Loss of the vitamin D receptor triggers senescence in chronic myeloid leukemia via DDIT4-mediated DNA damage

Abstract Chronic myeloid leukemia (CML) is a hematopoietic malignancy driven by the fusion gene BCR::ABL1. Drug resistance to tyrosine kinase inhibitors (TKIs), due to BCR::ABL1 mutations and residual leukemia stem cells (LSCs), remains a major challenge in CML treatment. Here, we revealed the requirement of the vitamin D receptor (VDR) in the progression of CML. VDR was upregulated by BCR::ABL1 and highly expressed in CML cells. Interestingly, VDR knockdown inhibited the proliferation of CML cells driven by both BCR::ABL1 and TKI-resistant BCR::ABL1 mutations. Mechanistically, VDR transcriptionally regulated DDIT4 expression; reduced DDIT4 levels upon VDR knockdown triggered DNA damage and senescence via p53 signaling activation in CML cells. Furthermore, VDR deficiency not only suppressed tumor burden and progression in primary CML mice but also reduced the self-renewal capacity of CML-LSCs. Together, our study demonstrated that targeting VDR is a promising strategy to overcome TKI resistance and eradicate LSCs in CML.


Introduction
Chronic myeloid leukemia (CML) is a hematopoietic malignancy in which hematopoietic stem cells (HSCs) are transformed by the BCR::ABL1 fusion gene generated from a reciprocal translocation [t(9;22)(q34;q11.2)](Cortes et al., 2021 ).This fusion gene encodes the BCR::ABL1 protein with constitutively activated tyrosine kinase, resulting in the malignant expansion of myeloid cells in the bone marrow (BM) and peripheral blood (Druker, 2008 ).The successful application of tyrosine kinase inhibitors (TKIs) targeting the adenosine triphosphate binding GAPDH was used as a loading control.( C and D ) Immunoblotting analysis of VDR protein levels ( C ) and relative VDR mRNA levels ( D ) in BM cells from CML patients and healthy donors (normal, NM).GAPDH was used as a loading control.( E and F ) Immunoblotting analysis of the indicated protein levels ( E ) and relative VDR mRNA levels ( F ) in K562 cells treated with IM for 10 h.GAPDH was used as a loading control.( G and H ) Immunoblotting analysis of VDR and BCR::ABL1 proteins ( G ) and relative VDR mRNA levels ( H ) in K562 cells transduced with retroviruses encoding the indicated shRNA.shNC represents a nontargeting shRNA; shBA represents a shRNA targeting BCR::ABL1.GAPDH was used as a loading control.( I and J ) Immunoblotting analysis of the indicated proteins ( I ) and relative VDR mRNA levels ( J ) in KBM5-T315I cells treated with Po for 10 h.GAPDH was used as a loading control.( K and L ) Immunoblotting analysis of VDR protein levels ( K ) and relative VDR mRNA levels ( L ) in Ba/F3 cells transduced with retroviruses encoding the indicated BCR::ABL1 mutations or empty vector.GAPDH was used as a loading control.All P values were determined by an unpaired two-tailed Student's t-test except where indicated.Data are presented as mean ± SD from three independent experiments.See also Supplementary Figure S1.
receptor to bind to vitamin D-responsive elements of target genes, playing key roles in numerous biological processes (Deeb et al., 2007 ;Medrano et al., 2018 ).Due to its altered expression and function, the role of VDR in cancers has also been extensively studied (Campbell and Trump, 2017 ).Vitamin Dmediated VDR signaling activation exhibits antiproliferative effects against multiple cancers (Larriba et al., 2011 ;Li et al., 2012 ;Zhang et al., 2020 ;Ling et al., 2022 ).However, there is also increasing evidence to support an oncogenic role for VDR in tumor progression.VDR knockdown inhibited the growth of glioblastoma and T-ALL cells (Shirvani-Farsani and Behmanesh, 2019 ), and lung metastatic cancer growth was extremely reduced in VDR-null mice (Nakagawa et al., 2004 ).Loss of VDR eliminated breast and prostate cancers through downregulation of Wnt/ β-catenin signaling (Zheng et al., 2017 ).Furthermore, VDR functions as a master regulator of MYCN, which is highly expressed in over 70% of malignancies (Salehi-Tabar et al., 2012 ;Liu et al., 2020 ).These findings indicated that the distinct roles of VDR in cancers depend on the tumor cell type or whether it is independent of its ligand.
VDR expression and activity have been documented to be associated with several types of leukemia, such as acute lymphoblastic leukemia (ALL) and acute myelocytic leukemia (AML) (Pezeshki et al., 2018 ).Targeting VDR provides antileukemic activity by acting on cell differentiation and decreasing the stemness of AML cells (Paubelle et al., 2020 ).Accordingly, vitamin D and its analogs have been considered potent drugs for AML therapy via VDR activation (Paubelle et al., 2020 ;Sabatier et al., 2021 ).However, the role of VDR in CML is still incompletely described.In this study, we present an unexpected role of VDR in the progression of CML.Targeting VDR provides an effective therapy for CML, representing a novel therapeutic target for overcoming resistance to TKIs.

