Leukemia (2014) 28, 1844–1850 & 2014 Macmillan Publishers Limited All rights reserved 0887-6924/14 www.nature.com/leu

ORIGINAL ARTICLE

Haploinsufficiency of Sf3b1 leads to compromised stem cell function but not to myelodysplasia M Matsunawa1,6, R Yamamoto2,6, M Sanada1,6, A Sato-Otsubo1, Y Shiozawa1, K Yoshida1, M Otsu2, Y Shiraishi3, S Miyano3,4, K Isono5, H Koseki5, H Nakauchi2,7 and S Ogawa1,7 SF3B1 is a core component of the mRNA splicing machinery and frequently mutated in myeloid neoplasms with myelodysplasia, particularly in those characterized by the presence of increased ring sideroblasts. Deregulated RNA splicing is implicated in the pathogenesis of SF3B1-mutated neoplasms, but the exact mechanism by which the SF3B1 mutation is associated with myelodysplasia and the increased ring sideroblasts formation is still unknown. We investigated the functional role of SF3B1 in normal hematopoiesis utilizing Sf3b1 heterozygous-deficient mice. Sf3b1 þ /  mice had a significantly reduced number of hematopoietic stem cells (CD34  cKit þ ScaI þ Lin  cells or CD34  KSL cells) compared with Sf3b1 þ / þ mice, but hematopoiesis was grossly normal in Sf3b1 þ /  mice. When transplanted competitively with Sf3b1 þ / þ bone marrow cells, Sf3b1 þ /  stem cells showed compromised reconstitution capacity in lethally irradiated mice. There was no increase in the number of ring sideroblasts or evidence of myeloid dysplasia in Sf3b1 þ /  mice. These data suggest that SF3B1 plays an important role in the regulation of hematopoietic stem cells, whereas SF3B1 haploinsufficiency itself is not associated with the myelodysplastic syndrome phenotype with ring sideroblasts. Leukemia (2014) 28, 1844–1850; doi:10.1038/leu.2014.73

INTRODUCTION Frequent pathway mutations involving multiple components of the RNA splicing machinery are a cardinal feature of myeloid neoplasms, particularly those showing myeloid dysplasia in which the major mutational targets include SF3B1, U2AF1, SRSF2 and ZRSR2.1–4 SF3B1 mutations are one of the most common genetic alterations in myelodysplastic syndromes (MDS) and have also been reported in 5  15% of chronic lymphocytic leukemia cases,5 and at lower frequencies in a variety of solid cancers such as endometrial cancers,2 pancreatic carcinoma,6 breast cancers7 and uveal melanoma.8 SF3B1 mutations are considered to be one of the founding genetic events in MDS and define a benign clinical phenotype.2,9 The frequency of SF3B1 mutations is particularly high among the unique subtypes of MDS that are characterized by increased ring sideroblasts, such as refractory anemia with ring sideroblasts (RARS) or refractory cytopenia with multiple lineage dysplasia with ring sideroblasts as well as RARS associated with thrombocytosis9,10 in which mutation frequencies of 66.7  79% have been reported. These genetic findings strongly suggest a close relationship between SF3B1 mutation and the presence of ring sideroblasts. However, the molecular mechanism by which SF3B1 mutation leads to myelodysplasia and promotes the formation of ring sideroblasts is unknown. SF3B1 encodes subunit 1 of the splicing factor 3b complex that is a core component of U2 small nuclear ribonucleoprotein. The U2 small nuclear ribonucleoprotein complex recognizes the 30 splice site at intron–exon junctions in normal pre-mRNA splicing 1

machinery,11 in which SF3B1 is involved in recognition of the branchpoint sequence. It has been demonstrated that Sf3b1 knockout mice are embryonic lethal at very early stages, whereas Sf3b1 heterozygous knockout (Sf3b1 þ /  ) mice exhibit mild skeletal alterations.12 However, a detailed analysis of the functional role of Sf3b1 in hematopoiesis in these mice has not been reported. In this study we investigated the hematological phenotype of Sf3b1 þ /  mice to clarify the role of SF3B1 in hematopoiesis and to obtain insights into how deregulation of SF3B1 leads to the development of MDS phenotypes. MATERIALS AND METHODS Ethical approval of the study protocol Animal experiments were undertaken with the approval of the Animal Care and Use Committee of the Institute of Medical Science, University of Tokyo (Tokyo, Japan).

Mice Generation of Sf3b1 þ /  mice was as previously described.12 C57BL/6(CD45.1 þ ) mice and C57BL/6 F1-CD45.1 þ CD45.2 þ mice were purchased from Japan SLC (Shizuoka, Japan) and Sankyo-Lab Service (Tsukuba, Japan), respectively.

