Skip to main content
NCBI home page
The American Journal of Pathology logoLink to The American Journal of Pathology
. 2012 Feb;180(2):811–818. doi: 10.1016/j.ajpath.2011.10.028

A Novel Population of Cells Expressing Both Hematopoietic and Mesenchymal Markers Is Present in the Normal Adult Bone Marrow and Is Augmented in a Murine Model of Marrow Fibrosis

Masanobu Ohishi ⁎,†, Wanida Ono ⁎, Noriaki Ono ⁎, Richa Khatri ⁎, Marilena Marzia ⁎, Emma K Baker ‡, Sierra H Root §, Tremika Le-Shan Wilson ⁎,, Yukihide Iwamoto †, Henry M Kronenberg ⁎, Hector L Aguila §, Louise E Purton ‡,∥,⁎, Ernestina Schipani ⁎,
PMCID: PMC3349873  PMID: 22155108

Abstract

Bone marrow (BM) fibrosis is a feature of severe hyperparathyroidism. Consistent with this observation, mice expressing constitutively active parathyroid hormone (PTH)/PTH-related peptide receptors (PPR) in osteoblasts (PPR*Tg) display BM fibrosis. To obtain insight into the nature of BM fibrosis in such a model, a double-mutant mouse expressing constitutively active PPR and green fluorescent protein (GFP) under the control of the type I collagen promoter (PPR*Tg/GFP) was generated. Confocal microscopy and flow cytometry revealed the presence of a cell population expressing GFP (GFP+) that was also positive for the hematopoietic marker CD45 in the BM of both PPR*Tg/GFP and control animals. This cell population was expanded in PPR*Tg/GFP. The existence of cells expressing both type I collagen and CD45 in the adult BM was confirmed by IHC and fluorescence-activated cell sorting. An analysis of total RNA extracted from sorted GFP+CD45+ cells showed that these cells produced type I collagen and PTH/PTH-related peptide receptor and receptor activator for NF-κB mRNAs, further supporting their features of being both mesenchymal and hematopoietic lineages. Similar cells, known as fibrocytes, are also present in pathological fibroses. Our findings, thus, indicate that the BM is a permissive microenvironment for the differentiation of fibrocyte-like cells and raise the possibility that these cells could contribute to the pathogenesis of BM fibrosis.

The bone marrow (BM) is a complex microenvironment composed of both hematopoietic and mesenchymal cells. Hematopoietic cells, including bone-resorbing osteoclasts, derive from self-renewing hematopoietic stem cells (HSCs) and constitute most of the adult BM cellularity. The mesenchymal population, comprising osteoblasts (ie, the cells making bone) and adipocytes, is thought to originate from mesenchymal stem cells (MSCs).1 Recent studies2–5 have challenged the simple dichotomy hematopoietic versus mesenchymal by raising the intriguing possibility that cells of hematopoietic origin could differentiate into mesenchymal lineages.

The BM stroma consists of a heterogeneous population that provides support to hematopoietic cells and contains cells that can differentiate into various mesenchymal cell lineages. Marrow fibrosis is a pathological condition characterized by expansion of the BM stroma, with consequent abnormal accumulation in the BM of fibroblastoid cells and collagen fibers. It can be idiopathic, although in humans it is often a feature of a variety of malignancies of the hematopoietic system, such as myelofibrosis,6,7 and of nonmalignant pathological conditions, such as severe primary and secondary hyperparathyroidism8–10 and fibrous dysplasia.11 Consistent with these findings, transgenic mice expressing constitutively active parathyroid hormone (PTH)/PTH-related peptide receptors in osteoblasts (PPR*Tg) show a considerable accumulation of fibroblastoid cells in their BM.12

Marrow fibrosis could be a disease of MSCs.13 In this regard, the number of CD146+CD45− skeletal stem cells14 is augmented in the BM of patients with myelofibrosis, which suggests that MSCs/mesenchymal progenitors could be involved in the pathogenesis of BM fibrosis.15 Interestingly, a heterogeneous group of cells of BM origin, expressing both hematopoietic and mesenchymal markers, has recently been identified in the circulation and at sites of pathological fibroses.16–18 To this end, no evidence that such cells are, indeed, present in the BM in vivo has been reported.

To gain more insights into the complexity of the BM stroma in a model of BM fibrosis, we, thus, studied whether the stroma expansion observed in the BM of PPR*Tg mice was concomitant to a modulation of the pool of immature mesenchymal cells or whether it shared features of cells involved in pathological fibroses.

Materials and Methods

Mice

Transgenic mice expressing constitutively active PTH/PTH-related peptide receptors (PPR*) under the control of the 2.3-kb fragment of the mouse a1 (I) collagen gene promoter (PPR*Tg) and transgenic mice expressing green fluorescent protein (GFP) under the control of the 2.3-kb fragment rat a1 (I) collagen gene promoter [wild type (WT)/GFP] have been previously described.19 Hemizygous WT/GFP mice were crossed with hemizygous PPR*Tg mice, and double-hemizygous mutant mice were generated (PPR*Tg/GFP). C57BL/6 mice were obtained from Animal Resources Center, Perth, WA, Australia. All studies were approved by an animal care committee of each institution.

