In bone marrow, special microenvironments, known as niches, are essential for the maintenance of hematopoietic stem cells (HSCs). A population of mesenchymal stem cells, termed CXC chemokine ligand 12 (CXCL12)-abundant reticular (CAR) cells or leptin receptor-expressing cells are the major cellular component of HSC niches. The molecular regulation of HSC niche properties is not fully understood. The role of Runx transcription factors, Runx1 and Runx2 in HSC cellular niches remains unclear. Here we show that Runx1 is predominantly expressed in CAR cells and that mice lacking both Runx1 and Runx2 in CAR cells display an increase in fibrosis and bone formation with markedly reduced hematopoietic stem and progenitor cells in bone marrow. In vitro, Runx1 is induced by the transcription factor Foxc1 and decreases fibrotic gene expression in CAR cells. Thus, HSC cellular niches require Runx1 or Runx2 to prevent their fibrotic conversion and maintain HSCs and hematopoiesis in adults.
In the adult bone marrow, special microenvironments, known as niches, are essential for the maintenance of hematopoietic stem cells (HSCs) that sustain hematopoiesis and the production of blood cells, including immune cells1,2,3. HSCs are in contact with their niches that provide critical signals. The identity of HSC niches has been a subject of longstanding debate, but recent studies have demonstrated that a population of self-renewing mesenchymal stem cells, termed CXC chemokine ligand 12 (CXCL12)-abundant reticular (CAR) cells, which overlap strongly with leptin receptor-expressing (LepR)+ cells, are the major cellular component of HSC niches in the bone marrow, giving rise to adipocytes and osteoblasts4,5,6,7,8,9,10,11,12. However, the molecular regulation of CAR/LepR+ cell development and properties is not fully understood.
CAR cells are the major producer of hematopoietic cytokines, including CXCL12 and stem cell factor (SCF), which are critical for the maintenance of hematopoietic stem and progenitor cells (HSPCs)6,8,10,13,14. In addition, the transcription factors, Foxc1 and Ebf1/3 are specifically expressed in CAR cells and play a critical role in the formation and maintenance of HSC niches, inhibiting adipocyte and osteoblast differentiation of CAR cells, respectively9,15. In contrast, CAR cells have much lower levels of fibrotic gene expression compared with fibrogenic cells in primary myelofibrosis16; however, the transcription factors involved in this remain unclear.
We have previously shown that CAR cells express high levels of the Runx transcription factor Runx2 (ref. 8), prompting us to investigate the role of Runx transcription factors in formation and maintenance of HSC niches. Members of the Runx transcription factors share a highly conserved DNA binding region with the Drosophila pair-rule gene runt, which has the ability to make contact with the same DNA motif. Runx1 is known to be an important regulator of HSPCs. In mouse embryos, Runx1 is expressed in hemogenic endothelial cells and mice lacking Runx1 have no HSCs, indicating that Runx1 is essential for the establishment of definitive hematopoiesis in hemogenic endothelial cells17,18. In adults, it has been shown that loss of Runx1 leads to an increase in HSCs, myeloid progenitors with upregulation of stem cell and megakaryocytic transcription programs, and aberrant megakaryocyte progenitors and a decrease in B cell progenitors and megakaryocytes in the bone marrow as well as T cell progenitors in the thymus19,20,21. Clinically, Runx1 is one of the most frequently mutated genes in hematopoietic cells in hematological malignancies22. Somatic mutations and chromosomal rearrangements involving Runx1 are frequently observed in abnormal hematopoietic cells in myelodysplastic syndrome (MDS), acute myeloid leukemia (AML), acute lymphoblastic leukemia (ALL), and chronic myelomonocytic leukemia (CMML) patients22. On the other hand, the expression of Runx2 occurs at sites of bone formation early during skeletal development23,24,25. Embryos deficient for Runx2 have an almost perfectly patterned skeleton composed entirely of cartilage but lack bones24,25, indicating that Runx2 is essential for generation of osteoblasts. In adults, loss of Runx2 in hematopoietic cells leads to a decrease in plasmacytoid dendritic cells that mediate type I interferon responses to viral infection26.
Here, we have shown that Runx1 is highly and predominantly expressed in CAR cells and that mice lacking both Runx1 and Runx2 in CAR cells exhibit a marked reduction in HSPCs and increased bone formation and fibrosis in the bone marrow. These findings indicate that hemogenic Runx1 or osteogenic Runx2 inhibits fibrotic conversion of marrow mesenchymal stem cells and maintain HSC niches.
Runx1 is predominantly expressed in CAR cells in the marrow
We analyzed relative mRNA expressions of Runx1 in adult bone marrow cell populations. Quantitative real-time polymerase chain reaction with reverse transcription (qRT-PCR) analysis of 15-week-old CXCL12-GFP mice showed that expression levels of Runx1 mRNA in CXCL12-GFPhi CAR cells were much higher than those in other bone marrow nonhematopoietic cell populations, including alkaline phosphatase (ALP)hiCXCL12-GFPlo osteoblasts, Sca-1+CD31+ endothelial cells, and platelet-derived growth factor receptor α (PDGFRα)+Sca-1+CD45-Ter119- (PαS) cells27, as well as hematopoietic cell populations, including CD34-CD150+CD48-Lin-Sca-1+c-kit+ cells, which are highly enriched for long-term repopulating HSCs (LT-HSCs), c-kit+CD19+IgM- pro-B cells, c-kit-CD19+IgM- pre-B cells, and F4/80+ macrophages (Fig. 1a). Transcription of the Runx1 gene is under the control of two alternative promoters, which generate distal and proximal Runx1 transcripts that produce Runx1c and Runx1b protein isoforms, respectively28. By qRT-PCR, we found that Runx1b mRNA was dominant in CAR cells whereas Runx1c was dominant in hematopoietic cells, including LT-HSCs and pre-B cells (Fig. 1b). We next analyzed relative mRNA expressions of Runx1 as well as Runx2, which was shown to be abundantly expressed in CAR cells8, in fetal bone marrow cell populations. qRT-PCR analysis of embryonic day (E) 16.5 transgenic mice expressing the green fluorescent protein (GFP) reporter gene under the control of Osterix (Osx/Sp7) regulatory elements (Osx-GFP) mice showed that expression levels of Runx1 and Runx2 in CAR cell progenitors15 were similar to those in Osx-GFP+ osteoblast progenitors but higher than those in other nonhematopoietic cell populations (Fig. 1c).