VDR is upregulated by the fusion gene BCR::ABL1 in CML
To explore the potential role of VDR in CML, we used a wellestablished BM transplantation model with ectopic expression of BCR::ABL1 (referred to as BCR::ABL1-driven CML mice), which mimics the pathogenesis in human CML patients.Immunoblotting analysis demonstrated that VDR was upregulated in BM cells from BCR::ABL1-driven CML mice compared with that of the control group, whereas the mRNA level of VDR showed no detectable changes in either group of mice ( Figure 1 A and B).In line with these findings, similar upregulation of VDR was observed in Ba/F3 cells with ectopic expression of BCR::ABL1 (referred to as Ba/F3 BCR :: ABL1 cells) ( Supplementary Figure S1A and B).Importantly, the protein but not mRNA levels of VDR were also significantly upregulated in BM samples from CML patients compared with the corresponding healthy donors ( Figure 1 C and D).These data suggest that BCR::ABL1 regulates VDR expression in CML.
We then challenged K562 and KBM5, two CML cell lines harboring the fusion gene BCR::ABL1 , with IM, which specifically inhibits the tyrosine kinase activity of BCR::ABL1.As expected, VDR expression was notably reduced by IM treatment ( Figure 1 E and F; Supplementary Figure S1C and D).Furthermor e, si l encin g BCR::ABL1 with shRNAs also led to the downregulation of VDR in K562 cells ( Figure 1 G and H).Considering that the protein but not mRNA level of VDR was greatly upregulated in Ba/F3 BCR :: ABL1 cells, we challenged these cells with cycloheximide (CHX) to inhibit protein synthesis and found that BCR::ABL1 overexpression led to the retaining abundance of VDR compared to that of control cells ( Supplementary Figure S1E and F).These findings demonstrated that BCR::ABL1 regulated VDR mRNA expression and stabilized VDR.
BCR::ABL1 T315I is a common point mutation resistant to all earlier-generation TKIs, remaining one major obstacle to vanquishing CML (Braun et al., 2020 ;Hughes and Shanmuganathan, 2022 ).Interestingly, we found that VDR was also highly expressed in BCR::ABL1 T315I -driven CML mice and Ba/F3 cells ( Figure 1 A and B; Supplementary Figure S1A  and B).Both the protein and mRNA levels of VDR were dramatically reduced in KBM5-T315I cells harboring the IMresistant BCR::ABL1 T315I mutation by treatment with ponatinib (Po), which efficiently inhibited the activity of BCR::ABL1 T315I ( Figure 1 I and J).We also challenged Ba/F3 T315I cells with CHX but found that BCR::ABL1 T315I overexpression led to similar changes in VDR protein levels compared to those in control cells ( Supplementary Figure S1G and H).
To further confirm that VDR expression is regulated by BCR::ABL1 independent of its mutations, we generated corresponding Ba/F3 cells stably expressing several documented drug-resistant BCR::ABL1 mutations and found that VDR exhibited high expression in these mutant BCR::ABL1 compared with control cells ( Figure 1 K and L).Notably, similar upregulation of VDR was also observed in Ba/F3 cells with ectopic expression of BCR::ABL1 E255V+ T315I , which confers resistance to Po ( Supplementary Figure S1I and J).These data demonstrated that VDR expression is upregulated in CML independent of BCR::ABL1 mutations.