Preparation of BM Cells for Flow Cytometry Analysis or Cell Sorting

Tibiae, femurs, and iliac crests were isolated from either 6- to 8-week-old WT and PPR*Tg mice or 6- to 8-week-old WT/GFP and PPR*Tg/GFP mice. After scraping the periosteum with scalpels, bone specimens were digested with a solution containing collagenase 1 and collagenase 2 (Worthington, Lakewood, NJ), for 15 minutes at 37°C, to completely eliminate the periosteum. Epiphyses were detached, and specimens were crushed in PBS using mortal and pestle. Cells were then flushed through cell strainers (BD/Falcon, Franklin Lakes, NJ), layered over Ficoll-Paque PLUS (GE Healthcare, Upsala, Sweden), and centrifuged for 10 minutes at 4°C. Low-density cell fractions were collected and resuspended in one times PBS–2% fetal bovine serum at a concentration of 1 × 107 cells/mL. Thereafter, collected cells were stained and analyzed by flow cytometry using an LSRII, an FACSCalibur, or an LSRFortessa analyzer (BD/Falcon for all), or sorted using an FACS Aria cell sorter (BD).

PTH Administration and Preparation of Endosteal-Enriched BM Cells for Flow Cytometry Analysis

C57BL6 WT mice, aged 6 to 8 weeks, were injected s.c. with either PBS or PTH (1–34) at 80 μg/kg per day. Muscular tissue was removed, and BM was flushed with 10 mL of HBSS media, pH 7.4, supplemented with 2% fetal calf serum (FCS). Erythrocytes were lysed by hypotonic shock, and flushed cells were then spun down and finally resuspended in HBSS/FCS (fraction 1). Flushed bones were cut into small chips, placed in 50-mL conical tubes, and rinsed in PBS to eliminate loosely attached cells. Bone chips were then suspended in 2.5 mL of PBS, containing 2 mg/mL collagenase A and 2 mg/mL hyaluronidase, and incubated at 37°C, with intermittent vortex mixing. After 25 minutes, supernatant was removed, and bone chips were then subjected to a second digestion with 2 mL of 2 mg/mL collagenase A and 2 mg/mL hyaluronidase dissolved in 0.25% trypsin-EDTA, for 45 minutes at 37°C, with intermittent vortex mixing. After incubation, supernatant was removed and trypsin was inactivated with 15% serum. Both digested fractions were combined, washed several times in HBSS-FCS, and filtered through a 70-mm cup strainer. This fraction (fraction 2) was enriched for cells of the endosteal compartment.

Flow Cytometry

BM cells, isolated from WT or PPR*Tg mice, were incubated with CD31–activated protein C (APC), CD45-APC, and stem cell antigen 1 (ScaI)–phosphatidylethanolamine (PE)–Cy5.5 antibodies (eBioscience, San Diego, CA) on ice for 30 minutes. Cells were then washed with PBS–2% fetal bovine serum and analyzed on an LSRII or an FACSCalibur analyzer (BD for both). BM cells isolated from WT/GFP mice, with or without PTH treatment, or PPR*Tg/GFP mice were incubated with CD45-PE or CD45-APC and CD11b-APC antibodies (eBioscience) and analyzed as previously described.

Analysis of collagen 1–positive cells in lineage-negative BM cells of WT mice was performed by staining BM cells isolated from C57BL/6 WT mice with a cocktail of PE-conjugated rat anti-mouse antibodies against Ter119, CD4, CD8, B220, and Gr-1 for lineage depletion. Analysis was also performed with CD45.2–Alexa Fluor 700, CD11b–fluorescein isothiocyanate, ScaI-PE-Cy7, c-Kit-APC-eFluor780, and F4/80-eFluor450 antibodies (all from eBioscience), together with either a biotinylated rabbit anti-mouse type 1 collagen antibody (Rockland, Gilbertsville, PA) or a biotinylated rabbit IgG-(Fc) isotype control (α Diagnostic International Inc., San Antonio, TX) on ice for 30 minutes. Cells were then washed, incubated with streptavidin PerCP-Cy5.5 antibody (eBioscience) on ice for 30 minutes, and analyzed as previously described. The cell populations shown were gated on lineage-negative cells from the BM.

RT-PCR Data

RNA was isolated from sorted cells (up to 1 × 105 cells) by using an RNeasy kit (Qiagen, Hilden, Germany). First-strand cDNA was synthesized using the QuantiTect Reverse Transcription kit (Qiagen). The RT-PCR was performed in a 25-μL reaction mixture of Illustra PureTaq Read-to-go beads (GE Healthcare) containing 1 μL of cDNA samples. The PCR conditions were as follows: 95°C for 1 minute, 58°C for 45 seconds, and 72°C for 1 minute, for a total of 35 cycles. The primer sequences are as follows: type 1 collagen, 5′-TGTGTGCGATGACGTGCAAT-3′ (forward) and 5′-GGGTCCCTCGACTCCTACA-3′ (reverse); GFP, 5′-CAAGATCCGCCACAACATCGAG-3′ (forward) and 5′-ATGTGATCGCGCTTCTCGTTGGGG-3′ (reverse); receptor activator for NF-κB (RANK), 5′-GGACGGTGTTGCAGCAGAT-3′ (forward) and 5′-CGAGTCTGAGTTCCAGTG-GTA-3′ (reverse); PTH1R, 5′-GTGGGACAACATCGTTTGCTG-3′ (forward) and 5′-CGGTCAAATACCTCCCG-TTC-3′ (reverse); β-actin, 5′-GGCTGTATTCCCCTCCATCG-3′ (forward) and 5′-CCAGTTGGTAACAATG-CCATGT-3′ (reverse); and PPR*Tg, 5′-CACCTGCCCTGCTACAGGAAGAG-3′ (forward) and 5′-TTCCACCACTG-CCTCCCATTCATC-3′ (reverse).