HSC niches are formed in mice lacking Runx1 or Runx2
To examine the role of Runx1 or Runx2 in CAR cell development, we analyzed mice with conditional floxed Runx1 or Runx2 alleles (Runx1f/f or Runx2f/f mice) in conjunction with transgenic mice expressing Cre recombinase under the control of Prx1 regulatory elements (Prx1-Cre mice)29, in which Runx1 or Runx2 is deleted in all mesenchymal cells in developing limbs (Prx1-Cre;Runx1f/f or Prx1-Cre;Runx2f/f mice). Prx1-Cre;Runx1f/f mice were viable30 and showed normal development of bone, bone marrow, and CAR cells at 3 weeks of age (Fig. 2a and Supplementary Fig. 1). On the other hand, Prx1-Cre;Runx2f/f mice were born and viable but exhibited a severe reduction in the length of all limbs compared to control animals at 3 weeks of age. Histological analysis of femurs from Prx1-Cre;Runx2f/f mice crossed to mice with GFP reporter gene knocked into the CXCL12 locus (CXCL12-GFP mice) showed that there existed bone-like structures, which appear as woven bone and surround the bone marrow cavity (Fig. 2b). Although bone marrow cavity was reduced in length, CXCL12-GFP+ CAR cells, which exhibited a normal morphology with long processes, were present and their density appeared to be normal in the marrow cavity (Fig. 2c). Bone marrow volume and the total hematopoietic cell counts were reduced in the mutant marrow. However, by flow cytometry, there were no significant differences in frequencies of LT-HSCs, Lin-Sca-1-c-kit+CD34-FcγRII/IIIlo megakaryocyte/erythrocyte progenitors (MEPs), c-kit+CD71+Ter119lo proerythroblasts, Lin-Sca-1-c-kit+CD34+FcγRII/IIIhi granulocyte/macrophage progenitors (GMPs), Lin-IL-7Rα+Flt3+ common lymphoid progenitors (CLPs), or pro-B cells between mutants and control animals. The frequencies of pre-B cells were only slightly reduced in the mutants (Fig. 2d). By qRT-PCR, CAR cells from the mutants had relatively normal expression of Runx1, platelet-derived growth factor receptor β (PDGFRβ), LepR, CXCL12, SCF (Kitl) Foxc1, Ebf1, and Ebf3 (Supplementary Fig. 2). They had slightly reduced expression of Interleukin-7 (IL-7) (ref. 13) and much lower expression of Osx in CAR cells compared with control animals (Supplementary Fig. 2). Mice lacking Osx, genetically downstream of Runx2 (ref. 31), (Prx1-Cre;Osxf/f mice) showed similar phenotypes to Runx2 mutants (Supplementary Fig. 3).
HSC niches are maintained in mice lacking Runx1 or Runx2
To examine the role of Runx1 or Runx2 in CAR cell maintenance, we next crossed Runx1f/f or Runx2f/f mice with mice expressing the CreERT2 transgene under the control of the endogenous Ebf3 locus, in which Cre recombinase can be activated in CAR cells but not in other bone marrow hematopoietic or nonhematopoietic populations upon tamoxifen treatment (Ebf3-CreERT2 mice)9. Ebf3-CreERT2;Runx1f/f or Ebf3-CreERT2;Runx2f/f mice were subjected to a tamoxifen pulse for a week beginning at 10 weeks of age and analyzed 10–14 weeks after tamoxifen treatment. Flow cytometric analysis showed that the frequencies of LT-HSCs and the numbers of LT-HSCs, MEPs, c-kit+CD71+Ter119lo proerythroblasts, GMPs, and CLPs in the bone marrow from tamoxifen-treated Ebf3-CreERT2;Runx1f/f or Ebf3-CreERT2;Runx2f/f mice were similar to those in control animals (Supplementary Fig. 4a, d).The numbers of pro-B cells and pre-B cells were unaltered in tamoxifen-treated Ebf3-CreERT2;Runx1f/f mice (Supplementary Fig. 4a) but slightly reduced in tamoxifen-treated Ebf3-CreERT2;Runx2f/f mice compared with control animals (Supplementary Fig. 4d). qRT-PCR analysis of the mutants showed that CAR cells had relatively normal expression of PDGFRβ, LepR, CXCL12, SCF, Ebf1, Ebf3, which are preferentially expressed in CAR cells, as well as fibrotic genes, including type I collagen α1 (Col1a1), type III collagen α1 (Col3a1), and type VI collagen α3 (Col6a3), and reduced expression of Osx (Supplementary Fig. 4b, c, e, f). Expression of Foxc1 and IL-7 was unaltered in tamoxifen-treated Ebf3-CreERT2;Runx1f/f mice (Supplementary Fig. 4b) but slightly reduced in tamoxifen-treated Ebf3-CreERT2;Runx2f/f mice (Supplementary Fig. 4e). Mice lacking Osx (Ebf3-CreERT2;Osxf/f mice) showed similar phenotypes to Runx1 mutants (Supplementary Fig. 5).
Mice lacking Runx1 and Runx2 in CAR cells have reduced HSPCs
Since the expression of Runx2 or Runx1 was unaltered in CAR cells in the bone marrow from tamoxifen-treated Ebf3-CreERT2;Runx1f/f or Ebf3-CreERT2;Runx2f/f mice, respectively (Supplementary Fig. 4b, e), we analyze mice, in which both Runx1 and Runx2 were deleted from CAR cells. Prx1-Cre;Runx1f/fRunx2f/f mice die perinatally with protruded internal organs due to the absence of sternum in contrast to the normal sternal fusion in Prx1-Cre;Runx2f/f mice30. Histological analysis showed that femurs from Prx1-Cre;Runx1f/fRunx2f/f mice lacked bone and bone marrow like Prx1-Cre;Runx2f/f mice at E18.5 (ref. 30). We, therefore, generated Ebf3-CreERT2;Runx1f/fRunx2f/f mice, in which both Runx1 and Runx2 were deleted from CAR cells upon tamoxifen treatment. These mice were subjected to a tamoxifen pulse for a week beginning at 10 weeks of age and analyzed 10–14 weeks after tamoxifen treatment. Flow cytometric analysis showed that the frequencies of LT-HSCs were unaltered but the total hematopoietic cell counts and numbers of phenotypic LT-HSCs, MEPs, proerythroblasts, Lin-Sca-1-c-kit+CD150+CD41+ megakaryocyte-committed progenitors (MkPs)32, GMPs, Gr-1hiCD11b+ granulocytes, CLPs, pro-B, pre-B cells, B220hiIgD+ mature B cells, plasmacytoid dendritic cells (pDCs), and NK cells were markedly reduced in the bone marrow of mutants compared with control animals (Fig. 3a). Additionally, we estimated the numbers of functional HSCs using repopulating units (RU), based on a competitive repopulation assay and found that RUs were markedly reduced in the mutant bone marrow (Fig. 3b). Histological analysis showed that numbers of morphologically identifiable CD41+CD31+ megakaryocytes per high powered field from bone marrow were modestly reduced in the mutants (Fig. 3c). Furthermore, the numbers of phenotypic LT-HSCs, MEPs, and GMPs were increased in the spleen of the mutants (Fig. 3d), suggesting that impaired hematopoiesis in the bone marrow induced extramedullary hematopoiesis. Peripheral blood (PB) platelet counts were unaltered in the mutants (Fig. 3e).