VDR knockdown inhibits CML cell proliferation
To examine whether high expression of VDR is involved in the progression of CML, we first performed proliferation assays in K562 cells transduced with retrovirus-encoding VDR shRNA.Indeed, we examined BCR::ABL1 expression in K562 cells with VDR silencing and found that VDR knockdown had no effect on BCR::ABL1 ( Supplementary Figure S2A and B).Comp ar ed t o the control group, VDR silencing led to substantial inhibition of K562 cell proliferation ( Figure 2 A and B).This was further determined by the repressed DNA replication indicated by the reduced EdU staining in K562 cells upon VDR knockdown ( Figure 2 C).Conversely, overexpression of VDR led to a mild increase in K562 cell growth ( Supplementary Figure S2C and D).
To explore the underlying mechanism by which VDR regulates CML cell proliferation, we analyzed the cell cycle profile of K562 cells and found that VDR knockdown had minimal effects on the cell cycle ( Supplementary Figure S2E).In addition, K562 cells with VDR silencing also exhibited comparable cell viability to  that of the control group (data not shown).Since cellular senescence is characterized by permanent proliferation arrest, we performed a senescence-associated-β-galactosidase (SA-β-gal, a biomarker of senescence) staining assay and found that VDR knockdown triggered senescence, as indicated by the marked increase in SA-β-gal-positive K562 cells ( Figure 2 D).
Compound mutations in BCR::ABL1 contribute to resistance to TKIs in CML patients, which remains a clinical challenge for CML therapy (Cortes et al., 2021 ).Interestingly, similar inhibition of cell proliferation induced by VDR knockdown was also observed in KBM5 and IM-resistant KBM5-T315I cells ( Figure 2 A and B).In line with the upregulation of VDR in BCR::ABL1 mutation-driven Ba/F3 cells, VDR knockdown significantly inhibited the proliferation of all these transformed CML cells ( Supplementary Figure S2F).Importantly, even Po-resistant mutation-driven proliferation was dramatically reduced by VDR knockdown in Ba/F3 cells ( Supplementary Figure S2G).These data demonstrated that inhibiting VDR effectively suppressed the proliferation of CML cells independent of BCR::ABL1 mutations, avoiding the resistance of CML to TKIs.
To examine whether VDR depletion increases CML cell sensitivity to TKIs, we performed proliferation assays in KBM5 and KBM5-T315I cells transduced with retrovirus-encoding VDR shRNA, followed by treatment with IM and Po, respectively.Compared to the corresponding controls, both TKIs showed similar inhibition of CML cell proliferation upon VDR silencing.However, combined VDR knockdown and TKIs exhibited synergistic inhibition of CML cell proliferation ( Supplementary Figure S2H).

