Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2016 Jan 5;7:10258.
doi: 10.1038/ncomms10258.

The KDM3A-KLF2-IRF4 Axis Maintains Myeloma Cell Survival

Affiliations
Free PMC article

The KDM3A-KLF2-IRF4 Axis Maintains Myeloma Cell Survival

Hiroto Ohguchi et al. Nat Commun. .
Free PMC article

Abstract

KDM3A is implicated in tumorigenesis; however, its biological role in multiple myeloma (MM) has not been elucidated. Here we identify KDM3A-KLF2-IRF4 axis dependence in MM. Knockdown of KDM3A is toxic to MM cells in vitro and in vivo. KDM3A maintains expression of KLF2 and IRF4 through H3K9 demethylation, and knockdown of KLF2 triggers apoptosis. Moreover, KLF2 directly activates IRF4 and IRF4 reciprocally upregulates KLF2, forming a positive autoregulatory circuit. The interaction of MM cells with bone marrow milieu mediates survival of MM cells. Importantly, silencing of KDM3A, KLF2 or IRF4 both decreases MM cell adhesion to bone marrow stromal cells and reduces MM cell homing to the bone marrow, in association with decreased ITGB7 expression in MAF-translocated MM cell lines. Our results indicate that the KDM3A-KLF2-IRF4 pathway plays an essential role in MM cell survival and homing to the bone marrow, and therefore represents a therapeutic target.