Iron staining Prussian blue stain (Muto Pure Chemicals, Tokyo, Japan) and nuclear red counterstain with nuclear fast red were performed by standard procedures. Light microscopic images were acquired on an OLYMPUS BX45 microscope

Departments of Pathology and Tumor Biology, Graduate School of Medicine, Kyoto University, Kyoto, Japan; 2Division of Stem Cell Therapy, Center for Stem Cell Biology and Medicine, Institute of Medical Science, University of Tokyo, Tokyo, Japan; 3Laboratory of DNA Information Analysis, Human Genome Center, Institute of Medical Science, The University of Tokyo, Tokyo, Japan; 4Laboratory of Sequence Analysis, Human Genome Center, Institute of Medical Science, The University of Tokyo, Tokyo, Japan and 5Laboratory for Developmental Genetics, RIKEN Research Center for Integrative Medical Sciences, Yokohama, Japan. Correspondence: Dr S Ogawa, Departments of Pathology and Tumor Biology, Kyoto University, Yoshida Konoe-cho, Sakyo-ku, Kyoto 606-8501, Japan. E-mail: [email protected] 6 These authors contributed equally to this work. 7 These authors jointly directed this work. Received 27 January 2014; accepted 10 February 2014; accepted article preview online 18 February 2014; advance online publication, 25 March 2014

Haploinsufficiency of Sf3b1 M Matsunawa et al

1845 and an OLYMPUS DP25 camera using DP2-BSW software (version 2.2; Olympus, Tokyo, Japan).

Table 1.

Peripheral blood counts of Sf3b1 þ / þ and Sf3b1 þ /  mice at 8

weeks

Colony-forming assays

Parameter 4

þ/

Sf3b1 þ / þ

Sf3b1 þ / 

N

P-value

1.32±0.56 13.67±6.40 80.58±7.14 5.75±1.54 18.23±1.37 110.63±22.68

1.09±0.28 15.83±6.28 79.08±6.50 5.08±2.84 17.61±1.00 119.30±18.44

6 6 6 6 6 6

0.39 0.57 0.71 0.62 0.39 0.48

þ/þ

Bone marrow (BM) cells (2.5  10 cells) from Sf3b1 or Sf3b1 mice at the age of 8 weeks were seeded into methylcellulose-containing medium (MethoCult M3234; Stem Cell Technologies, Vancouver, BC, Canada) supplemented with 10 ng/ml murine interleukin-3, 10 ng/ml murine interleukin-6, 20 ng/ml murine thrombopoietin, 50 ng/ml murine stem cell factor (Wako Pure Chemical, Osaka, Japan) and 3 U/ml human recombinant erythropoietin (R&D Systems, Minneapolis, MN, USA). The number of colonies was counted after 14 days of culture.

4

WBC count (  10 /ml) Neutrophil (%) Lymphocyte (%) Monocyte (%) Hb level (g/dl) PLT (  104/ml)

Abbreviations: Hb, hemoglobin; PLT, platelets; WBC, white blood cells. Data are the mean±s.d. (n ¼ 6).

Flow cytometry Measurement of hematopoietic stem cells (HSCs) and hematopoietic progenitor cells was conducted in 8-week-old male mice as previously described.13 Stained cells were analyzed with FACSAria II or FACSCanto II flow cytometers (BD Bioscience, Franklin Lakes, NJ, USA). Cell sorting was performed on a MoFlo system (Beckman Coulter, Fullerton, CA, USA). Data were analyzed by FlowJo software (Tree Star, Ashland, CA, USA). The antibodies used in this study are listed in the Supplementary Table 1.

Competitive repopulation assay Unfractionated pooled BM cells (1  106 cells) from 8-week-old Sf3b1 þ /  or Sf3b1 þ / þ mice (CD45.2 þ ) were transplanted into 8–12-week-old female CD45.1 þ recipient mice lethally irradiated at 2 doses of 4.9 Gy together with the same number of BM cells from 8–12-week-old male CD45.1 þ /CD45.2 þ mice as competitors. Sorted 120 CD34  KSL cells as well as 80 CD150 þ CD34  KSL cells obtained from Sf3b1 þ /  or Sf3b1 þ / þ mice were transplanted into lethally irradiated recipients together with competitor whole BM cells (5  105 cells). At 40 weeks after transplantation, BM cells (1  107) were harvested from the recipient mice and were serially transplanted into second recipients. The chimerism of donor-derived cells was evaluated by flow cytometry as previously described.14 The antibodies used in this study are listed in the Supplementary Table 1.

RESULTS Hematologic findings are not disturbed in Sf3b1 þ /  mice No Sf3b1-null mice were obtained, confirming the previous observation that Sf3b1  /  mice should be embryonic lethal. However, Sf3b1 þ /  mice were obtained at an expected frequency compared with Sf3b1 þ / þ littermates and appeared grossly normal.12 The complete peripheral blood counts in Sf3b1 þ /  mice at 8 weeks of age were comparable with those in Sf3b1 þ / þ littermate mice with normal differential counts of white blood cells (Table 1). There was no significant change in the peripheral blood counts between Sf3b1 þ / þ and Sf3b1 þ /  mice at any time points up to 54 weeks (Supplementary Figures 1a and b). Sf3b1 þ /  mice did not show any significant differences in total BM cellularity or the number of megakaryocytes, and their lineage composition was comparable with that of Sf3b1 þ / þ mice (Table 2). No splenomegaly was observed in any mice tested in these experiments, and spleen weights were also similar to those of Sf3b1 þ / þ mice (Table 2). No significant morphologic abnormalities were recognized in peripheral blood and BM cells in May– Gru¨nwald–Giemsa staining. Taken together, these findings suggested that steady-state hematopoiesis was maintained almost normally in Sf3b1 þ /  mice.