IHC on Undecalcified Cryosections

Femora isolated from 6- to 8-week-old PPR*Tg/GFP mice were excised, snap frozen in n-hexane/dry ice (approximately −70°C), and embedded in super cryosection embedding medium (Section-Lab, Hiroshima, Japan). Nondecalcified longitudinal sections of femurs were cut using a tungsten blade and a tape transfer system (CryofilmIIC; Section-Lab). Nondecalcified sections were adhered on the adhesive side of cryofilms and analyzed as attached throughout. Sections on cryofilms were rapidly dehydrated in 100% ethanol for 1 minute, air dried completely, and kept at −20°C until use. Slides were rinsed in 0.1% Tween 20 in Tris-buffered saline three times, blocked with 2% BSA–0.1% Tween 20 in Tris-buffered saline for 30 minutes at room temperature, and incubated with primary antibodies diluted in blocking solution overnight at 4°C. Secondary antibodies were diluted in blocking solution and incubated for 3 hours at 4°C, followed by nuclear staining with TO-PRO-3 (1:1000; Invitrogen, Carlsbad, CA) for 15 minutes at room temperature. The following antibodies were used: anti-GFP polyclonal (1:200; Abcam, Cambridge, MA), anti-CD45 monoclonal (1:200; Abcam), anti-collagen 1 polyclonal (1:200; Abcam), Alexa 488–conjugated goat anti-rabbit IgG (1:400; Abcam), and Alexa 546–conjugated goat anti-rat IgG (1:400; Abcam). A confocal laser microscope (LSM510 Pascal; Carl Zeiss, Oberkochen, Germany) was used for signal visualization, with an excitation wavelength of 488/520/610 nm and band-pass filters of 515/550/630 nm, respectively.

Detection of GFP Signaling and TRAP Staining on Decalcified Cryosections

Tibias isolated from WT/GFP or PPR*Tg/GFP mice were fixed in 4% paraformaldehyde for 48 hours at 4°C. Specimens were decalcified in 20% EDTA solution at 4°C for 7 days, and the EDTA solution was changed daily. Decalcified bones were then soaked in 30% sucrose/one times PBS for 24 hours, and embedded in Optimal Cutting Temperature compound (Sakura, Tokyo, Japan). Sections (10 μm thick) were air dried, rehydrated, and mounted. Pictures were then taken for documentation of GFP expression. In selected cases, sections were immersed in water to wash out the mounting media, and tartrate-resistant acid phosphatase (TRAP) staining was performed using an acid phosphatase detection kit (Sigma-Aldrich Corp, St Louis, MO).

IHC and in Situ Hybridization on Paraffin Sections

The tibias of WT and PPR*Tg mice were fixed in 4% paraformaldehyde at 4°C for 48 hours. After decalcification with 20% EDTA at 4°C for 7 days, paraffin blocks were prepared by standard procedures. The sections were deparaffinized, and antigen retrieval was performed using proteinase K (20 mg/mL; Roche, Indianapolis, IN), followed by 10-minute treatment with 3% H2O2 to block endogenous peroxidase. The presence of immature progenitors was detected using anti-mouse ScaI Ab (BD) at a dilution of 1:500 for 1 hour at room temperature and the Tyramide Signal Amplification biotin system (Perkin-Elmer, Waltham, MA), according to the manufacturer's protocol. In situ hybridization analyses were performed as previously described12 using 35S-labeled riboprobes on paraffin sections of bone specimens prepared, as previously described.

Results

Marrow Fibrosis in PPR*Tg Mice Is Formed by Type I Collagen–Producing Cells but Not by Terminally Differentiated Osteoblasts

A considerable expansion of a fibroblastoid population or marrow fibrosis was observed in the BM of 2-week-old PPR*Tg mice, particularly in their metaphyseal regions, and persisted until at least the age of 16 weeks (Figure 1A).12 In situ hybridization analysis showed that these cells were heterogeneously expressing type I collagen mRNA (Col I), whereas osteocalcin mRNA, which is a classical marker of terminally differentiated osteoblasts, was detectable exclusively in osteoblasts adjacent to bone surfaces but not in the marrow stroma (Figure 1B).

Figure 1.