Figures

Figure 1
Figure 1. KDM3A expression in MM cells.
(a) KDM3A mRNA expression in patient MM samples. Publicly available microarray data sets (GSE5900 and GSE6691) were analysed for mRNA expression of KDM3A in normal plasma cells (NPC), monoclonal gammopathy of undetermined significance (MGUS), smoldering multiple myeloma (Smoldering MM) and MM cells. **P<0.01, ***P<0.001. ns, not significant; analysis of variance (ANOVA) followed by Tukey's test. (b) Immunoblot analysis of KDM3A in patient MM cells and MM cell lines. Arrowhead represents KDM3A. Data are representative of at least two independent experiments, except for patient samples' blot.
Figure 2
Figure 2. KDM3A is indispensable for the survival of MM cells.
(a) Immunoblot analysis of KDM3A, H3K9me2 and H3K9me3 in KDM3A-knockdown RPMI8226 and MM.1S cells. Cells were transduced with either KDM3A-specific shRNA (shKDM3A #1 and #2) or shLuc by lentivirus. Whole-cell lysates were subjected to immunoblot analysis. Actin or H3 served as the loading control for each membrane. (b) MM cell lines (RPMI8226, MM.1S and U266) were transduced with shKDM3A or control shLuc. Three days post infection, which was designated as day 0, cells were plated in 96-well plates. Cell viability was measured on days 0 and 5 by MTT assay, and the cell growth rate (day 5/day 0) relative to shLuc was determined. (c) RPMI8226 cells were retrovirally transduced with the KDM3A cDNA carrying synonymous mutations in the shKDM3A #2 target sequence or with empty vector. Cells stably expressing the KDM3A cDNA or empty vector were then lentivirally transduced with shKDM3A #2 or shLuc. The cell growth rate (day 5/day 0) after lentiviral infection was determined for shKDM3A relative to shLuc. The growth rate for control shLuc in each cell type expressing the KDM3A cDNA or empty vector is set as 100%. (d) MM.1S cells transduced with shKDM3A #2 or shLuc (4 × 106 viable cells) were subcutaneously injected into SCID mice. Data represent mean±s.e.m. n=7 mice per group. An image of a representative tumour in each group is shown (top panel). (e) Apo2.7 immunostaining in RPMI8226 cells after knockdown of KDM3A. Cells were analysed 3 days post infection (day 0) and on days 2, 4, 6 for apoptosis by measuring apo2.7-positive cells. (f) Immunoblot analysis of caspase 8, caspase 7 and PARP in whole-cell lysates from KDM3A-knockdown RPMI8226 and MM.1S cells. Arrowheads indicate cleaved form. Actin served as the loading control. For b,c,e, data represent mean±s.d. of quintuplicate cultures (b,c) or duplicate cultures (e). For ac,e,f, data are representative of at least two independent experiments. **P<0.01, ***P<0.001 compared with control; Student's t-test.
Figure 3
Figure 3. KDM3A directly regulates KLF2 and IRF4 expression through H3K9 demethylation at their promoters in MM cells.
(a) Heatmap depicting the relative gene expression in RPMI8226 cells transduced with two independent shRNAs targeting KDM3A (shKDM3A #1 and shKDM3A #2) or shLuc. A total of 305 probes were selected based on ≥1.5-fold downregulation in KDM3A-knockdown cells and clustered. (bd) Quantitative real-time PCR (b) and immunoblot (c,d) analysis of KLF2 and IRF4 in RPMI8226 cells transduced with either shKDM3A or shLuc. After 3 days of infection, which is defined as day 0 in Fig. 2, cells were harvested for isolation of total RNA or whole-cell lysates. (b) Values represent the amount of mRNA relative to shLuc, defined as 1. (c) Arrowhead represents KLF2. (d) Signal intensity of each immunoblot was quantified using the ImageJ software. Results were normalized by Actin and are shown as relative signal intensity (shLuc=1). Error bars represent s.e.m of three independent experiments. (e,f) Quantitative real-time PCR (e) and immunoblot (f) analysis of KLF2 and IRF4 in MM.1S and U266 cells transduced with either shKDM3A or shLuc. After 3 days of infection, cells were harvested for isolation of total RNA or whole-cell lysates. (g) ChIP analysis showing KDM3A occupancy on KLF2 and IRF4 core promoters in RPMI8226 cells. Results were normalized to control IgG. The MYOD1 promoter region was used as negative control. (h) KDM3A occupancy is abrogated by KDM3A knockdown on KLF2 and IRF4 promoter regions in RPMI8226 cells. RPMI8226 cells transduced with either shKDM3A or shLuc were used for ChIP, followed by quantitative real-time PCR. (i) Enrichment of H3K9 methylation by KDM3A knockdown on KLF2 and IRF4 promoter regions in RPMI8226 cells. ChIP assays were performed with RPMI8226 cells transduced with either shKDM3A or shLuc. The relative enrichment over the input was assessed. Results are shown as fold enrichment compared with control shLuc. For b,e,gi, error bars represent s.d. of triplicate measurements. For bi, data are representative of at least two independent experiments. **P<0.01, ***P<0.001 compared with control; Student's t-test.