Flow cytometric analysis showed that CAR cells had increased cell surface expression of Sca-1 in the bone marrow of tamoxifen-treated Ebf3-CreERT2;Runx1f/fRunx2f/f mice (Fig. 3f). qRT-PCR analysis showed that CAR cells from the mutants displayed reduced expression of Osx, PDGFRβ, LepR, and key HSC niche factors, including CXCL12, SCF, IL-7, Foxc1, Ebf1, and Ebf3, compared with control animals (Fig. 3g). Of note, some mutants had a modest reduction in CXCL12, SCF, and Ebf1 expression in CAR cells but had a marked reduction in HSC number and CAR cells from the mutants displayed markedly increased expression of fibrotic genes, including Col1a1, Col3a1, Col6a3, and Gli1 (ref. 33) as well as podoplanin (PDPN), which is highly expressed in fibroblastic reticular cells in lymph nodes34, and slightly increased expression of PDGFRα, which is required for bone marrow fibrosis16 (Fig. 3h). These results indicate that CAR cells undergo fibrotic differentiation and their HSPC niche function is impaired in the absence of Runx1 and Runx2.
Mice lacking Runx1 and Runx2 in CAR cells have increased fibrosis in the marrow
Although embryos lacking Runx2 have no bones, histological analysis of the bone marrow from tamoxifen-treated Ebf3-CreERT2;Runx1f/fRunx2f/f mice showed that the trabecular bone numbers in the metaphysis were increased without any cartilaginous structure compared with control animals (Fig. 4a). Micro-computed tomography (μCT) analyses confirmed that the trabecular bone mass was increased and marrow area was slightly reduced in the long bones of the mutants (Fig. 4b). The numbers of cells lining the bone surface and expressing an osteoclast marker tartrate-resistant acid phosphatase (TRAP) were unaltered in the marrow sections of the mutants (Fig. 4c). Additionally, serum levels of bone resorption marker carboxy-terminal telopeptides of type I collagen (CTX-I), which reflect the whole-body amount of cleaved type I collagen by osteoclasts, were unaltered in the mutants (Fig. 4d). These results suggest that increased bone formation in the mutants does not result from defects in osteoclasts.
Histological analysis of the bone marrow from tamoxifen-treated Ebf3-CreERT2;Runx1f/fRunx2f/f;CXCL12-GFP mice showed that CXCL12-GFPhi cells, which exhibited a CAR cell-like morphology, were present and their numbers were increased about 2-fold compared with those of CXCL12-GFPhi CAR cells in control animals (Fig. 4e). The protein expression of CXCL12 and SCF was relatively normal in the mutants, presumably because of a decrease in their mRNA expression in CAR cells and an increase in the numbers of CAR cells (Supplementary Fig. 6). Importantly, in consistent with the increased expression of fibrotic genes in CAR cells lacking Runx1 and Runx2, silver staining showed massive fibrosis in bone marrow sections from the mutants whereas no significant reticulin deposition was detected in control animals (Fig. 4f). In addition, immunohistochemical analysis with antibodies against Col3a1 revealed that CXCL12-GFPhi CAR cells and CXCL12-GFP- cells expressing high levels of Col3a1 protein were almost absent in control animals but were markedly increased in the mutants (Fig. 4g). These fibrotic phenotypes were not observed in Runx1- or Runx2-deficient mice. Histological analysis showed that the numbers of adipocytes were increased in the bone marrow of the mutants (Fig. 4h).
Runx1 or Runx2 decreased fibrotic gene expression in vitro
To examine the in vitro functions of Runx1 or Runx2 in fibrotic gene expression in CAR cells, we infected sorted CAR cells with retroviruses expressing Runx1 or Runx2. The enforced expression of Runx1 or Runx2 in CAR cells increased mRNA expression of Col1a1 and decreased mRNA expression of fibrotic genes, including Col3a1, Col6a3, and PDPN as well as Sca-1 (Ly6a) but did not alter Foxc1 expression based on qRT-PCR in the culture (Fig. 5a). We next infected sorted CAR cells with retroviruses expressing Runx1 or Runx2 in addition to Ebf3, which inhibits osteoblast differentiation of CAR cells9. The enforced expression of Ebf3, Ebf3 and Runx1, or Ebf3 and Runx2 in CAR cells decreased the expression of Col1a1 and Col3a1 as assessed by qRT-PCR in the culture (Fig. 5b). Cultured CAR cells from tamoxifen-treated Ebf3-CreERT2;Runx1f/fRunx2f/f mice had an increase in the expression of fibrotic genes than that of control CAR cells in the culture (Fig. 5a). The increase was modest presumably because of reduction in the expression of Runx1 and Runx2 in control CAR cells after in vitro culture. The enforced expression of Runx1 or Runx2 in mutant CAR cells reduced the expression of fibrotic genes to normal or lower levels in the culture (Fig. 5a). It has been previously shown that PDGFRα in CAR/LepR+ cells is involved in bone marrow fibrosis16. We found that PDGFs reduced the expression of Runx1 and Runx2 in CAR cells in culture (Fig. 5c).