Figure 1

Open in a new tab

A: Expansion of stromal cells in trabecular bone areas of PPR*Tg mice. H&E staining of paraffin sections of decalcified mouse tibias isolated from 16-week-old WT and PPR*Tg mice is shown. Original magnification, ×20. B: Decalcified paraffin sections of tibias isolated from adult PPR*Tg mice were hybridized with specific riboprobes for detection of type I collagen (Col I) and osteocalcin (OCN) mRNAs, by in situ hybridization. Bright fields (top panels) and dark fields are shown. Original magnification, ×20. Arrows indicate that stromal cells express Col1 but not OCN. C: Histological analysis of bone specimens isolated from PPR*Tg/GFP mice. Decalcified frozen section of tibias isolated from 6-week-old PPR*Tg/GFP mice were first analyzed to identify cells expressing GFP (left panels). Sections were then stained for TRAP and H&E (right panels). Bottom panels: Magnified images of the boxed areas in the top panels. Original magnification: ×20 (top panels); ×40 (bottom panels). Areas surrounded by dotted lines in the bottom panels indicate bone matrix. Bone-lining osteoblasts and osteocytes are GFP+. Yellow arrowheads indicate TRAP-positive osteoclasts. White arrowheads indicate GFP− stromal cells. D: Immunostaining for ScaI. Histological sections of tibias isolated from WT and PPR*Tg mice were incubated with ScaI antibody; antibody binding was detected by diaminobenzidine. Images of trabecular bone areas are shown. In both genotypes, cells in the wall of blood vessels (double arrowheads) were positive for ScaI. Osteoblasts on bone surfaces (arrows) were negative for ScaI in both WT and PPR*Tg mice. Most stromal cells in PPR*Tg mice were negative for ScaI (circled areas). Original magnification, ×40.

To gain more insights into the nature of the marrow fibrosis observed in PPR*Tg mice, we took advantage of a transgenic mouse line that expresses GFP in mature osteoblasts under the control of a 2.3-kb fragment of the rat a1 (I) collagen gene promoter (WT/GFP),19 which is similar to the promoter fragment driving expression of PPR* in PPR*Tg mice. PPR*Tg and WT/GFP mice were crossed to generate PPR*Tg/GFP double-hemizygous mice. Histological analysis of decalcified frozen sections of adult tibias isolated from PPR*Tg/GFP mice revealed that, as previously reported,19 osteoblasts closely adjacent to bone surfaces expressed considerable amounts of GFP (Figure 1C). Conversely, most stromal cells did not express GFP to a detectable level, although a few fibroblastoid cells were positive for this fluorochrome (Figure 1C). In contrast to findings in transgenic mice expressing GFP under the control of the 3.6-kb fragment of the a1 (I) collagen gene promoter,20 no expression of GFP was detectable in TRAP-positive multinucleated osteoclasts (Figure 1C).

Taken together, these data indicate that most fibroblastoid cells present in the marrow stroma of PPR*Tg mice are not terminally differentiated osteoblasts, although they produce type I collagen, which is suggestive of a mesenchymal identity of these cells.

Marrow Fibrosis in PPR*Tg Mice Is Not Contributed by Immature Progenitor Cells

We next determined whether the expansion of fibroblastoid cells observed in the BM of PPR*Tg mice was the result of an increase of immature progenitors of the mesenchymal lineage. Several lines of evidence suggest that ScaI is expressed in mesenchymal progenitors.21–23 We, thus, performed immunohistochemical (IHC) staining for ScaI on histological sections of bone specimens isolated from WT and PPR*Tg mice. As expected, terminally differentiated osteoblasts in either genotype did not express detectable levels of ScaI (Figure 1D). Most fibroblastoid cells in the BM were also negative for ScaI (Figure 1D). Interestingly, an intense ScaI-positive staining was observed in and adjacent to the wall of sinusoidal blood vessels (Figure 1).

A BM population positive for ScaI and negative for the pan-leukocyte marker CD45 and for the endothelial cell marker CD31 (ScaI+CD45−CD31− cells) is enriched in mesenchymal progenitors.22 A distinct ScaI+CD45−CD31− population was, indeed, present in the BM of either WT or PPR*Tg mice by flow cytometry analysis performed on crushing of adult bone specimens (Figure 2A); however, no significant difference in the percentage of ScaI+CD45−CD31− cells per total BM cells could be detected between PPR*Tg and WT mice (Figure 2B). In agreement with the observation that ScaI+CD45−CD31− population is enriched in immature progenitors,22 flow cytometry analysis of BM cells isolated from either WT/GFP mice or PPR*Tg/GFP mice indicated that the ScaI+CD45−CD31− population does not express detectable levels of GFP (data not shown).

Figure 2.

Figure 2

Open in a new tab

Flow cytometry analysis of BM cells isolated from WT and PPR*Tg mice. BM cells were collected as described in Materials and Methods, and were then incubated with ScaI-PE-Cy5.5, CD45-APC, and CD31-APC antibodies. A: Representative analysis of BM cells isolated from WT and PPR*Tg mice. B: A summary of the percentage of ScaI+CD45−CD31− cells in WT (n = 6) and PPR*Tg (n = 6) mice from two independent sets of experiments is shown. Each symbol represents a separate mouse. The same symbol indicates the data came from the same experiment within the two replicate experiments. P = 0.69.

Taken together, our data suggest that the fibroblastoid population forming the marrow fibrosis in PPR*Tg mice is not a population of immature mesenchymal progenitor cells.

