YTHDF2 as a critical mediator of AML whose inhibition selectively compromises leukemogenesis.

[0305] This suggests that YTHDF2 can be a therapeutic target in AML.

Example 13. YTHDF2 Decreases m.SUP.6.A RNA Stability in AML

[0306] We next sought to understand the mechanism by which YTHDF2 loss impedes LSC function. YTHDF2 is known to promote transcript decay through deadenylation (Du et al. Nat Commun 7:12626, 2016; Wang et al., Nature 505:117-120, 2014). We performed gene expression analysis (as described by Ivanova et al., ibid) in Ythdf2.sup.CKO and Ythdf2.sup.CTL pre-leukemic cells. The loss of YTHDF2 resulted in deregulated gene expression with 754 upregulated and 582 downregulated genes, P<0.05) in Ythdf2.sup.CKO compared to Ythdf2.sup.CTL pre-leukemic cells (FIG. 8A). To understand which of the deregulated transcripts could be direct targets of YTHDF2, we determined transcriptome-wide mRNA m.sup.6A in Ythdf2.sup.CTL and Ythdf2.sup.CKO pre-leukemic cells (as previously described (Lin et al., Mol Cell 62, 335-345, 2016)). This revealed the expected m.sup.6A consensus motif as well as distribution of m.sup.6A within the transcriptome and enrichment around the stop codon within transcripts in both genotypes. YTHDF2 loss is expected to result in the upregulation of direct target transcripts, indeed we observed an enrichment for m.sup.6A occupancy in the significantly upregulated genes (P<0.05, 754 genes) in Ythdf2.sup.CKO pre-leukemic cells compared to the corresponding unchanged or downregulated gene sets (FIG. 8B). Reciprocally, we analyzed the transcriptome based on RNA m.sup.6A modification and found that transcripts that contain m.sup.6A show increased expression in the Ythdf2.sup.CKO pre-leukemic cells (FIG. 8C-D).

[0307] The upregulation of m.sup.6A-containing transcripts in the absence of YTHDF2 may arise from an increase in their half-life. We therefore measured mRNA half-life transcriptome-wide in pre-leukemic cells using SLAM-seq (as previously described (Herzog et al., Nat Methods 14:1198-1204, 2017)) which revealed an overall modest increase in mRNA half-life in Ythdf2.sup.CKO cells (FIG. 8E). Interestingly, m.sup.6A-containing transcripts displayed overall shorter half-lives compared to non-m.sup.6A transcripts in Ythdf2.sup.CTL cells (FIG. 8F). YTHDF2 loss extended the half-life of m.sup.6A-containing transcripts (FIG. 8F-G). These data indicate that m.sup.6A-directed YTHDF2-mediated mRNA decay regulates the leukemic transcriptome.

Example 14. YTHDF2 Decreases Stability in Tnfrsf1b Encoding TNFR2

[0308] Inspecting m.sup.6A modified transcripts that contain m.sup.6A in both mouse and human AML, are upregulated in Ythdf2.sup.CKO LSCs we found TNF receptor 2 (TNFR2) encoded by Tnfrsf1b gene (FIG. 8H). Tnfrsf1 transcript negatively correlated with YTHDF2 expression and was highly associated with the loss of leukemogenic potential (FIG. 8I). We found that TNFRSF1B is highly methylated in mouse pre-leukemic cells and human AML cells (FIG. 8J). RIP-qPCR revealed co-precipitation of the Tnfrsf1b transcript with YTHDF2 (FIG. 8K). Concurrent with the increased half-life of Tnfrsf1b transcript (FIG. 8L), the surface expression of TNFR2 is upregulated on Ythdf2.sup.CKO pre-leukemic cells (FIG. 8M). We therefore tested if TNF stimulation had differential impact on Ythdf2.sup.CTL and Ythdf2.sup.CKO pre-leukemic cells. YTHDF2 loss rendered cells more sensitive to TNF-induced apoptosis (FIG. 8N). This indicates that TNFR2 upregulation may, at least in part, mediate decreased leukaemogenic capacity of YTHDF2 deficient AML cells.

Sequence Listing Notes:

[0309] SEQ ID NO: 1=YTHDF2 protein sequence (NCBI ACCESSION: NP_057342)

[0310] YTHDF2 protein (579 amino acids) consists of two key domains, namely P/Q/N-rich domain (residues 1-400) which binds CNOT (of the CCR4-NOT deadenylase complex) and the YTH domain (residues 401-579) responsible for binding to m.sup.6A of m.sup.6A-modified transcripts.

[0311] Residues 101-200 within the P/Q/N-rich domain of YTHDF2 are sufficient to bind CNOT (doi: 10.1038/ncomms12626).

[0312] YTHDF2 recognises m.sup.6A through the aromatic cage within its YTH domain. The aromatic cage is formed by Trp432, Trp486 and Trp491 residues (doi:10.1038/cr.2014.153).

[0313] SEQ ID NO: 2=YTHDF2 transcript sequence (NCBI ACCESSION: NM_016258)