We next analyze whether Foxc1, which is essential for formation and maintenance of HSC niches in CAR cells15, induce Runx1 or Runx2 expression in CAR cells. The enforced expression of Foxc1 markedly increased mRNA expression of Runx1 but not Runx2, based on qRT-PCR in the culture (Fig. 5d). Considering that Foxc1 but not Runx1 is predominantly expressed in CAR cell progenitors in E16.5 embryos (Fig. 1c)15, Foxc1 might enhance the expression of Runx1 in CAR cells during development.
Runx1 and Runx2 are decreased in CAR cells in a PMF model
Primary myelofibrosis (PMF) patients demonstrate myelofibrosis during the course of their disease35. We thus examine expression levels of Runx1 and Runx2 in CAR cells in PMF, using a PMF mouse model where HSPCs infected with retroviruses expressing thrombopoietin (TPO) were transplanted into irradiated mice16. Recipients of TPO transduced HSPCs developed PMF-like disease 8–9 weeks after transplantation. Immunohistochemical analysis with antibodies against CD41 revealed that megakaryocyte lineage cells were increased (Fig. 6a). Silver staining showed massive fibrosis in bone marrow sections from PMF mice, whereas no significant reticulin deposition was detected in control animals (Fig. 6b). Immunohistochemical analysis with antibodies against Col3a1 revealed that cells expressing high levels of Col3a1 protein were almost absent in control animals but were markedly increased in PMF mice (Fig. 6c). qRT-PCR analysis showed that CAR cells from PMF mice displayed reduced expression of Runx1 and Runx2 as well as key HSC niche factors, including CXCL12, SCF, Foxc1, Ebf1, and Ebf3, compared with control animals (Fig. 6d). In consistent with massive fibrosis in bone marrow sections, CAR cells displayed markedly increased expression of fibrotic genes, including Col1a1, Col3a1, Col6a3, and Gli1 in PMF mice (Fig. 6e).
We show that Runx1 and Runx2 are highly expressed in CAR/LepR+ cells in the bone marrow. Mice lacking Runx2 failed to generate osteoblasts and showed delayed invasion of blood vessels into hypertrophic cartilage and bone marrow formation probably due to defective development of chondrocytes during endochondral bone development36,37,38. However, we show that CAR cells were largely unaltered in the absence of Runx2 (ref. 39) or Osterix, which is genetically downstream of Runx2. Considering that osteoblasts and primordial CAR cells are generated from heterogeneous perichondrial cells during embryogenesis15,40, these results indicate that Runx2 is essential for the development of osteoblasts in the fetus but not for the development and maintenance of CAR cells.
In contrast, CAR cell-specific deletion of both Runx1 and Runx2 induced massive fibrosis with increased expression of fibrotic genes in CAR cells in the bone marrow. Consistent with this, enforced expression of Runx1 or Runx2 decreased the expression of fibrotic genes, including Col3a1 and Col6a3, in CAR cells in vitro. Runx1/2 overexpression increased expression of a fibrotic gene Col1a1 in cultured CAR cells, in which Ebf3 expression was much lower than that in bone marrow CAR cells; however, enforced expression of Ebf3 blocked this effect and reduced the expression of Col1a1 in cultured CAR cells (Fig. 5b). Thus, in the absence of Runx1 and Runx2, Col1a1 expression was increased presumably because of a decrease in Ebf3 expression in mutant CAR cells in the marrow. Together, Runx1/2 are required for preventing fibrotic conversion of CAR cells and they would act in a cell-autonomous fashion (Fig. 5e).
The stromal cells responsible for fibrosis are thought to be activated by the abnormal megakaryocytes in myelofibrosis41. Previous studies have shown that genetic ablation of Runx1 in hematopoietic cells caused an increase in HSCs, GMPs with upregulation of stem cell and megakaryocytic transcription programs, and aberrant progenitors of megakaryocytes as well as a reduction in PB platelet counts, indicating that Runx1 in HSPCs is involved in megakaryocytic differentiation19,20,21,42. In contrast, megakaryocytes cell size appeared normal, frequencies of megakaryocytes were modestly reduced, and numbers of HSCs and MkPs were markedly reduced in tamoxifen-treated Ebf3-CreERT2;Runx1f/fRunx2f/f mice. In addition, expression of PDGFs and TGF-β1, which are suggested to be produced by megakaryocytes and induce myelofibrosis in PMF mouse models16,43, in HSPCs, including MkPs, is unaltered in the mutants (Supplementary Fig. 7). Although we cannot entirely exclude the possibility that megakaryocytes are altered and affect CAR cells in the mutants, these results suggest that megakaryocytes and their progenitors are not responsible for myelofibrosis in the mutants, supporting the idea that effects of loss of function of Runx1/2 in CAR cells on the marrow fibrotic phenotype are cell autonomous.
Even after adjusting for bone marrow volume, the numbers of HSPCs and immune cells were markedly reduced in mice lacking Runx1 and Runx2 in CAR cells (Figs. 3a and 4b), indicating that Runx1/2 play a critical role in the maintenance of niches for HSCs and hematopoiesis. There is the possibility that increased fibrotic proteins, including Col1a1, Col3a1, and Col6a3, in CAR cells prevent CAR cells from providing critical signals to HSPCs and immune cells. Additionally, defective hematopoiesis in the absence of Runx1/2 would be partially attributed to decreased expression levels of CXCL12, SCF, Foxc1, Ebf1, and Ebf3, which are essential for the maintenance of HSPCs6,9,15,44. However, some Ebf3-CreERT2;Runx1f/fRunx2f/f mice had a modest reduction in CXCL12, SCF, Foxc1, Ebf1, and Ebf3 expression in CAR cells but had a marked increase in fibrotic gene expression and a marked reduction in HSC numbers. This result and the finding that CAR cell numbers were doubled in these mutants, which might be beneficial for HSCs, support the idea that defects in HSC niche functions other than reduced expression of HSC niche factors in CAR cells caused a marked reduction in HSC numbers in these mutants. How fibrotic conversion of CAR cells affects HSCs and hematopoiesis is an important question for the future.
Unexpectedly, mice lacking both Runx1 and Runx2 in adult CAR cells displayed an increase in bone formation as well as adiposity in the marrow. Considering that the majority of CAR cells remained undifferentiated in these mutants, this raises the possibility that Runx1/2 exert inhibitory effects on sensitivity to unknown extracellular and/or intracellular cue that stimulates osteoblast or adipocyte differentiation of CAR cells in adult bone marrow.