VDR knockdown triggers senescence in CML cells via DDIT4-mediated DDR signaling of γ -H2AX/p53/p21
To understand the molecular mechanism underlying the cellular senescence induced by VDR silencing, we performed bulk RNA sequencing of K562 cells transduced with retrovirus-encoding VDR shRNA.A total of 1502 differentially expressed genes (DEGs) were found ( ≥1.2-fold, P < 0.05), including 639 upregulated and 863 downregulated genes ( Supplementary Figure S3A).Enrichment analysis showed that these DEGs were involved in DNA damage, cell cycle regulation, and p53 signaling ( Supplementary Figure S3B and Table S1).Consistently, immunoblotting analysis demonstrated that VDR knockdown led to the elevated phosphorylation of H2AX, a well-known DNA damage marker, accompanied by increased p53 phosphorylation and p21 expression ( Figure 3 A).Furthermore, p53 inhibition significantly recovered the senescence and cell proliferation inhibition driven by VDR knockdown ( Supplementary Figure S3C and D), suggesting that VDR knockdown triggered DNA damage response (DDR) signaling, which consequently resulted in senescence in K562 cells.
Among these DEGs, DNA damage-inducible transcript 4 ( DDIT4 ), known to be induced by various cellular stresses (Ellisen et al., 2002 ), was downregulated upon VDR knockdown.This was further confirmed by its reduced protein and mRNA levels in K562 cells transduced with VDR shRNA compared to the control group ( Figure 3 A and B).In contrast, VDR overexpression resulted in a remarkable increase in the protein and mRNA levels of DDIT4 in K562 cells ( Figure 3 C and D).Considering that VDR is a member of the nuclear receptor superfamily and regulates the expression of numerous genes, we directly explored whether DDIT4 is transcriptionally regulated by VDR.A luciferase assay showed that ectopic expression of VDR markedly increased luciferase activity in HEK293T cells transduced with the DDIT4 promoter and VDR ( Figure 3 E).We also performed a chromatin immunoprecipitation assay to validate the binding of VDR with the DDIT4 promoter.By sequentially using ChIP followed by quantitative PCR, we found that VDR was preferentially bound to the regions most distal to the start codon of DDIT4 ( Figure 3 F; Supplementary Figure S3E).These data demonstrated that DDIT4 is a transcriptional target of VDR.
To assess the role of DDIT4 in CML cell proliferation mediated by VDR, we first examined DDIT4 expression in CML cells.Immunoblotting analysis demonstrated that DDIT4 was upregulated in BM cells from both BCR::ABL1-driven and BCR::ABL1 T315I -driven CML mice compared to the corresponding control group ( Supplementary Figure S3F).Similarly, high expression of DDIT4 was also observed in BM samples from CML patients compared to the corresponding healthy donors ( Supplementary Figure S3G and H).In addition, the high expression of DDIT4 was dramatically abolished by IM treatment ( Supplementary Figure S3I and J).We then performed rescue experiments in K562 cells and found that enforced expression of DDIT4 obviously restored the activated DDR signaling of γ-H2AX/p53/p21 induced by VDR knockdown ( Figure 3 G).As a downstream consequence, cellular senescence and subsequent cell proliferation inhibition were also successfully recovered

Loss of VDR ameliorates BCR::ABL1-driven CML phenotypes
To clarify the functional roles of VDR in the progression of BCR::ABL1-driven CML, we utilized a VDR knockout (KO) mouse model, in which VDR deficiency was confirmed by the undetectable expression of VDR in BM cells compared to the wildtype (WT) control ( Supplementary Figure S4A and B).Indeed, KO mice exhibited comparable blood cell counts in peripheral blood and BM, excluding a minor decrease in platelets ( Supplementary Figure S4C and D), as well as a similar frequency of hematopoietic stem and progenitor cells (HSPCs) in BM compared to that of WT controls ( Supplementary Figure S4E).These data indicated that VDR does not contribute to steadystate hematopoiesis in vivo .
We then utilized a CML mouse model to assess the impact of VDR deletion on the disease, in which VDR-KO or WT BM was transduced with BCR::ABL1 and then transplanted into irradiated recipient mice (referred to as WT-CML and KO-CML mice) ( Figure 4 A).Homing assays demonstrated that VDR deficiency did not affect the engraftment of HSCs ( Supplementary Figure S4F).The leukemia burden was evaluated by assessing the degree of organ lesions and CML cell infiltration.As previously reported (Peng and Li, 2016 ), the most common blast crisis in CML is myeloid ( Supplementary Figure S4G).KO-CML mice showed a substantial reduction in leukemic cells (GFP + Gr1 + ) in the peripheral blood and BM after 14 days of CML model establishment, accounting for the reduced white blood cell count ( Figure 4 B-D).Similar amelioration of the CML blast crisis was also observed in the spleens of KO-CML mice, including reversed splenomegaly ( Figure 4 E) and decreased infiltration ( Figure 4 F and G).Correspondingly, loss of VDR significantly reduced BCR::ABL1induced lethality, even in the secondary transplanted mice ( Figure 4 H; Supplementary Figure S4H).
Furthermore, VDR deficiency led to a lower frequency of leukemia stem cells (LSCs; GFP + Lin − Sca-1 + c-Kit + ) in BM ( Figure 4 I).This was further confirmed by the substantial reduction in the colony-forming cell (CFC)/replating capacity of LSCs from KO-CML mice ( Figure 4 J and K).Consistent with findings in CML cells in vitro , KO-CML mice exhibited prominent activation of DDR signaling and reduced DDIT4 expression, accompanied by a remarkable increase in cellular senescence in BM compared with WT-CML mice ( Supplementary Figure S4I-K).The se r e s u lts demonstrated that VDR plays important roles in the pathogenesis of BCR::ABL1-driven CML.