A Population of Cells Expressing Both Hematopoietic and Mesenchymal Markers Is Present in the Normal Adult BM and Is Significantly Augmented in the PPR*Tg

To distinguish hematopoietic and nonhematopoietic cells in the BM of PPR*Tg mice, we performed confocal microscopy analysis of frozen bone sections isolated from PPR*Tg/GFP mice by staining with the pan-leukocyte marker CD45. No CD45 signal was observed in specimens stained with an isotype control antibody or in the absence of a primary antibody (data not shown). As expected, GFP+ bone-lining osteoblasts were negative for CD45 (Figure 3A). However, surprisingly, we identified fibroblastoid cells that were positive for both GFP and CD45 (GFP+CD45+) (Figure 3A). Similar findings were obtained when confocal microscopy analysis was performed to detect native type I collagen and CD45 proteins on equivalent specimens obtained from PPR*Tg mice (Figure 3B).

Figure 3.

Figure 3

Open in a new tab

A: Confocal microscopy image of a frozen section of a tibia isolated from PPR*Tg/GFP mice. Tibias of 6-week-old mice were dissected and processed as described in Materials and Methods. Sections were stained with anti-GFP polyclonal, anti-CD45 monoclonal, Alexa 488–conjugated goat anti-rabbit IgG and Alexa 546–conjugated goat anti-rat IgG. TO-PRO-3 was used for nuclear staining. The arrows indicate the presence of GFP+CD45+ cells. B: A representative frozen section of a tibia of a PPR*Tg mouse was incubated with anti-Col I polyclonal, anti-CD45 monoclonal, Alexa 488–conjugated anti-rabbit IgG and Alexa 546–conjugated anti-rat IgG. TO-PRO-3 was used for nuclear staining. Arrows indicate the presence of CD45+Col I+ cells.

To further characterize the GFP+CD45+ cells identified in the BM of PPR*Tg/GFP mice by confocal microscopy, we then performed flow cytometry analysis of BM cells isolated from either WT/GFP or PPR*Tg/GFP mice, on crushing of the long bones. Notably, we preferred to crush, rather than to flush, the specimens of interest because in previous pilot experiments we had not been able to isolate a sufficient number of BM cells from PPR*Tg by flushing, which was likely the result of the complex trabecular architecture of the mutant mice (data not shown). To enhance the specificity of our findings, all duplex events were carefully gated out (data not shown). Flow cytometry analysis confirmed the presence of GFP+CD45+ cells in the BM of both mutant (Figure 4B) and control (Figure 4A) mice. When isotype control antibody was used, no CD45+ events could be identified within the GFP+ population (data not shown). The GFP+CD45+ cell population was significantly expanded in the BM of PPR*Tg/GFP mice (Figure 4, A–C). Interestingly, most GFP+CD45+ cells isolated from either WT/GFP or PPR*Tg/GFP mice expressed CD11b (Figure 4, A and B).

Figure 4.

Figure 4

Open in a new tab

A–C: Identification and characterization of GFP+CD45+ cells in WT and PPR*Tg mice. BM cells were isolated as described in Materials and Methods from WT/GFP (A) and PPR*Tg/GFP (B) mice, and analyzed for the expression of GFP, CD45, and CD11b. Note the expansion of the GFP+CD45+ cell population in PPR*Tg (B, left panel) compared with WT (A, left panel) mice. A and B, right panels: Most of the GFP+CD45+ cells are positive for CD11b. C: Comparison of percentage of GFP+CD45+ cells per total BM cells between WT/GFP (n = 4) and PPR*Tg/GFP (n = 5) mice. The results of four independent experiments were summed, and the average values are shown. A Student's t-test was performed to validate the significance of difference. P < 0.01. D: GFP+CD45− cells, GFP+CD45+CD11b−cells, and GFP+CD45+CD11b+ cells were sorted from the BM of WT/GFP mice, and total RNA was collected. RT-PCR was performed for β-actin (154 bp), Col I (133 bp), GFP (159 bp), PPR (225 bp), and RANK (243 bp) (product size is in parentheses). The MspI digest of pBR322 DNA was used as mol. wt. standards. E and F: Detection of CD45+CD11b+ Col1+ F4/80+ and CD45+CD11b+ Col1+ F4/80− cells in the BM of WT mice. CD45+CD11b+ cells in lineage-negative gated BM are shown (left panels). These can be further subdivided based on their expression of F4/80 and Col 1 (F). An isotype control for Col 1 is shown in E.

RT-PCR analysis of total RNA extracted from GFP+CD45+ cells, isolated from WT/GFP mice and sorted twice to eliminate any potential contamination with other cell types, indicated that these cells expressed both Col I and GFP mRNAs (Figure 4D). Moreover, native PPR, which is expressed in MSCs and osteoblasts,24,25 and RANK, which is present in osteoclast lineage cells and in dendritic cells,26 were also detectable in GFP+CD45+ cells by RT-PCR, confirming that these cells had mixed features of both mesenchymal and hematopoietic lineages (Figure 4D). Similar results were obtained in GFP+CD45+ cells isolated from PPR*Tg/GFP mice (data not shown). Not surprisingly, these cells also expressed PPR* transgene mRNA, which, like GFP transgene, is driven by a similar 2.3-kb fragment of the a1 (I) collagen gene promoter (data not shown). A scarce number of GFP+CD45− cells were also identified by flow cytometry in the BM of either WT/GFP or PPR*Tg/GFP mice, which were likely to be mature osteoblasts (Figure 4, A and B). Consistent with this conclusion, GFP+CD45− cells did not express RANK, as shown by RT-PCR (Figure 4D).