Gene expression analyses Total RNA was prepared from CD34  KSL cells, pooled from 3 female mice at the age of 11–13 weeks, using NucleoSpin RNA XS (Macherey-Nagel, Du¨ren, Germany). For RNA sequencing analyses, the synthesis and amplification of complementary DNA was done using a SMARTer Ultra Low RNA kit for Illumina sequencing (Clontech Laboratories, Mountain View, CA, USA) according to the manufacturer’s protocol. Sequencing libraries were generated using the NEBNext DNA Library Prep Reagent Set for Illumina (New England BioLabs, Ipswich, MA, USA) and analyzed using Illumina HiSeq 2000 (Illumina, San Diego, CA, USA) according to the manufacturer’s protocol. Data processing was performed as described previously.1,15 All sequence reads were mapped to the mouse transcriptome based on the UCSC known gene (downloaded in June 2013) using bowtie ver. 0.12.7 (http://bowtie-bio.sourceforge.net/index.shtml),16 and unmapped or poorly mapped reads were realigned to mouse reference genome (mm10) using BLAT (http://genome.ucsc.edu/).17 The expression level of each transcript was quantified with normalized fragments per kilobase of transcript per million fragments sequenced18 that were calculated using bedtools ver. 2.17.0 (https://code.google.com/p/bedtools/)19 with a transcriptome reference (RefSeq Genes, downloaded in June 2013). Gene set enrichment analyses (GSEA) were performed with GSEA20 ver. 2.0.13 software from the Broad Institute (http://www.broad.mit.edu/gsea). For quantitative reverse transcriptase-PCR, RNA was subjected to reverse transcription using the ReverTra Ace qPCR RT kit (Toyobo, Osaka, Japan) according to the manufacturer’s protocol. Quantitative expression levels of mRNA were measured as described previously.1,15 Primers used for quantitative reverse transcriptase-PCR are listed in the Supplementary Table 2.

Statistical analyses Statistical significance was evaluated by Student’s t-test, and Po0.05 was considered significant. & 2014 Macmillan Publishers Limited

The number of ring sideroblasts is not increased in Sf3b1 þ /  mice To determine whether loss of Sf3b1 alone can lead to increased production of ring sideroblasts, we examined cytospin BM specimens by Prussian blue staining for iron. In contrast to the previous report describing increased formation of ring sideroblasts in the same Sf3b1 þ /  strain of mouse,21 we observed very few sideroblasts characterized by nuclei encircled by a small number of iron granules. Specifically, there was no significant difference in the number of sideroblasts between Sf3b1 þ /  and Sf3b1 þ / þ mice at 8 weeks (Figures 1a and b) and also at 54 weeks (Supplementary Figure 1c). Decreased HSC fraction in Sf3b1 þ /  mice We evaluated HSCs and progenitor cells by flow cytometric analyses to further assess the hematopoietic system in the BM of 8-week-old mice. Interestingly, the frequency and the absolute number of HSCs, defined as CD34  /low c-Kit þ Sca1 þ Lineage  (CD34  KSL),22 were significantly decreased in Sf3b1 þ /  mice (Figures 2a–c). On the other hand, there were no significant differences in the number of hematopoietic progenitor cell fraction (CD34 þ KSL cells), megakaryocyte– erythroid progenitors, common myeloid progenitors or granulocyte–monocyte progenitors between Sf3b1 þ / þ and Sf3b1 þ /  mice (Figures 2a–f). SF3B1 mutations were also seen in a subset of chronic lymphocytic leukemia cases, but no obvious changes in the common lymphoid progenitor population in BM were observed (Figures 2g–i). Leukemia (2014) 1844 – 1850

Haploinsufficiency of Sf3b1 M Matsunawa et al

1846 Table 2.

Bone marrow cell counts and spleen size of Sf3b1 þ / þ and Sf3b1 þ /  mice at 8 weeks Sf3b1 þ / þ

Sf3b1 þ / 

N

P-value

5.04±1.62 65.42±4.35 17.08±4.34 13.58±3.17 3.92±2.85 7.00±2.00 54.64±9.83

6.20±0.84 66.67±4.36 12.58±3.43 18.33±5.17 2.42±1.88 8.33±1.75 64.64±14.10

6 6 6 6 6 6 6

0.15 0.63 0.07 0.08 0.31 0.27 0.24

Parameter 7

Total nucleated cells (  10 ) Myeloid (%) Erythroid (%) Lymphoid (%) Monocyte (%) Megakaryocyte (6  104 nucleated cells) Spleen weight (mg) Data are the mean±s.d. (n ¼ 6).

distribution of colony size or colony types were observed. These data also suggested that haploinsufficiency of Sf3b1 leads to a decrease in the number of HSCs/immature progenitor cells, although there was no significant difference in the numbers of differentiated or mature blood cells.

type I

Sf3b1 +/+

Sf3b1 +/-

Sf3b1 +/+

Sf3b1 +/-

type II

6 4 N.S.