Figure 4
Figure 4. KLF2 is required for MM cell survival.
(a) Immunoblot analysis of KLF2 in patient MM cells and MM cell lines. Arrowhead represents KLF2. Whole-cell lysates from KLF2-overexpressed or knockdown RPMI8226 cells were used as positive or negative control for KLF2 antibody (N2212), respectively. (b) MM cell lines (RPMI8226, MM.1S and U266) were transduced with either KLF2-specific shRNA (shKLF2 #1 and #2) or control shLuc by lentivirus. Cell viability was measured on days 0 and 5 by MTT assay, and the cell growth rate (day 5/day 0) relative to shLuc was determined. (c) Immunoblot analysis of caspase 8, caspase 7 and PARP in KLF2-knockdown RPMI8226 and MM.1S cells. Arrowheads indicate cleaved form. Actin served as the loading control. (d) RPMI8226 cells expressing the KLF2 cDNA carrying synonymous mutations in the shKLF2 #2 target sequence or empty vector were transduced with shKLF2 #2 or shLuc by lentivirus. The growth rate for control shLuc in each cell type expressing the KLF2 cDNA or empty vector is set as 100%. (e) MM.1S cells transduced with shKLF2 #2 or shLuc (4 × 106 viable cells) were subcutaneously injected into SCID mice. Data represent mean±s.e.m. n=7 mice per group. An image of a representative tumour in each group is shown (right panel). Data of control shLuc group are the same data shown in Fig. 2d since subcutaneous xenograft studies of MM.1S cells treated with shKDM3A and shKLF2 were performed simultaneously with a common control shLuc group. For b,d, data represent mean±s.d. of quintuplicate cultures. For ad, data are representative of at least two independent experiments, except for patient samples' blot (a). *P<0.05, ***P<0.001 compared with control; Student's t-test.
Figure 5
Figure 5. IRF4 is a direct transcriptional target of KLF2 in MM cells.
(a) Microarray analysis in RPMI8226 cells transduced with shKDM3A, shKLF2 or shLuc was performed. The genes significantly downregulated by KDM3A or KLF2 knockdown compared with control were used as gene sets for the GSEA. Normalized enrichment score (NES) and P values are shown. (b,c) Quantitative real-time PCR of KLF2 and IRF4 after knockdown (b) and overexpression (c) of KLF2 in RPMI8226 cells. (d) Immunoblot analysis of KLF2 and IRF4 after knockdown of KLF2 in RPMI8226 cells. Shown are the relative signal intensity (shLuc=1) normalized by Actin. Error bars represent s.d. of two independent experiments. (e) RPMI8226 cells expressing the KLF2 cDNA carrying synonymous mutations in the shKLF2 #2 target sequence or empty vector were transduced with shKLF2 #2 or shLuc. Whole-cell lysates were subjected to immunoblot analysis. (f) Quantitative real-time PCR and immunoblot analysis of KLF2 and IRF4 after knockdown of KLF2 in MM.1S and U266 cells. (g) Transactivation of the IRF4 promoter by KLF2. The indicated amounts of KLF2 expression plasmids were transfected into 293T cells together with the human IRF4 promoter-luciferase reporter. The value above each bar indicates the induction level compared with empty vector. (h) 293T cells were co-transfected with the indicated IRF4-luciferase reporter and 0.04 μg of KLF2 expression plasmid or empty vector, and then assayed for luciferase activity. The fold activation (normalized luciferase activity co-transfected with KLF2 expression plasmid relative to empty vector) is shown. (i) ChIP analysis showing KLF2 occupancy on IRF4 promoter in RPMI8226 cells. GAPDH promoter was used as negative control. (j) KLF2 occupancy is abrogated by KLF2 knockdown on IRF4 promoter in RPMI8226 cells. RPMI8226 cells transduced with either shKLF2 or shLuc were used for ChIP. For b,c,f,i,j, error bars represent s.d. of triplicate measurements. For g,h, data represent mean±s.d. of three (g) or two (h) biological replicates. For bj, data are representative of at least two independent experiments. *P<0.05, **P<0.01, ***P<0.001 compared with control; Student's t-test.
Figure 6
Figure 6. KLF2 is a direct target of IRF4 transactivation in MM cells.