VDR deletion is an effective therapeutic strategy for overcoming TKI resistance in CML
Considering that the upregulated VDR expression and the inhibition of cell proliferation upon VDR knockdown in CML were independent of BCR::ABL1 mutations, we investigated whether inhibiting VDR could overcome the resistance of CML to TKIs in vivo .We established a similar CML mouse model with an IMresistant BCR::ABL1 T315I mutation (referred to as CML T315I mice) ( Figure 5 A) and found that VDR deletion significantly enhanced the survival of CML T315I mice, which died 20 days after model establishment ( Figure 5 B), even in secondary transplanted mice ( Supplementary Figure S5).KO-CML T315I mice exhibited a remarkable reduction in white blood cell count and frequencies of leukemic cells (GFP + Gr1 + ) in peripheral blood and BM compared to WT-CML T315I mice ( Figure 5 C-E).In addition, VDR deficiency also led to the substantial amelioration of the CML blast crisis in the spleen, as indicated by reversed splenomegaly and decreased infiltration ( Figure 5 F-H).
The frequencies of LSCs (GFP + Lin − Sca-1 + c-Kit + ) were significantly reduced upon VDR deletion ( Figure 5 I), although they showed similar engraftments (data not shown).In line with the in vivo findings, a reduced number and size of colonies were also observed in LSCs from KO-CML T315I mice ( Figure 5 J and K), suggesting that loss of VDR efficiently repressed the proliferation of LSCs.