N.S.

2

type I

type II

/-

3b 1 Sf +/+ 3b 1+

Sf

/-

3b 1 Sf +/+ 3b 1+

Sf

3b 1 Sf +/+ 3b 1+

/-

0 Sf

Number of sideroblasts in 1.0×103 erythroblasts

N.S.

ring sideroblast

Figure 1. The number of sideroblasts is not increased in Sf3b1 þ /  mice. (a) Representative images of BM cytospin slides from Sf3b1 þ / þ and Sf3b1 þ /  mice stained with Prussian Blue iron staining for the detection of sideroblasts. Sideroblasts were defined as follows: type I, sideroblasts with o5 siderotic granules in the cytoplasm; type II, sideroblasts with X5 siderotic granules, but not in a perinuclear distribution; and ring sideroblasts with X5 granules in a perinuclear position, surrounding the nucleus or encompassing at least one-third of the nuclear circumference. Original magnification  1000. (b) The number of cells with siderotic granules counted per 1.0  103 erythroblasts; n ¼ 6 mice per genotype. Data represent the mean±s.d. NS, not significant.

In addition, splenic B cell populations were not significantly changed between Sf3b1 þ / þ and Sf3b1 þ /  mice (Supplementary Figures 2a and b). Next, we performed in vitro colony-forming cell assays using whole BM cells. Consistent with the reduction in the HSC fraction in Sf3b1 þ /  mice, the number of hematopoietic colonies in Sf3b1 þ /  mice BM cells was significantly lower than that in Sf3b1 þ / þ mice (Figure 3a). No significant differences in the Leukemia (2014) 1844 – 1850

Reduced number and impaired function of HSCs in Sf3b1 þ /  mice Next, we assessed the reconstitution capacity of total BM cells from Sf3b1 þ /  mice using competitive repopulation assays. In these assays, 1.0  106 total BM cells from Sf3b1 þ / þ or Sf3b1 þ /  mice (CD45.1  /CD45.2 þ ) were transplanted into lethally irradiated recipient mice (CD45.1 þ /CD45.2  ) with the same number of competitor cells from CD45.1 þ /CD45.2 þ Sf3b1 þ / þ mice. Then, the chimerism of donor-derived CD45.1  /CD45.2 þ cells in the peripheral blood of recipient mice was measured by flow cytometry up to 40 weeks after transplantation. The chimerism of Sf3b1 þ /  -derived CD45.1  /CD45.2 þ cells in peripheral blood was significantly lower than that of Sf3b1 þ / þ derived cells (Figure 3b), suggesting the compromised hematopoietic repopulation capacity of Sf3b1 þ /  mice. To confirm this finding further, we performed competitive repopulation assays using purified HSC fractions (CD34  KSL cells; Figure 3b). Similar to the result of competitive reconstitution assay using whole BM cells, the chimerism of donor-derived CD45.2 þ cells in peripheral blood was also reduced in the mice transplanted with Sf3b1 þ /  mice-derived HSCs compared with that in mice transplanted with Sf3b1 þ / þ mice-derived HSCs. These observations suggested that the HSCs from Sf3b1 þ /  mice had significantly reduced reconstitution capacity compared with those from Sf3b1 þ / þ mice (Figure 3b). The lineage contribution of Sf3b1 þ /  cells in peripheral blood was comparable with that of Sf3b1 þ / þ cells (Figure 3b). These findings were confirmed by competitive repopulation assays using enriched long-term HSCs (CD150 þ CD34  KSL cells; Figure 3c).23,24 Furthermore, we performed serial transplantation experiments of whole BM and HSCs to assess the long-term reconstitution capacity of Sf3b1 þ /  HSCs more precisely. Sf3b1 þ /  mice showed reduced chimerism of donor-derived CD45.2 þ cells in the primary transplantations of competitive whole BM and competitive HSCs, and the reduced chimerism was even more pronounced after secondary transplantations (Figure 3d). In summary, HSCs in Sf3b1 þ /  mice reduce not only their number but also their competitive repopulation capacity of hematopoiesis. The effect of Sf3b1 haploinsufficiency on gene expression To investigate the molecular mechanisms of the impaired function of HSCs induced by Sf3b1 haploinsufficiency, we conducted gene expression analyses by RNA sequencing using CD34  KSL cells isolated from Sf3b1 þ /  and Sf3b1 þ / þ mice (Supplementary Table 3). Differentially expressed genes in Sf3b1 þ /  mice, including 1059 upregulated and 828 downregulated genes, from those of & 2014 Macmillan Publishers Limited

Haploinsufficiency of Sf3b1 M Matsunawa et al

1847

CD34+KSL Sf3b1 +/+ 105 2.52

104

103

103

102 0

102 0

0.002 0.000 +/ +

c-Kit

104

Sf3b1 +/− 3.16

0.05 0.00

Sf 3b 1 Sf 3b 1

105

0.004

0.10

0102 103 104 105

0.06

5.0

*

0.04 0.02 0.00

0.8 0.6 0.4 0.2 0.0 +/ −

Sca-1

0.006

N.S.