Foxc1 and Ebf1/3, which are specifically expressed in CAR cells and critical for their HSPC niche functions9,15, were reduced but detectable in CAR cells lacking both Runx1 and Runx2. However, enforced expression of Runx1 did not increase expression of Foxc1 and Ebf3. Considering that enforced expression of Foxc1 enhanced expression of Runx1, Foxc1 might enhance the expression of Runx1 in CAR cells (Fig. 5e). How expression of Runx1/2, Foxc1, and Ebf1/3 is enhanced in CAR cells is an important question for the future.
Expression levels of Runx1 are much higher in CAR cells than those in osteoblasts, although expression levels of Runx2 are similar between these cell populations. Moreover, HSPC niche functions of CAR cells were dramatically altered in the absence of both Runx1 and Runx2 but largely unaltered in the absence of Runx2, whereas fetal osteoblasts were depleted in the absence of Runx2. Thus, Runx1 compensates for the inactivation of Runx2 in CAR cells but not in fetal osteogenic progenitors.
Clinically, it is known that myelofibrosis is driven by mutations in HSPCs that alter the proliferation properties of the HSPCs. Our results raise the possibility that hematopoietic cells with the driver mutations found in myelofibrosis act on HSC cellular niches (CAR cells) to reduce their Runx1/2 expression, resulting in an increase in myelofibrosis. Consistent with this, CAR cells from PMF mice with myelofibrosis displayed reduced expression of Runx1 and Runx2 compared with control animals (Fig. 6c). In addition, PDGFs, which induce myelofibrosis in the PMF mouse model16, reduced expression of Runx1 and Runx2 in cultured CAR cells. We analyzed available data sets comparing gene expression in bone marrow cells from PMF patients with healthy donors45. Although expression of RUNX1 and RUNX2 was unaltered, expression of fibrotic genes, including COL1A1, COL3A1, COL6A3, and GLI1 as well as PDPN, was also unaltered in bone marrow cells from PMF patients, suggesting that the proportion of fibrogenic cells and/or fibroblasts was not high in the cryopreserved cells from the bone marrow provided as samples from PMF patients45. Thus, it will be important to analyze expression of RUNX1 and RUNX2 in fibrogenic cells as well as CAR cells46 of PMF patients to understand pathogenesis of fibrosis; however, it is technically difficult to obtain informative bone marrow fibrogenic cells and fibroblasts from PMF patients. TGF-β signaling activated in marrow stromal cells has been hypothesized to drive the progression of Philadelphia translocation (Ph)-negative myeloproliferative neoplasms (MPN) to PMF43,45. If so, although it has been reported that TGF-β signaling pathways transcriptionally activate Runx proteins in tumor microenvironments47, our results raise the possibility that TGF-β signaling might inhibit the Runx1/2 expression and/or function in CAR cells to induce myelofibrosis. Thus, this study proposes a mechanism implicated in the pathogenesis of myelofibrosis; however, further studies are needed to address this issue.
On the other hand, RUNX1 mutations are observed in the malignant stem cell clones of a minority of myelofibrotic patients, and only when these patients have progressed to AML35 and enforced expression of Runx2 in HSPCs has been shown to induce AML in mouse48. In addition, RUNX1 and/or RUNX2 mutations in bone marrow nonhematopoietic cells, including CAR cells, have not been found in PMF patients. Thus, mutations or chromosomal rearrangements involving Runx1 and/or Runx2 would not contribute to progression of myelofibrosis.
For clinical application, our results raise the possibility that activators of CAR cell-specific RUNX1 can decrease bone marrow fibrosis, which might be applied to non-cell-autonomous therapies targeting the HSC niches, in hematological disorders, including MPN. In addition, Runx1 activators might enhance the formation of HSPC niches, which support ex vivo hematopoiesis, from non-niche cells, including dermal fibrogenic cells or induced pluripotent stem cell (iPSC)-derived fibrogenic cells in vitro. This study provides significant advances in our understanding of mesenchymal cells providing HSPC niches and bone.
Male and female mice were used between 3–24 weeks of age or at the age of embryonic day 16.5, depending on the respective experiments. C57BL/6 mice were purchased from Japan SLC, Inc. Runx1f/f mice30, Sp7f/f mice37, Ebf3-CreERT2 knock-in mice9, CXCL12-GFP knock-in mice49, and Osx-GFP knock-in mice50 have been previously described. Runx2f/f mice were generated with two loxP sites flanking exon 4 of Runx2 by electroporation of a targeting vector into embryonic stem cells. These mice were backcrossed at least six times onto a C57BL/6 background. Prx1-Cre mice29 were obtained from The Jackson Laboratory. To induce CreERT2-mediated recombination, mice were injected i.p. with 2 mg tamoxifen (Sigma) four times every other day, and analyzed 10–14 weeks after the initial tamoxifen injection. Peripheral blood was measured on a hematology analyzer Celltac-α (Nihon Kohden). All mice were bred and maintained under specific pathogen-free conditions at the animal facilities of Osaka University. These mice were maintained in 12 h light/dark cycle, and the housing temperature and humidity were 23 ± 1.5 °C and 45 ± 15%, respectively. All animal experiments were performed in accordance with approved protocols of the Institutional Animal Care and Use Committees at Osaka University and Kyoto University.