We then determined whether CD45+ cells express native type 1 collagen. For this purpose, we stained BM cells with a cocktail of lineage (Ter119, CD4, CD8, B220, and Gr-1) antibodies, together with CD45.2, CD11b, ScaI, c-Kit, and type 1 collagen antibodies, and could detect a distinct Col1+ population in lineage-negative CD45+CD11b+ cells (Figure 4E). When isotype control antibody was used, no Col1+ events could be identified within the CD45+CD11b+ population (Figure 4E). Notably, the Col1+CD45+CD11b+ population could be further divided into two subpopulations, based on the expression of F4/80 (Figure 4F). Staining the cells for ScaI and c-Kit at the same time revealed that most of the CD45+CD11b+ cells were negative for both c-Kit and ScaI and that the Col1+F4/80− and Col1+F4/80+ cell populations were predominantly in this HSC-devoid population (see Supplemental Figure S1 at http://ajp.amjpathol.org). Taken together, our data indicate that the adult murine BM contains a novel cell population that displays mixed features of mesenchymal and hematopoietic cells; this population is significantly expanded in PPR*Tg mice.

PTH Administration Expands the Population of GFP+CD45+ in the BM Endosteal Fraction of WT/GFP Mice

Based on the previous findings, we reasoned that expansion of the GFP+CD45+ population observed in GFP/PPR*Tg mice was the result of increased PTH signaling. To pharmacologically test this possibility, WT/GFP mice were treated with daily s.c. injections of PTH (1–34) at 80 μg/kg or PBS for 28 days. After treatment, mice were sacrificed and BM cells were isolated from long bones in a two-step process that included the mechanical flushing of the BM (fraction 1), followed by digestion of the flushed bones with collagenase (fraction 2). As depicted in Figure 5, we observed an increase in the number of GFP+CD45+ cells in fraction 2 collected from PTH-treated mice compared with control. However, this increase did not reach statistical significance. No differences were observed between PTH- and vehicle-treated specimens in fraction 1 (data not shown).

Figure 5.

Figure 5

Open in a new tab

Identification and characterization of GFP+CD45+ cells in Col2.3GFP mice after PTH treatment. An endosteal cell fraction (fraction 2) was isolated from Col2.3/GFP (treated and untreated) mice. Cells were stained with anti-CD45 antibody and analyzed by flow cytometry. A: Representative samples of each group. B: The percentage of GFP+ CD45+ cells in vehicle- versus PTH- treated samples (n = 3). P = 0.09.

Discussion

In recent years, we have generated a mouse model of marrow fibrosis by expressing a constitutively active receptor for PTH in osteoblasts (PPR*Tg).12 In the current study, analysis of PPR*Tg mouse, a model of BM fibrosis, has led us to the discovery of a novel BM cell population constituted by cells that express both mesenchymal and hematopoietic markers. This novel population, which we have also successfully identified in normal mice, is significantly expanded in PPR*Tg mice.

Cells expressing both mesenchymal and hematopoietic markers have been identified at sites of pathological fibroses and have been named fibrocytes. Fibrocytes have mixed features of both hematopoietic and mesenchymal cells, because they express both cell surface antigens of the hematopoietic lineage, such as CD45 and CD11b, and matrix proteins, particularly type 1 collagen,16 and they derive from BM precursors. They contribute to wound healing27 and to pathological fibrosis28 by secreting matrix and pro-angiogenic factors, and by differentiating into myofibroblasts.29 Another cell population, called circulating osteoprogenitor cells, which also expresses CD45 and type 1 collagen, has differentiated into osteoblasts and potentially plays a critical role in the pathogenesis of fibrodysplasia ossificans progressiva.30

According to the notion that the minimum criteria of co-expression of collagen and hematological markers are sufficient to define fibrocytes as such, our findings indicate that the novel cells we have identified in the BM are fibrocyte-like. It is intriguing that these BM fibrocyte-like cells are expanded in PPR*Tg mice, which is a model of marrow fibrosis. PTH treatment increased the number of GFP+CD45+ cells in the endosteal fraction but not in the rest of the BM. This finding is particularly interesting considering that marrow fibrosis is observed mainly in the endosteal region in PTH-treated rodents (data not shown).31 Notably, as previously reported and consistent with the constitutive activity of the transgene, serum PTH levels are partially suppressed in PPR*Tg mice.12