0.15

+/ +

0102 103 104 105

*

CD34+KSL N.S.

Sf 3b 1 Sf 3b 1

0102 103 104 105

0.008

+/ +

102 0

CD34-KSL

CD34+KSL

Sf 3b 1 Sf 3b 1

102 0

CD34-KSL Frequency (%) in BM nucleated cells

103

+/ −

103

Absolute number in BM (×107 cells)

0.0666

+/ −

0.12

+/ +

104

Sf 3b 1 Sf 3b 1

Sf3b1 +/−

104

Absolute number in BM (×107 cells)

105

Absolute number in BM (×107 cells)

Sf3b1 +/+

+/ −

105

Frequency (%) in BM nucleated cells

c-Kit

CD34

-KSL

0 102 103 104 105

Sca-1

2.0 1.0 − +/

+ +/

3b Sf 1 3b 1

Sf

+

−

+/

+/

Sf 1 3b 1

3b

Sf

Sf

3b Sf 1 3b 1

+/

+/

−

+

0.0

CLP N.S.

0.5 0.4 0.3 0.2 0.1

− +/

3b 1

+

Sf

3b 1

+/

0.0 Sf

Absolute number in BM (×107 cells)

MEP N.S.

3.0

−

Sf 1 3b 1

GMP N.S.

4.0

+/

+/

−

3b

+/

0.00 Sf

IL-7R

−

0 102 103 104 105

+/

0102 103 104 105

0.02

3b 1

102 0

+

102 0

0.04

+/

103

48.8

Sf

103

Sf3b1 +/−

0.06

3b 1

Flt-3

104

63.5

Sf

+/

3b Sf 1 3b 1

Sf Sf3b1 +/+ 105

104

CMP N.S.

CLP N.S.

CLP 105

+

0.0

CD34

+

0 102 103 104 105

0.2

+/

0102 103 104 105

CMP 28.1

3b Sf 1 3b 1

102 0

MEP N.S.

0.4

Sf

102 0

MEP 55.4

GMP 14.2

Frequency (%) in BM nucleated cells

CMP 29.8

103

103

Sf3b1 +/−

−

GMP 19.0

104

GMP N.S.

+/

MEP 48.3

CMP N.S.

+

FcgR

104

Frequency (%) in BM nucleated cells

Myeloid progenitor Sf3b1 +/+ 105

105

0.6

Figure 2. HSC population decreases in Sf3b1 þ /  mice. (a) Representative flow cytometric plot of HSCs (CD34  KSL cells) and hematopoietic progenitor cells (HPCs; CD34 þ KSL cells) from Sf3b1 þ / þ and Sf3b1 þ /  mice. (b) Frequency of CD34  KSL cells and CD34 þ KSL cells in Sf3b1 þ / þ and Sf3b1 þ /  mice BM. (c) Absolute number of CD34  KSL cells and CD34 þ KSL cells in Sf3b1 þ / þ and Sf3b1 þ /  mice BM. (d) Representative flow cytometric plot of common myeloid progenitors (CMPs), granulocyte–monocyte progenitors (GMPs) and megakaryocyte–erythroid progenitors (MEPs) in Sf3b1 þ / þ and Sf3b1 þ /  mice BM cells. CMPs are defined as Lin  cKIT þ Sca1  CD34 þ /lowFcgRint,  GMPs as Lin cKIT þ Sca1  CD34 þ FcgR þ and MEPs as Lin  cKIT þ Sca1  CD34  FcgR  . (e) Frequency of CMPs, GMPs and MEPs in Sf3b1 þ / þ and Sf3b1 þ /  mice BM. (f ) Absolute number of CMPs, GMPs and MEPs in Sf3b1 þ / þ and Sf3b1 þ /  mice BM. (g) Representative flow cytometric plot of common lymphoid progenitors (CLPs) in Sf3b1 þ / þ and Sf3b1 þ /  mice BM cells. CLPs are defined as Lin  ckitlow Sca1lowIL7R þ Flt3 þ . (h) Frequency of CLPs in Sf3b1 þ / þ and Sf3b1 þ /  mice BM. (i) Absolute number of CLPs in Sf3b1 þ / þ and Sf3b1 þ /  mice BM. Data are the mean±s.d.; n ¼ 6 per genotype. *Po0.05. NS, not significant.

wild-type mice were detected (Supplementary Figure 3a and Supplementary Tables 4 and 5). Expression of Sf3b1 in HSCs from Sf3b1 þ /  mice was reduced by B45% compared with that in Sf3b1 þ / þ HSCs, and this was confirmed by quantitative reverse transcriptase-PCR (Supplementary Figure 3b). Next, we performed the pathway analysis for differential expressed genes using GSEA, but GSEA identified no significant biological pathways, which explained the functional impairments in HSCs of Sf3b1 þ /  mice. & 2014 Macmillan Publishers Limited