Monoclonal antibodies were purchased from BioLegend, BD Bioscience, eBioscience, or Miltenyi Biotec, unless otherwise noted: B220-PE/Cy5 (1:800, RA3-6B2, 15-0452-83, eBioscience), B220-PE (1:200, RA3-6B2, 103208, Biolegend), B220-PB (1:100, RA3-6B2, 103227, Biolegend), CD3ε-PE/Cy5 (1:400, 145-2C11, 100310, Biolegend), CD3ε-APC (1:100, 145-2C11, 100312, Biolegend), CD11b-PE/Cy5 (1:1600, M1/70, 101210, Biolegend), CD11b-PE/Cy7 (1:100, M1/70, 101216, Biolegend), CD19-PE/Cy5 (1:100, 1D3, 115510, Biolegend),CD19-PE (1:100, 1D3, 2016867, eBioscience), CD31-APC (1:200, MEC13.3, 102516, Biolegend), CD34-FITC (1:25, RAM34, 11-0341-85, eBioscience), CD41-FITC (1:100, MWReg30, 553848, BD Pharmingen), CD45-PE/Cy5 (1:800, 30-F11, 103110, Biolegend), CD45.1-FITC (1:100, A20, 110706, Biolegend), CD45.2-APCe780 (1:100, 104, 47-0454-82, eBioscience), CD48-PB (1:200, HM48-1, 103418, Biolegend), CD71-PE (1:400, C2, 553267, BD Pharmingen), CD150-BV421 (1:50, TC15-12F12.2, 115925, Biolegend), CD150-PE (1:100, TC15-12F12.2, 115904, Biolegend), c-Kit-APC (1:200, 2B8, 2078220, eBioscience), c-Kit-PE/Cy7 (1:200, 2B8, 105814, Biolegend), FcγRII/III-PE (1:100, 2.4G2, 553145, BD Pharmingen), Flt3-Biotin (1:100, A2F10, 135308, Biolegend), F4/80-Alexa 647 (1:500, BM8, 123122, Biolegend), Gr-1-PB (1:400, RB6-8C5, 108430, Biolegend), IgM-APC (1:100, II/41, 2056825, eBioscience), IgD-FITC (1:200, 11-26 c.2a, 405704, Biolegend), IL-7Rα-PE/Cy5 (1:50, A7R34, 15-1271-83, eBioscience), IL-7Rα-PE/Cy7 (1:100, A7R34, 135014, Biolegend), NK1.1-PE (1:100, PK136, 553165, BD Pharmingen), PDCA-1-FITC (1:50, JF05-1C2.4.1, 130-102-229, Miltenyi Biotec), Sca-1-PE/Cy7 (1:100, E13-161.7, 108114, Biolegend), Ter119-PE/Cy5 (1:400, Ter119, 116210, Biolegend), Ter119-APC (1:100, Ter119, 116212, Biolegend). The following polyclonal antibodies were used: PDGFRβ (1:200, R&D Systems, BAF1042), TRAP (1:100, Abcam, ab185716), and Col3a1 (1:100, Abcam, ab7778).
Flow cytometric analysis and cell sorting
Bone marrow cells were isolated by flushing or crushing from femurs and tibias. Bone marrow nonhematopoietic cells were isolated by flushing or crushing from femurs, tibias, and humeri followed by enzymatic digestion with collagenase type I (Gibco) and DNase I (Sigma). Cells in bone fractions, including osteoblasts and PαS cells, were isolated by mechanical disruption and collagenase digestion of bones as described previously8. CAR cells were isolated as CXCL12-GFPhiSca-1-CD31-CD45-Ter119- cells from CXCL12-GFP mice or PDGFRβ+Sca-1-CD31-CD45-Ter119- cells. Endothelial cells were isolated as CD31+Sca-1+CD45-Ter119- cells. Osteoblasts were isolated as ALPhiCXCL12-GFPloCD31-CD45-Ter119- cells from CXCL12-GFP mice using ELF-97 phosphatase substrate (Invitrogen). All flow cytometric experiments and cell sorting were performed using a BD FACS Aria and FACS Diva 8.0.1 software (BD Biosystems). Gating strategies are included in Supplementary Fig. 8.
Bone marrow sections were analyzed by immunofluorescence as described previously10. In brief, bone samples were fixed in 4% paraformaldehyde and equilibrated in 30% sucrose/phosphate-buffered saline (PBS). Fixed samples were embedded in OCT medium (Sakura) and frozen in cooled hexane. Sections (12 μm) of undecalcified femoral bone were generated by Kawamoto’s film method51 (Cryofilm transfer kit; Section-Lab). Sections were stained with von Kossa and toluidine blue. Reticulin staining was performed with ammoniacal silver solution (Muto Pure Chemicals) according to the manufacturer’s instructions. For immunohistochemistry, sections were first blocked with 5% FCS and then stained with antibodies in blocking buffer. Counts of CXCL12-GFP+ CAR cells were obtained from femur sections. Confocal microscopy was performed with an LSM 510 META (Carl Zeiss). Image analysis and cell quantification were performed using Zeiss ZEN 3.0 SR software (Carl Zeiss).
Micro-computed tomography (μCT) analysis
The attached soft tissue in femurs was removed thoroughly and fixed in 4% paraformaldehyde. Micro-computed tomography scanning was performed using ScanXmate-RX Scanner (Comscantechno). Three-dimensional microstructural image data were reconstructed, and structural indices were calculated using TRI/3D-BON software (RATOC Systems).
As a bone resorption marker, serum CTX-I was measured with the RatLaps ELISA kit (Immunodiagnostic Systems) according to the manufacturer’s instructions. CXCL12 and SCF of humeri were measured as described previously8 using Quantikine ELISA Kit (R&D Systems) according to the manufacturer’s instructions.
Relative mRNA expression was analyzed by qRT-PCR analysis performed with a Step One Plus (Applied Biosystems) using Thunderbird SYBR qPCR Mix (Toyobo). Total RNA was isolated from sorted cells using Isogen (Nippon Gene) and treated with DNase I (Invitrogen), and cDNA was synthesized using SuperScript VILO (Invitrogen) following the manufacturer’s instructions. Values for each gene were normalized to the relative quantity of Gapdh mRNA in each sample. The primers used for qRT-PCR are listed in Supplementary Table 1.
Competitive repopulation assays
Competitive repopulation assays were performed using the CD45.1/CD45.2 congenic system. Unfractionated 1/20 of bone marrow cells (CD45.2) were transplanted into lethally irradiated (8 Gy) recipient mice (CD45.1) with 1×106 competitor cells (CD45.1/CD45.2). Myeloid, B, and T cells in peripheral blood of the recipient mice were analyzed by flow cytometry using antibodies against Gr-1, B220, and CD3 for 16 weeks after transplantation. High turnover of myeloid cells provides a good measure of HSC activity, and repopulating units were calculated using Harrison’s formula as described previously52.
Cell cultures and retroviral infections
Runx1, Runx2, Ebf3, and Foxc1 cDNA were cloned into MSCV-based retroviral vectors. CAR cells from wild-type mice were isolated by flow cytometry, plated into 96-well plates (500-1,000 cells/well) in MF-start medium (Toyobo) supplemented with Y-27632 (Fujifilm Wako Pure Chemical) at 10 μM, and cultured with 5% O2 and 5% CO2 for 4–5 days. Retroviral transductions into cultured CAR cells were performed with plat-E cells as producers of viral supernatants as previously described53. After transduction, media were changed into MF-start medium. Four days after transduction, transduced (GFP+ and/or Kusabira Orange+) cells were sorted for qRT-PCR analysis.