Discussion
In this study, we identified VDR as a potent target for CML.VDR knockdown suppresses the proliferation of CML cells independent of BCR::ABL mutations and reduces LSCs.Although VDR is upregulated by the CML-driven gene BCR::ABL1 via transcriptional activation or protein stability that is dependent on the corresponding mutations, inhibiting VDR is sufficient to inhibit CML cell proliferation driven by BCR::ABL1 independent of its mutations in vitro and in vivo .
Although BCR::ABL1 inhibition led to a decrease in VDR mRNA expression, VDR mRNA levels were unchanged or mildly increased in the CML mouse model and Ba/F3 cells with ectopic expression of BCR::ABL1 and CML patients.Furthermore, BCR::ABL1 enhanced the protein stability of VDR.These findings indicated that BCR::ABL1 regulates VDR mRNA expression and  stabilizes VDR.It is possible that the accumulated VDR provides negative feedback to suppress the transcriptional activity of VDR, leading to its low basis of mRNA expression in CML cells.It has been reported that VDR undergoes multiple posttranslational modifications (phosphorylation, SUMOylation, and ubiquitination) that mediate the degradation of VDR (Zhi et al., 2011 ;Wei et al., 2018 ).Although BCR::ABL1 is a fusion protein with kinase activity, we did not observe the interaction between BCR::ABL1 and VDR via co-IP assays (data not shown).Therefore, BCR::ABL1 may interact with and activate an unknown protein that mediates posttranslational modification of VDR to protect it from degradation.
Our data showed that VDR was also highly expressed in BCR::ABL1 T315I -driven CML mice and Ba/F3 cells.However, BCR::ABL1 T315I upregulated its mRNA expression but not its protein stability, suggesting that BCR::ABL1 and its mutations regulate VDR expression via different regulatory mechanisms.This was further confirmed by the distinct VDR expression in other BCR::ABL1 mutation-driven Ba/F3 cells.
DNA damage and compromised DNA repair lead to suppression of tumor development through the induction of apoptosis, possibly with contributions by cell cycle arrest and cell senescence (Jin and Robertson, 2013 ).Consistent with one previous study showing that vitamin D/VDR axis regulates DNA repair during oncogene-induced senescence (Graziano et al., 2016 ), we found that VDR knockdown led to DDR and subsequent senescence, accounting for the proliferative inhibition in CML cells.Indeed, VDR regulates the expression of the DNA repair genes RAD50 and ATM in response to DNA damage (Ting et al., 2012 ).
DDIT4, also known as regulated in development and DNA damage response-1 (Redd1), is highly expressed and associated with a worse prognosis in acute myeloid leukemia and solid tumors (Pinto et al., 2017 ).Our data demonstrated that DDIT4 is the target gene of VDR in CML cells, which is responsible for the DNA damage and senescence mediated by VDR.In line with our findings, DDIT4 has recently been reported to be downregulated during oxidative stress-and UV-induced senescence, and its overexpression reverses this cellular senescence (Lee et al., 2022 ).Therefore, our data suggest that the VDR/DDIT4 axis plays important roles in CML cell proliferation via DNA damage repair, and its deficiency leads to DNA damage accumulation and senescence.
Alternatively, it is also possible that DDIT4 directly regulates p53 signaling to induce senescence.p53 is a central regulator of the DDR and genomic integrity, and its activation induces the transcription of its critical effector p21 to drive fundamental cellular processes, including senescence (Vousden and Prives, 2009 ;Brady et al., 2011 ).DDIT4 is a crucial inhibitor of mTOR that exerts different effects on cell proliferation and senescence (Brugarolas et al., 2004 ;DeYoung et al., 2008 ), and its knockdown exerts mTORC1-dependent control to activate p53 signaling (Vadysirisack et al., 2011 ;Du et al., 2018 ).Consistently, we observed that VDR knockdown led to mTOR signaling activation in K562 cells (data not shown) in parallel with p53 signaling activation.This may be due to reduced DDIT4 expression.
Notably, our data also showed that targeting VDR reduced CML-LSCs.Residual leukemic stem cells are the major barrier against CML therapy, leading to disease relapse and progression (Braun et al., 2020 ).Indeed, VDR has been reported to regulate HSPC expansion and survival via IL-8 activity (Cortes et al., 2016 ).On the other hand, VDR functions as a critical modulator of HSC trafficking via the neuronal control of the HSC niche (Kawamori et al., 2010 ).In line with these findings, we also elucidated a role for VDR in LSC expansion, and the deficiency of VDR led to reduced LSCs in CML.Given the similar features of the quiescence of stem cells and cellular senescence characterized by limited proliferative activity (Hernandez-Segura et al., 2018 ), VDR knockdown may also trigger senescence in LSCs, which contributes to the impaired self-renewal of LSCs.
It is evident that senescence-inducing therapies are a promising strategy for cancer (Wang et al., 2020 ).Senescence not only directly controls tumor growth via proliferative inhibition but also inhibits tumor progression via activation of immunity mediated by senescence-associated secretory phenotype factors produced by senescent cells.On the other hand, senescenceinducing therapies can also be combined with senolytic treatments (senolytic agents) as a 'one-two punch' approach to cancer therapy (Wang et al., 2022 ).Recently, this 'one-two punch' strategy by sequential use of one-two punch prosenescence and senolytic therapy has been successfully used to kill cancer cells (Wang et al., 2019 ).Our data provide evidence that inhibiting VDR triggers senescence in CML cells, which leads to proliferative inhibition in vitro .Loss of VDR exhibited mild but significant amelioration of the disease burden and progression in primary CML mice.This may be due to the accumulation of senescent leukemia cells in vivo .Therefore, our study identified VDR as a potential target for prosenescence therapy, the combination of which with senolytic therapy can provide remarkable benefit for CML therapy.
In summary, we demonstrated that VDR is needed for BCR::ABL1-driven CML progression and that targeting VDR