The presence of fibrocytes in the BM has not been previously reported in a direct fashion, although it is notable that similar cells have been reported to be present in the BM. In these studies, single enhanced green fluorescence protein plus HSCs were transplanted into mice, and fibroblast-like cells expressing enhanced green fluorescence protein and type 1 collagen were observed either in BM culture studies17 or in vivo, contributing to tumor stroma.18 Our studies expand on these findings, demonstrating in three different mouse models (WT, PPR*Tg, and GFP mouse strains) that lineage-negative cells that express CD45, CD11b, and type 1 collagen are present in the BM in homeostasis conditions, and are elevated in a mouse model of fibrosis. Furthermore, these fibrocyte-like cells can be further subdivided by their expression of the macrophage marker, F4/80, and most of these cells do not express the HSC markers, ScaI and c-Kit, although they are progeny of transplantable BM cells (data not shown), in agreement with the findings of Ebihara et al.17

Interestingly, in previous studies, Ebihara et al17 found that GFP-positive colony-forming cells differentiated into Col1+CD45− cells or Mac1+ macrophage-like cells, depending on culture conditions. Our studies suggest that the Col1+CD45+CD11b+ population contains a subpopulation with characteristics of fibroblasts (F4/80− cells) and a subpopulation with characteristics of macrophages (F4/80+ cells). Our current working hypothesis is that the BM, like wounds and pathological fibroses, is a permissive microenvironment for the differentiation of fibrocyte-like cells from hematopoietic precursors, and that these cells may contribute to the marrow fibrosis and/or BM homeostasis with yet unknown mechanisms. It is also possible that these fibrocyte-like cells share the same precursor as macrophages. To this end, further investigation of the F480+ and F4/80− subpopulations of the Col1+CD45+CD11b+ BM cells will expand our understanding of these fibrocyte-like cells, which is of clinical importance.

Different from transgenic mice, in which GFP is driven by the 3.6-kb fragment of the a1(I) collagen gene promoter,20 TRAP-positive multinucleated osteoclasts did not express GFP in our model. However, to this end, we cannot fully exclude the possibility that GFP+CD45+ cells, which also express RANK, differentiate into osteoclasts and lose the expression of type 1 collagen (and, hence, GFP) in their differentiation process.

We have previously reported that the marrow fibrosis observed in PPR*Tg mice, as in hyperparathyroidism and fibrous dysplasia in humans, is mainly contributed by cells of the osteoblast lineage at early stages of differentiation.12 The present study has confirmed these observations by showing by both a genetic approach and by in situ hybridization analysis that marrow fibrosis in PPR*Tg mice is formed by type I collagen–producing cells, but it is not contributed by terminally differentiated osteoblasts.

No expansion of immature cells expressing ScaI was observed in PPR*Tg mice. This finding is consistent with the finding that MSCs/progenitors, defined as cells generating colony-forming unit fibroblastoid in vitro, decline with age in PPR*Tg mice.32 Interestingly, intermittent PTH administration has expanded the pool of MSCs/progenitors expressing nestin.25 The apparent discrepancies with our findings could be the result of the fact that ScaI defines a population distinct from nestin-expressing cells or that intermittent PTH treatment has different effects on MSCs/progenitors when compared with long-term activation of PPR as it occurs in PPR*Tg mice. Further studies are required to address this important issue.

In summary, we have identified a novel population of cells expressing both hematopoietic and mesenchymal markers in the normal adult BM, which is significantly augmented in a murine model of marrow fibrosis. The clinical relevance of our findings in pathological settings, such as primary and secondary hyperparathyroidism and fibrous dysplasia, requires further studies.

Footnotes

Supported by the Japan Society for the Promotion of Science (M.O.), the Mochida Memorial Foundation (M.O.), NIH grants RC1HL100569-01 (H.L.A.) and R21AR060689 (E.S.), a grant from the National Health and Medical Research Council (NHMRC; 1006485 to L.E.P.), the Victorian Government's OIS Program (L.E.P.), and an NHMRC Senior Research Fellowship (L.E.P.).

M.O. and W.O. contributed equally to this work.

Supplemental material for this article can be found at http://ajp.amjpathol.org or at doi: 10.1016/j.ajpath.2011.10.028.

Contributor Information

Louise E. Purton, Email: lpurton@svi.edu.au.

Ernestina Schipani, Email: eschipan@iupui.edu.

Supplementary data

Supplemental Figure S1

The expression of ScaI and c-Kit in the lineage-negative CD45+CD11b+ cells (middle panel). The expression of F4/80 and Col 1 is also shown in lin-CD45+CD11b+ScaI+c-Kit+ (top right panel) and lin-CD45+CD11b+ScaI−cKit− (bottom right panel) cells.

mmc1.pdf (119.8KB, pdf)