For PDGF stimulation, after 4–5 days initial culture as described above, CAR cells were stimulated with recombinant mouse PDGF-AA or PDGF-BB (Fujifilm Wako Pure Chemical) at 250 ng/mL in DMEM supplemented with 5% FCS. Three days after stimulation, PI− cells were sorted for qRT-PCR analysis.
PMF model mice
Mouse TPO cDNA was cloned into MSCV-based retroviral vectors. MSCV-IRES-GFP or MSCV-TPO-IRES-GFP was transiently transfected into Plat-E packaging cells to produce retroviruses53. Bone marrow CD48-Lin-Sca-1+c-kit+ cells were isolated from wild-type mice and cultured overnight in S-Clone SF-O3 media (Iwai North America) supplemented with mouse SCF (10 ng/ml, BioLegend), mouse TPO (20 ng/ml, R&D Systems), 100 μM 2-mercaptoethanol (Sigma-Aldrich), and 1% penicillin/streptomycin (Nacalai tesque). The pre-stimulated cells were infected with the retroviruses harboring MSCV-IRES-GFP or MSCV-TPO-IRES-GFP. The cells were harvested 48 h after the infection, and 6000-10,000 GFP+ cells were transplanted into lethally irradiated (8 Gy) recipient mice with 1 × 106 competitor cells.
One-way ANOVA with Dunnett’s multiple comparison test was used for all studies with more than two groups. For comparisons between two groups, unpaired two-tailed Student’s t-test was used. Statistical significances were calculated using GraphPad Prism 9.3.1 (GraphPad Software). All experiments were repeated at least three times with sufficient reproducibility.
Further information on research design is available in the Nature Research Reporting Summary linked to this article.
Source data are provided with this paper.
Morrison, S. J. & Scadden, D. T. The bone marrow niche for haematopoietic stem cells. Nature 505, 327–334 (2014).
Pinho, S. & Frenette, P. S. Haematopoietic stem cell activity and interactions with the niche. Nat. Rev. Mol. Cell Biol. 20, 303–320 (2019).
Sugiyama, T., Omatsu, Y. & Nagasawa, T. Niches for hematopoietic stem cells and immune cell progenitors. Int. Immunol. 31, 5–11 (2018).
Baccin, C. et al. Combined single-cell and spatial transcriptomics reveal the molecular, cellular and spatial bone marrow niche organization. Nat. Cell Biol. 22, 38–48 (2020).
Baryawno, N. et al. A cellular taxonomy of the bone marrow stroma in homeostasis and leukemia. Cell 177, 1915–1932.e16 (2019).
Ding, L., Saunders, T. L., Enikolopov, G. & Morrison, S. J. Endothelial and perivascular cells maintain haematopoietic stem cells. Nature 481, 457–462 (2012).
Gomariz, A. et al. Quantitative spatial analysis of haematopoiesis-regulating stromal cells in the bone marrow microenvironment by 3D microscopy. Nat. Commun. 9, 2532 (2018).
Omatsu, Y. et al. The essential functions of adipo-osteogenic progenitors as the hematopoietic stem and progenitor cell niche. Immunity 33, 387–399 (2010).
Seike, M., Omatsu, Y., Watanabe, H., Kondoh, G. & Nagasawa, T. Stem cell niche-specific Ebf3 maintains the bone marrow cavity. Genes Dev. 32, 359–372 (2018).
Sugiyama, T., Kohara, H., Noda, M. & Nagasawa, T. Maintenance of the hematopoietic stem cell pool by CXCL12-CXCR4 chemokine signaling in bone marrow stromal cell niches. Immunity 25, 977–988 (2006).
Tikhonova, A. N. et al. The bone marrow microenvironment at single-cell resolution. Nature 569, 222–228 (2019).
Zhou, B. O., Yue, R., Murphy, M. M., Peyer, J. G. & Morrison, S. J. Leptin-receptor-expressing mesenchymal stromal cells represent the main source of bone formed by adult bone marrow. Cell Stem Cell 15, 154–168 (2014).
Cordeiro Gomes, A. et al. Hematopoietic stem cell niches produce lineage-instructive signals to control multipotent progenitor differentiation. Immunity 45, 1219–1231 (2016).
Greenbaum, A. et al. CXCL12 in early mesenchymal progenitors is required for haematopoietic stem-cell maintenance. Nature 495, 227–230 (2013).
Omatsu, Y., Seike, M., Sugiyama, T., Kume, T. & Nagasawa, T. Foxc1 is a critical regulator of haematopoietic stem/progenitor cell niche formation. Nature 508, 536–540 (2014).
Decker, M. et al. Leptin-receptor-expressing bone marrow stromal cells are myofibroblasts in primary myelofibrosis. Nat. Cell Biol. 19, 677–688 (2017).
Chen, M. J., Yokomizo, T., Zeigler, B. M., Dzierzak, E. & Speck, N. A. Runx1 is required for the endothelial to haematopoietic cell transition but not thereafter. Nature 457, 887–891 (2009).
Okuda, T., Van Deursen, J., Hiebert, S. W., Grosveld, G. & Downing, J. R. AML1, the target of multiple chromosomal translocations in human leukemia, is essential for normal fetal liver hematopoiesis. Cell 84, 321–330 (1996).
Growney, J. D. et al. Loss of Runx1 perturbs adult hematopoiesis and is associated with a myeloproliferative phenotype. Blood 106, 494–504 (2005).
Ichikawa, M. et al. AML-1 is required for megakaryocytic maturation and lymphocytic differentiation, but not for maintenance of hematopoietic stem cells in adult hematopoiesis. Nat. Med. 10, 299–304 (2004).
Behrens, K. et al. Runx1 downregulates stem cell and megakaryocytic transcription programs that support niche interactions. Blood 127, 3369–3381 (2016).
Sood, R., Kamikubo, Y. & Liu, P. Role of RUNX1 in hematological malignancies. Blood 129, 2070–2082 (2017).
Ducy, P., Zhang, R., Geoffroy, V., Ridall, A. L. & Karsenty, G. Osf2/Cbfa1: A transcriptional activator of osteoblast differentiation. Cell 89, 747–754 (1997).
Komori, T. et al. Targeted disruption of Cbfa1 results in a complete lack of bone formation owing to maturational arrest of osteoblasts. Cell 89, 755–764 (1997).