Figure 1
Figure 1 VDR is highly expressed in CML cells, and BCR::ABL1 regulates VDR expression.( A and B ) Immunoblotting analysis of VDR and BCR::ABL1 protein levels ( A ) and relative VDR mRNA levels ( B ) in BM cells from the indicated CML mice and corresponding controls (Ctrl).GAPDH was used as a loading control.( C and D ) Immunoblotting analysis of VDR protein levels ( C ) and relative VDR mRNA levels ( D ) in BM cells from CML patients and healthy donors (normal, NM).GAPDH was used as a loading control.( E and F ) Immunoblotting analysis of the indicated protein levels ( E ) and relative VDR mRNA levels ( F ) in K562 cells treated with IM for 10 h.GAPDH was used as a loading control.( G and H ) Immunoblotting analysis of VDR and BCR::ABL1 proteins ( G ) and relative VDR mRNA levels ( H ) in K562 cells transduced with retroviruses encoding the indicated shRNA.shNC represents a nontargeting shRNA; shBA represents a shRNA targeting BCR::ABL1.GAPDH was used as a loading control.( I and J ) Immunoblotting analysis of the indicated proteins ( I ) and relative VDR mRNA levels ( J ) in KBM5-T315I cells treated with Po for 10 h.GAPDH was used as a loading control.( K and L ) Immunoblotting analysis of VDR protein levels ( K ) and relative VDR mRNA levels ( L ) in Ba/F3 cells transduced with retroviruses encoding the indicated BCR::ABL1 mutations or empty vector.GAPDH was used as a loading control.All P values were determined by an unpaired two-tailed Student's t-test except where indicated.Data are presented as mean ± SD from three independent experiments.See also Supplementary Figure S1.

Figure 2
Figure 2 VDR knockdown inhibits CML cell proliferation.( A ) Immunoblotting analysis of VDR protein levels in K562, KBM5, and KBM5-T315I cells transduced with retroviruses encoding the indicated shRNA.shNC represents a nontargeting shRNA.GAPDH was used as a loading control.( B ) Statistical analysis of cell proliferation in cells as in A .Data were obtained from three independent experiments.The P value was determined by two-way ANOVA.**** P < 0.0001.( C ) K562 cells transduced with retroviruses encoding the indicated shRNA were labeled with EdU for 2 h and stained with azide-conjugated Alexa567 (red) and DAPI (blue).shNC represents a nontargeting shRNA.Scale bar, 200 μm.Quantification of the percentage of EdU-positive cells is shown on the right.( D ) Representative images of SA-β-gal staining in K562 cells transduced with retroviruses encoding the indicated shRNAs.shNC represents a nontargeting shRNA.Quantitation of the percentage of SA-β-gal-positive cells in the indicated group is shown on the right.Scale bar, 25 μm.All P values were determined by an unpaired twotailed Student's t-test except where indicated.See also Supplementary Figure S2.

Figure 3
Figure 3 VDR knockdown triggers cell senescence in CML cells via DDIT4-mediated DDR signaling of γ-H2AX/p53/p21.( A and B ) Immunoblotting analysis of the indicated protein levels ( A ) and relative DDIT4 mRNA levels ( B ) in K562 cells transduced with retroviruses encoding shVDR or shNC.shNC represents a nontargeting shRNA.GAPDH was used as a loading control.( C and D ) Immunoblotting analysis of VDR and DDIT4 protein levels ( C ) and relative DDIT4 mRNA levels ( D ) in K562 cells transduced with retroviruses encoding VDR or empty

Figure 3 (
Figure 3 (Continued) vector (OE-C).GAPDH was used as a loading control.( D ) Quantification of DDIT4 mRNA expression in cells as in C .( E ) Luciferase reporter assays of the DDIT4 promoter in HEK293T cells cotransfected with a luciferase reporter construct bearing the -2000 DDIT4 promoter and a VDR plasmid or empty vector (OE-C).Data are presented as mean ± SD from three independent experiments.( F ) Chromatin immunoprecipitation analyses of VDR binding to the DDIT4 promoter in K562 cells.P1-P7 indicate fragments in the DDIT4 region amplified in ChIP-qPCR assays (see also Supplementary Figure S3E).( G ) Immunoblotting analysis of the indicated protein levels in K562 cells transduced with retroviruses encoding shVDR or shNC and DDIT4.shNC represents a nontargeting shRNA.GAPDH was used as a loading control.( H ) Ectopic expression of DDIT4 rescued VDR knockdown-induced senescence in K562 cells.Representative images of SA-β-gal staining are shown on the left.Quantitation of the percentage of SA-β-gal-positive cells in the indicated group is shown on the right.Scale bar, 12.5 μm.( I ) Statistical analysis of cell proliferation in cells as in G .Data were obtained from three independent experiments.The P value was determined by two-way ANOVA.* P < 0.05, *** P < 0.001.( J ) A schematic model depicting the role of the VDR-DDIT4-p53 axis in CML cells.All P values were determined by an unpaired two-tailed Student's t-test except where indicated.See also Supplementary Figure S3.