DISCUSSION The SF3B1 mutation is one of the most frequent genetic alterations in myelodysplasia and is also found in some chronic lymphocytic leukemia cases. Hence, the physiological role of SF3B1 in the regulation of normal hematopoiesis provides an important clue to understand the role of SF3B1 mutations in the pathogenesis of hematopoietic malignancies having the SF3B1 mutation. Here, we showed Leukemia (2014) 1844 – 1850

Haploinsufficiency of Sf3b1 M Matsunawa et al

1848 Colony number / 2.5×104 BM cells

* 80 60 40 20 0 Sf3b1+/+ CD34−KSL

Whole BM 100

100

90

90 80

80 *

50

T cell B cell Myeloid

*

*

40 T cell B cell Myeloid

70

60

60

50

50

40

40

30

30

20

20

10

10

% chimerism

% chimerism

70

Sf3b1+/-

30

20

10

0

0

0

5 16 32 40

5 16 32 40

5 16 32 40

Sf3b1+/+

Sf3b1+/-

Sf3b1+/+

5 16 32 40 (weeks) Sf3b1+/-

% chimerism

100

90

90

80

80

70

70 *

50

4 9 16 (weeks) Sf3b1+/-

CD34−KSL

Whole BM 100

60

4 9 16 Sf3b1+/+

* *

60 **

50

40

40

30

30

20

20

10

10

0

0 1° 2°

1° 2°

Sf3b1+/+ Sf3b1+/-

1° 2°

1° 2°

Sf3b1+/+ Sf3b1+/-

Figure 3. Reduced HSC numbers and impaired HSC function in Sf3b1 þ /  mice. (a) Total number of colony-forming units generated from whole BM cells. The number of colonies was counted after 14 days of culture. Data are the mean±s.d.; n ¼ 4 per genotype; *Po0.05. (b) Competitive transplantation using whole BM or HSCs; the vertical axis represents the average peripheral blood (PB) chimerism of donor-derived CD45.2 þ cells. Data are the mean±s.d.; n ¼ 13 for whole BM transplantation, n ¼ 8 for HSC transplantation; *Po0.05. (c) Competitive transplantation using enriched long-term HSCs (CD150 þ CD34  KSL cells); the vertical axis represents the average PB chimerism of donor-derived CD45.2 þ cells. Data are shown as mean±s.d.; n ¼ 5, *Po0.05. (d) Serial transplantation assay; at 40 weeks post primary transplantation, BM cells harvested from the recipient mice were serially transplanted into additional secondary recipient mice. Chimerism of donor-derived CD45.2 þ cells in peripheral blood at 40 weeks after primary transplantation (11) and at 21 weeks after secondary transplantation (21). Data are the mean±s.d.; n ¼ 13 for primary whole BM transplantation, n ¼ 8 for primary HSC transplantation and n ¼ 5 for both secondary transplantation per group; *Po0.05, **Po0.01.

unequivocally that SF3B1 plays an important role in the regulation of HSCs. In the present study, Sf3b1 þ /  mice showed a significantly reduced number of HSCs and compromised reconstitution capacity of hematopoiesis, although the underlying molecular mechanism remained elusive. SF3B1 is known as a core component of the mRNA splicing machinery, and therefore a possible mechanism would be altered RNA splicing caused by Leukemia (2014) 1844 – 1850

haploinsufficiency of the genes involved in stem cell regulation. Unfortunately, we failed to identify plausible genes whose splicing was specifically altered in Sf3b1 þ /  mice. The other possibility is that the mechanism is not related to RNA splicing. In fact, Sf3b1 þ /  mice have been reported to show posterior transformation of vertebrae, and this phenomenon was ascribed to deregulation of the expression of Hox genes without any accompanying defects in Hox gene splicing. The deregulated Hox gene expression is & 2014 Macmillan Publishers Limited