Otto, F. et al. Cbfa1, a candidate gene for cleidocranial dysplasia syndrome, is essential for osteoblast differentiation and bone development. Cell 89, 765–771 (1997).
Sawai, C. M. et al. Transcription factor Runx2 controls the development and migration of plasmacytoid dendritic cells. J. Exp. Med. 210, 2151–2159 (2013).
Morikawa, S. et al. Prospective identification, isolation, and systemic transplantation of multipotent mesenchymal stem cells in murine bone marrow. J. Exp. Med. 206, 2483–2496 (2009).
Sroczynska, P., Lancrin, C., Kouskoff, V. & Lacaud, G. The differential activities of Runx1 promoters define milestones during embryonic hematopoiesis. Blood 114, 5279–5289 (2009).
Logan, M. et al. Expression of Cre Recombinase in the developing mouse limb bud driven by a Prxl enhancer. Genesis 33, 77–80 (2002).
Kimura, A. et al. Runx1 and runx2 cooperate during sternal morphogenesis. Development 137, 1159–1167 (2010).
Nakashima, K. et al. The novel zinc finger-containing transcription factor Osterix is required for osteoblast differentiation and bone formation. Cell 108, 17–29 (2002).
Pronk, C. J. H. et al. Elucidation of the phenotypic, functional, and molecular topography of a myeloerythroid progenitor cell hierarchy. Cell Stem Cell 1, 428–442 (2007).
Schneider, R. K. et al. Gli1+ Mesenchymal stromal cells are a key driver of bone marrow fibrosis and an important cellular therapeutic target. Cell Stem Cell 20, 785–800.e8 (2017).
Peters, A. et al. Th17 cells induce ectopic lymphoid follicles in central nervous system tissue inflammation. Immunity 35, 986–996 (2011).
Rumi, E. & Cazzola, M. Diagnosis, risk stratification, and response evaluation in classical myeloproliferative neoplasms. Blood 129, 680–692 (2017).
Inada, M. et al. Maturational disturbance of chondrocytes in Cbfa1-deficient mice. Dev. Dyn. 214, 279–290 (1999).
Nishimura, R. et al. Osterix regulates calcification and degradation of chondrogenic matrices through matrix metalloproteinase 13 (MMP13) expression in association with transcription factor Runx2 during endochondral ossification. J. Biol. Chem. 287, 33179–33190 (2012).
Takarada, T. et al. An analysis of skeletal development in osteoblast-specific and chondrocyte-specific runt-related transcription factor-2 (Runx2) knockout mice. J. Bone Miner. Res. 28, 2064–2069 (2013).
Matsushita, Y. et al. A Wnt-mediated transformation of the bone marrow stromal cell identity orchestrates skeletal regeneration. Nat. Commun. 11, 332 (2020).
Liu, Y. et al. Osterix-Cre labeled progenitor cells contribute to the formation and maintenance of the bone marrow stroma. PLoS One 8, e71318 (2013).
Wen, Q., Goldenson, B. & Crispino, J. D. Normal and malignant megakaryopoiesis. Expert Rev. Mol. Med. 13, e32 (2011).
Kuvardina, O. N. et al. RUNX1 represses the erythroid gene expression program during megakaryocytic differentiation. Blood 125, 3570–3579 (2015).
Papadantonakis, N., Matsuura, S. & Ravid, K. Megakaryocyte pathology and bone marrow fibrosis: The lysyl oxidase connection. Blood 120, 1774–1781 (2012).
Comazzetto, S. et al. Restricted hematopoietic progenitors and erythropoiesis require SCF from leptin receptor+ niche cells in the bone marrow. Cell Stem Cell 24, 477–486 (2019).
Ishikawa, G. et al. Shared and tissue-specific expression signatures between bone marrow from primary myelofibrosis and essential thrombocythemia. Exp. Hematol. 79, 16–25 (2019).
Aoki, K. et al. Identification of CXCL12-abundant reticular cells in human adult bone marrow. Br. J. Haematol. 193, 659–668 (2021).
Zhao, K. et al. Runx1 contributes to the mesenchymal subtype of glioblastoma in a TGFβ pathway-dependent manner. Cell Death Dis. 10, 877 (2019).
Kuo, Y. H. et al. Runx2 induces acute myeloid leukemia in cooperation with Cbfp-SMMHC in mice. Blood 113, 3323–3332 (2009).
Ara, T. et al. Long-term hematopoietic stem cells require stromal cell-derived factor-1 for colonizing bone marrow during ontogeny. Immunity 19, 257–267 (2003).
Rodda, S. J. & McMahon, A. P. Distinct roles for Hedgehog and canonical Wnt signaling in specification, differentiation and maintenance of osteoblast progenitors. Development 133, 3231–3244 (2006).
Kawamoto, T. & Kawamoto, K. Preparation of thin frozen sections from nonfixed and undecalcified hard tissues using Kawamoto’s film method (2020). Methods Mol. Biol. 2230, 259–281 (2021).
Harrison, D. E., Jordan, C. T., Zhong, R. K. & Astle, C. M. Primitive hemopoietic stem cells: direct assay of most productive populations by competitive repopulation with simple binomial, correlation and covariance calculations. Exp.Hematol. 21, 206–219 (1993).
Morita, S., Kojima, T. & Kitamura, T. Plat-E: An efficient and stable system for transient packaging of retroviruses. Gene Ther. 7, 1063–1066 (2000).
We thank A. Okada and K. Nagahara for secretarial assistance, and members of the Laboratory of Animal Experiments for Regeneration of Kyoto University for animal care. This work was supported by grants from the JSPS KAKENHI (grant number 18H03998 and 17H05643 to T.N. and 19K06688 and 16H06232 to Y.O.), the Princess Takamatsu Cancer Reserch Fund to T.N., the Mochida Memorial Foundation for Medical and Pharmaceutical Research and the Inamori Foundation to Y.O., Cooperative Research Program (Joint Usage/Research Center program) of Institute for Frontier Life and Medical Sciences, Kyoto University to T.N..
The authors declare no competing interests.
Peer review information
Nature Communications thanks Anna Rita Migliaccio and the other anonymous reviewers for their contribution to the peer review of this work.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
About this article
Cite this article
Omatsu, Y., Aiba, S., Maeta, T. et al. Runx1 and Runx2 inhibit fibrotic conversion of cellular niches for hematopoietic stem cells. Nat Commun 13, 2654 (2022). https://doi.org/10.1038/s41467-022-30266-y