Figure 4
Figure 4 Loss of VDR ameliorates BCR::ABL1-driven CML phenotypes.( A ) Schematic representation of the BCR::ABL1-driven CML mouse model with WT or VDR KO mouse BM cells.CBC represents complete blood count.( B ) Quantification of the percentage of GFP + cells and GFP + Gr1 + cells in peripheral blood mononuclear cells (PBMCs) from the mice described in A ( N = 5).( C ) White blood cells (WBCs) in peripheral blood were measured 14 days after transplantation from the indicated mice as in A ( N = 5).( D ) Quantification of the percentage of GFP + cells and GFP + Gr1 + cells in the BM from the indicated CML mice measured by flow cytometry ( N = 5).( E ) Representative spleens of the indicated CML mice described in A .Statistical analysis of spleen weight is shown on the right ( N = 4).( F ) Representative hematoxylin and

Figure 4 (
Figure 4 (Continued) eosin (H&E) staining of spleens from the indicated CML mice 14 days after transplantation.Upper scale bar, 500 μm, lower scale bar, 50 μm.( G ) Quantification of the percentage of GFP + cells and GFP + Gr1 + cells in the spleen from the indicated mice measured by flow cytometry ( N = 5).( H ) Kaplan-Meier survival curves of the indicated CML mice as in A ( N = 8).P values were determined by the logrank (Mantel-Cox) test.( I ) Quantification of the percentage of GFP + LSK (Lin -Sca1 + c-Kit + ) in the BM from the indicated mice measured by flow cytometry ( N = 5).( J ) Representative colonies of GFP + c-Kit + BM cells from the indicated CML mice.Scale bar, 10 μm. ( K ) Quantification of the colony number in the indicated group as in J .All P values were determined by an unpaired two-tailed Student's t-test except where indicated.See also Supplementary Figure S4.

Figure 5
Figure 5 VDR deletion is an effective therapy for BCR::ABL1 T315I -induced CML. ( A ) Experimental procedure for the assessment of the in vivo effect of VDR deficiency on BCR::ABL1 T315I -driven CML (CML T315I ) mice.CBC represents complete blood count.( B ) Kaplan-Meier survival curves of the indicated CML mice as in A ( N = 7).P values were determined by the log-rank (Mantel-Cox) test.( C ) Quantification of the percentage of GFP + cells and GFP + Gr1 + cells in PBMCs from the mice described in A ( N = 5).( D ) WBCs in peripheral blood were measured 14 days after transplantation from mice as in C ( N = 5).( E ) Quantification of the percentage of GFP + cells and GFP + Gr1 + cells in the BM from the indicated mice measured by flow cytometry ( N = 5).( F ) Representative spleen of the indicated CML T315I mice described in A .Statistical analysis of spleen weight is shown on the right ( N = 6).( G ) Representative

Figure 5 (
Figure 5 (Continued) hematoxylin and eosin (H&E) staining of spleens from the indicated CML T315I mice 14 days after transplantation.Scale bar, 500 μm (upper) or 50 μm (lower).( H ) Quantification of the percentage of GFP + cells and GFP + Gr1 + cells in the spleen from the indicated mice measured by flow cytometry ( N = 5).( I ) Quantification of the percentage of GFP + LSK (Lin -Sca1 + c-Kit + ) in the BM from the indicated mice measured by flow cytometry ( N = 5).( J ) Representative colonies of GFP + c-Kit + BM cells from the indicated CML T315I mice.Scale bar, 10 μm. ( K ) Quantification of the colony number and colony area in the indicated group as in J .All P values were determined by an unpaired two-tailed Student's t-test except where indicated.See also Supplementary Figure S5.