(a,b) Quantitative real-time PCR and immunoblot analysis of IRF4 and KLF2 in RPMI8226 (a), MM.1S and U266 (b) cells transduced with either shRNA targeting IRF4 (shIRF4) or shLuc. (a, right panel) Shown are the relative signal intensity (shLuc=1) normalized by Actin. Error bars represent s.e.m. of three independent experiments. (c) Quantitative real-time PCR and immunoblot analysis of IRF4 and KLF2 after overexpression of IRF4 in RPMI8226 cells. (d) RPMI8226 cells expressing the IRF4 cDNA or empty vector were transduced with shIRF4 #1 that targets the 3′ untranslated region of IRF4 or shLuc by lentivirus. Whole-cell lysates were subjected to immunoblot analysis with antibodies against IRF4 and KLF2. (e) ChIP analysis showing IRF4 occupancy on KLF2 second intron in RPMI8226 cells. ChIP assays were performed for the regions that contain tandem IRF4 motifs on KLF2 promoter and intron (indicated PCR positions are relative to the transcriptional start site). Results were normalized to control IgG. The SUB1 promoter or GAPDH promoter was used as positive or negative control, respectively. (f) Significant correlation between the genes downregulated by KLF2 knockdown and IRF4 knockdown in MM cells. The genes significantly downregulated by IRF4 knockdown were used as gene set for GSEA. NES and P values are shown. For a (left panel), b,c,e, error bars represent s.d. of triplicate measurements. For ae, data are representative of at least two independent experiments. **P<0.01, ***P<0.001 compared with control; Student's t-test.
Figure 7
Figure 7. The KDM3A–KLF2–IRF4 axis contributes to adhesion and homing of MM cells to the bone marrow.
(a) RPMI8226, MM.1S and U266 cells expressing each shRNA were labelled with calcein-AM and incubated in bone marrow stromal cells (BMSC)- or fibronectin (FN)-coated plates. After washing plates, adherent cells were quantified using a fluorescence plate reader. Results are shown as per cent adhesion over the input. BSA-coated wells were used as negative control. (b) MM.1S cells transduced with each shRNA were placed in the upper chamber of a modified Boyden chamber and incubated with CXCL12 (20 nM) added to the lower chamber. After 4 h, cells that migrated to the lower chamber were quantified. Results are shown as per cent migration over the input. (c) MM.1S-Luc-GFP cells expressing each shRNA (15 × 106 viable cells) were intravenously injected into NOD-SCID mice. Twenty-four hours after injection, bone marrow cells of recipient mice were collected, and GFP-positive cells were detected using flow cytometry. Results are shown as homing efficacy relative to shRFP cells. Data represent mean±s.e.m. n=4–5 mice per group. *P<0.05, ***P<0.001 compared with control; ANOVA followed by Dunnett's test. (d) Immunoblot analysis of ITGB7 after knockdown of KDM3A, KLF2, IRF4 in RPMI8226 and MM.1S cells. Shown are the relative signal intensity (shLuc=1) normalized by Actin. Error bars represent s.d. of two (RPMI8226) or three (MM.1S) independent experiments. (e) RPMI8226 cells expressing each cDNA were transduced with each shRNA as shown. Whole-cell lysates were subjected to immunoblot analysis. Data are representative of two independent experiments. (f) Adhesion assays were performed with RPMI8226 and MM.1S cells treated with either shITGB7 or shLuc. (g) Model of the KDM3A–KLF2–IRF4 axis in MM cells. KDM3A demethylates H3K9 of KLF2 and IRF4 promoter, resulting in upregulation of KLF2 and IRF4. KLF2 and IRF4 mutually activate expression of each other. The KDM3A–KLF2–IRF4 axis maintains MM cell survival and homing to the bone marrow. For a,b,f, data represent mean±s.d. of quadruplicate incubations (a,f), or triplicate incubations (b). *P<0.05, **P<0.01, ***P<0.001 compared with control; Student's t-test.

Similar articles

See all similar articles

Cited by 31 articles

See all "Cited by" articles

References

    1. Greer E. L. & Shi Y. Histone methylation: a dynamic mark in health, disease and inheritance. Nat. Rev. Genet. 13, 343–357 (2012). - PMC - PubMed
    1. Black J. C., Van Rechem C. & Whetstine J. R. Histone lysine methylation dynamics: establishment, regulation, and biological impact. Mol. Cell 48, 491–507 (2012). - PMC - PubMed
    1. Dawson M. A. & Kouzarides T. Cancer epigenetics: from mechanism to therapy. Cell 150, 12–27 (2012). - PubMed
    1. Martinez-Garcia E. et al. The MMSET histone methyl transferase switches global histone methylation and alters gene expression in t(4;14) multiple myeloma cells. Blood 117, 211–220 (2011). - PMC - PubMed
    1. Kuo A. J. et al. NSD2 links dimethylation of histone H3 at lysine 36 to oncogenic programming. Mol. Cell 44, 609–620 (2011). - PMC - PubMed

Publication types

MeSH terms

Substances

Feedback