Haploinsufficiency of Sf3b1 M Matsunawa et al

1849 thought to be caused by compromised Polycomb activities because of Sf3b1 haploinsufficiency.12 Polycomb group proteins are epigenetic transcriptional repressors with an important role in the regulation of hematopoiesis, and recent studies have shown that mutations in Polycomb group genes occur in hematological neoplasms (including MDS).25,26 Furthermore, most Hox genes are expressed in HSCs and immature progenitors and are downregulated during differentiation and maturation, and the critical role of Hox gene clusters in normal hematopoiesis has been demonstrated repeatedly in many literatures.27,28 For example, overexpression of Hoxb4 leads to expansion of HSCs, and Hoxb4-deficient mice exhibit a reduced reconstitution capacity similar to that seen in Sf3b1 þ /  mice.29 Similarly, Hoxa 9 and 10 are also reported to be functional regulators of HSCs.30,31 However, in our gene expression data based on RNA sequencing, expression of Hoxb4 was increased in HSCs from Sf3b1 þ /  mice compared with that from Sf3b1 þ / þ mice, and other clustered Hox genes, including Hoxa9 and 10, were not significantly changed (Supplementary Figure 3b). Thus, the exact mechanism of action that leads to reduced numbers of HSCs and their compromised function remains unclear. The other important issue to be discussed regarding the phenotype of Sf3b1 þ /  mice is the impact of Sf3b1 haploinsufficiency upon the formation of ring sideroblasts because the SF3B1 mutation is closely associated with MDS with increased numbers of ring sideroblasts. Although a previous study reported an increased frequency of ring sideroblasts in Sf3b1 þ /  mice,21 no increase in ring sideroblasts was demonstrated in the present study, even though both studies analyzed the identical Sf3b1 þ /  mouse strain.12 However, we considered that the lack of increased numbers of ring sideroblasts, together with the compromised repopulation capacity of Sf3b1 þ /  stem cells, should be rather expected, because most of the SF3B1 mutations thus far reported in MDS are clustered in the 5th–9th HEAT domains, mainly involving 5 hot spot amino-acid positions (K700 and, to a lesser extent, K666, H662 and E662), and no nonsense or frameshift changes have been reported, suggesting that these SF3B1 mutations will not lead to simple loss of function but should be associated with some gain of function.1,2,9 We could not exclude the possibility that an SF3B1 mutation acts as a dominant negative mutation that leads to more severe functional deficiency than haploinsufficiency, which would be responsible for tumorigenesis. However, our findings suggest that the simple haploinsufficiency of SF3B1 may not be responsible for the development of the MDS phenotype with increased formation of ring sideroblasts and other clonal disorders. In this regard, Sf3b1 þ /  mice may not be a suitable animal model for MDS, but further functional experiments using a conditional knock-in Sf3b1 mutant allele is required to understand the molecular mechanisms of SF3B1 mutations. CONFLICT OF INTEREST The authors declare no conflict of interest.

ACKNOWLEDGEMENTS We thank Y Yamazaki for his excellent technical support on flow cytometric analyses and cell sorting. This work was supported in part of Grants-in-aid for Scientific Research (nos. 24390242 and 23249052) and Grant-in-Aid for Scientific Research on Innovative Areas (no. 4201), MEXT, Japan, and grants from the Princes Takamatsu Cancer Research Fund and the Japan Leukemia Research Fund.

AUTHOR CONTRIBUTIONS MM, RY, MS, MO and HN performed mouse experiments; MM, AS-O, Y Shiozawa, KY, Y Shiraishi and SM performed bioinformatics analyses of RNA sequencing data; KI and HK generated Sf3b1 knockout mice; MM, RY, MS and AS-O analyzed

& 2014 Macmillan Publishers Limited

the data and generated figures and table; and MM, MS and SO designed the experiments and wrote the manuscript. All authors participated in the discussion and interpretation of data and results.

REFERENCES 1 Yoshida K, Sanada M, Shiraishi Y, Nowak D, Nagata Y, Yamamoto R et al. Frequent pathway mutations of splicing machinery in myelodysplasia. Nature 2011; 478: 64–69. 2 Papaemmanuil E, Cazzola M, Boultwood J, Malcovati L, Vyas P, Bowen D et al. Somatic SF3B1 mutation in myelodysplasia with ring sideroblasts. N Engl J Med 2011; 365: 1384–1395. 3 Graubert TA, Shen D, Ding L, Okeyo-Owuor T, Lunn CL, Shao J et al. Recurrent mutations in the U2AF1 splicing factor in myelodysplastic syndromes. Nat Genet 2011; 44: 53–57. 4 Visconte V, Makishima H, Jankowska A, Szpurka H, Traina F, Jerez A et al. SF3B1, a splicing factor is frequently mutated in refractory anemia with ring sideroblasts. Leukemia 2012; 26: 542–545. 5 Wang L, Lawrence MS, Wan Y, Stojanov P, Sougnez C, Stevenson K et al. SF3B1 and other novel cancer genes in chronic lymphocytic leukemia. N Engl J Med 2011; 365: 2497–2506. 6 Biankin AV, Waddell N, Kassahn KS, Gingras MC, Muthuswamy LB, Johns AL et al. Pancreatic cancer genomes reveal aberrations in axon guidance pathway genes. Nature 2012; 491: 399–405. 7 Stephens PJ, Tarpey PS, Davies H, Van Loo P, Greenman C, Wedge DC et al. The landscape of cancer genes and mutational processes in breast cancer. Nature 2012; 486: 400–404. 8 Harbour JW, Roberson ED, Anbunathan H, Onken MD, Worley LA, Bowcock AM. Recurrent mutations at codon 625 of the splicing factor SF3B1 in uveal melanoma. Nat Genet 2013; 45: 133–135. 9 Malcovati L, Papaemmanuil E, Bowen DT, Boultwood J, Della Porta MG, Pascutto C et al. Clinical significance of SF3B1 mutations in myelodysplastic syndromes and myelodysplastic/myeloproliferative neoplasms. Blood 2011; 118: 6239–6246. 10 Cazzola M, Rossi M, Malcovati L. Associazione Italiana per la Ricerca sul Cancro Gruppo Italiano Malattie Mieloproliferative. Biologic and clinical significance of somatic mutations of SF3B1 in myeloid and lymphoid neoplasms. Blood 2013; 121: 260–269. 11 Wahl MC, Will CL, Luhrmann R. The spliceosome: design principles of a dynamic RNP machine. Cell 2009; 136: 701–718. 12 Isono K, Mizutani-Koseki Y, Komori T, Schmidt-Zachmann MS, Koseki H. Mammalian polycomb-mediated repression of Hox genes requires the essential spliceosomal protein Sf3b1. Genes Dev 2005; 19: 536–541. 13 Ema H, Morita Y, Yamazaki S, Matsubara A, Seita J, Tadokoro Y et al. Adult mouse hematopoietic stem cells: purification and single-cell assays. Nat Protoc 2006; 1: 2979–2987. 14 Yamamoto R, Morita Y, Ooehara J, Hamanaka S, Onodera M, Rudolph KL et al. Clonal analysis unveils self-renewing lineage-restricted progenitors generated directly from hematopoietic stem cells. Cell 2013; 154: 1112–1126. 15 Sato Y, Yoshizato T, Shiraishi Y, Maekawa S, Okuno Y, Kamura T et al. Integrated molecular analysis of clear-cell renal cell carcinoma. Nat Genet 2013; 45: 860–867. 16 Langmead B, Trapnell C, Pop M, Salzberg SL. Ultrafast and memory-efficient alignment of short DNA sequences to the human genome. Genome Biol 2009; 10: R25. 17 Kent WJ. BLAT—the BLAST-like alignment tool. Genome Res 2002; 12: 656–664. 18 Mortazavi A, Williams BA, McCue K, Schaeffer L, Wold B. Mapping and quantifying mammalian transcriptomes by RNA-Seq. Nat Methods 2008; 5: 621–628. 19 Quinlan AR, Hall IM. BEDTools: a flexible suite of utilities for comparing genomic features. Bioinformatics 2010; 26: 841–842. 20 Subramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL, Gillette MA et al. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci USA 2005; 102: 15545–15550. 21 Visconte V, Rogers HJ, Singh J, Barnard J, Bupathi M, Traina F et al. SF3B1 haploinsufficiency leads to formation of ring sideroblasts in myelodysplastic syndromes. Blood 2012; 120: 3173–3186. 22 Osawa M, Hanada K, Hamada H, Nakauchi H. Long-term lymphohematopoietic reconstitution by a single CD34-low/negative hematopoietic stem cell. Science 1996; 273: 242–245. 23 Kiel MJ, Yilmaz OH, Iwashita T, Yilmaz OH, Terhorst C, Morrison SJ. SLAM family receptors distinguish hematopoietic stem and progenitor cells and reveal endothelial niches for stem cells. Cell 2005; 121: 1109–1121. 24 Morita Y, Ema H, Nakauchi H. Heterogeneity and hierarchy within the most primitive hematopoietic stem cell compartment. J Exp Med 2010; 207: 1173–1182.

Leukemia (2014) 1844 – 1850

Haploinsufficiency of Sf3b1 M Matsunawa et al

1850 25 Radulovic V, de Haan G, Klauke K. Polycomb-group proteins in hematopoietic stem cell regulation and hematopoietic neoplasms. Leukemia 2013; 27: 523–533. 26 Shih AH, Abdel-Wahab O, Patel JP, Levine RL. The role of mutations in epigenetic regulators in myeloid malignancies. Nat Rev Cancer 2012; 12: 599–612. 27 Argiropoulos B, Humphries RK. Hox genes in hematopoiesis and leukemogenesis. Oncogene 2007; 26: 6766–6776. 28 Alharbi RA, Pettengell R, Pandha HS, Morgan R. The role of HOX genes in normal hematopoiesis and acute leukemia. Leukemia 2013; 27: 1000–1008.

29 Brun AC, Bjornsson JM, Magnusson M, Larsson N, Leveen P, Ehinger M et al. Hoxb4-deficient mice undergo normal hematopoietic development but exhibit a mild proliferation defect in hematopoietic stem cells. Blood 2004; 103: 4126–4133. 30 Lawrence HJ, Christensen J, Fong S, Hu YL, Weissman I, Sauvageau G et al. Loss of expression of the Hoxa-9 homeobox gene impairs the proliferation and repopulating ability of hematopoietic stem cells. Blood 2005; 106: 3988–3994. 31 Magnusson M, Brun AC, Miyake N, Larsson J, Ehinger M, Bjornsson JM et al. HOXA10 is a critical regulator for hematopoietic stem cells and erythroid/ megakaryocyte development. Blood 2007; 109: 3687–3696.

Supplementary Information accompanies this paper on the Leukemia website (http://www.nature.com/leu)

Leukemia (2014) 1844 – 1850

& 2014 Macmillan Publishers Limited