References

  • 1.Caplan A.I. Mesenchymal stem cells. J Orthop Res. 1991;9:641–650. doi: 10.1002/jor.1100090504. [DOI] [PubMed] [Google Scholar]
  • 2.Ogawa M., LaRue A.C., Drake C.J. Hematopoietic origin of fibroblasts/myofibroblasts: its pathophysiologic implications. Blood. 2006;108:2893–2896. doi: 10.1182/blood-2006-04-016600. [DOI] [PubMed] [Google Scholar]
  • 3.Olmsted-Davis E.A., Gugala Z., Camargo F., Gannon F.H., Jackson K., Kienstra K.A., Shine H.D., Lindsey R.W., Hirschi K.K., Goodell M.A., Brenner M.K., Davis A.R. Primitive adult hematopoietic stem cells can function as osteoblast precursors. Proc Natl Acad Sci U S A. 2003;100:15877–15882. doi: 10.1073/pnas.2632959100. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Rogers I., Yamanaka N., Bielecki R., Wong C.J., Chua S., Yuen S., Casper R.F. Identification and analysis of in vitro cultured CD45-positive cells capable of multi-lineage differentiation. Exp Cell Res. 2007;313:1839–1852. doi: 10.1016/j.yexcr.2007.02.029. [DOI] [PubMed] [Google Scholar]
  • 5.Seta N., Kuwana M. Human circulating monocytes as multipotential progenitors. Keio J Med. 2007;56:41–47. doi: 10.2302/kjm.56.41. [DOI] [PubMed] [Google Scholar]
  • 6.Kuter D.J., Bain B., Mufti G., Bagg A., Hasserjian R.P. Bone marrow fibrosis: pathophysiology and clinical significance of increased bone marrow stromal fibers. Br J Haematol. 2007;139:351–362. doi: 10.1111/j.1365-2141.2007.06807.x. [DOI] [PubMed] [Google Scholar]
  • 7.Tefferi A. Primary myelofibrosis. Cancer Treat Res. 2008;142:29–49. [PubMed] [Google Scholar]
  • 8.Kumbasar B., Taylan I., Kazancioglu R., Agan M., Yenigun M., Sar F. Myelofibrosis secondary to hyperparathyroidism. Exp Clin Endocrinol Diabetes. 2004;112:127–130. doi: 10.1055/s-2004-817820. [DOI] [PubMed] [Google Scholar]
  • 9.Lim D.J., Oh E.J., Park C.W., Kwon H.S., Hong E.J., Yoon K.H., Kang M.I., Cha B.Y., Lee K.W., Son H.Y., Kang S.K. Pancytopenia and secondary myelofibrosis could be induced by primary hyperparathyroidism. Int J Lab Hematol. 2007;29:464–468. doi: 10.1111/j.1365-2257.2006.00877.x. [DOI] [PubMed] [Google Scholar]
  • 10.Rao D.S., Shih M.S., Mohini R. Effect of serum parathyroid hormone and bone marrow fibrosis on the response to erythropoietin in uremia. N Engl J Med. 1993;328:171–175. doi: 10.1056/NEJM199301213280304. [DOI] [PubMed] [Google Scholar]
  • 11.Weinstein L.S. G(s) alpha mutations in fibrous dysplasia and McCune-Albright syndrome. J Bone Miner Res. 2006;21(Suppl 2):P120–P124. doi: 10.1359/jbmr.06s223. [DOI] [PubMed] [Google Scholar]
  • 12.Calvi L.M., Sims N.A., Hunzelman J.L., Knight M.C., Giovannetti A., Saxton J.M., Kronenberg H.M., Baron R., Schipani E. Activated parathyroid hormone/parathyroid hormone-related protein receptor in osteoblastic cells differentially affects cortical and trabecular bone. J Clin Invest. 2001;107:277–286. doi: 10.1172/JCI11296. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Riminucci M., Robey P.G., Saggio I., Bianco P. Skeletal progenitors and the GNAS gene: fibrous dysplasia of bone read through stem cells. J Mol Endocrinol. 2011;45:355–364. doi: 10.1677/JME-10-0097. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Sacchetti B., Funari A., Michienzi S., Di Cesare S., Piersanti S., Saggio I., Tagliafico E., Ferrari S., Robey P.G., Riminucci M., Bianco P. Self-renewing osteoprogenitors in bone marrow sinusoids can organize a hematopoietic microenvironment. Cell. 2007;131:324–336. doi: 10.1016/j.cell.2007.08.025. [DOI] [PubMed] [Google Scholar]
  • 15.Tripodo C., Di Bernardo A., Ternullo M.P., Guarnotta C., Porcasi R., Ingrao S., Gianelli U., Boveri E., Iannitto E., Franco G., Florena A.M. CD146(+) bone marrow osteoprogenitors increase in the advanced stages of primary myelofibrosis. Haematologica. 2009;94:127–130. doi: 10.3324/haematol.13598. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Bellini A., Mattoli S. The role of the fibrocyte, a bone marrow-derived mesenchymal progenitor, in reactive and reparative fibroses. Lab Invest. 2007;87:858–870. doi: 10.1038/labinvest.3700654. [DOI] [PubMed] [Google Scholar]
  • 17.Ebihara Y., Masuya M., Larue A.C., Fleming P.A., Visconti R.P., Minamiguchi H., Drake C.J., Ogawa M. Hematopoietic origins of fibroblasts, II: in vitro studies of fibroblasts, CFU-F, and fibrocytes. Exp Hematol. 2006;34:219–229. doi: 10.1016/j.exphem.2005.10.008. [DOI] [PubMed] [Google Scholar]
  • 18.Larue A.C., Masuya M., Ebihara Y., Fleming P.A., Visconti R.P., Minamiguchi H., Ogawa M., Drake C.J. Hematopoietic origins of fibroblasts, I: in vivo studies of fibroblasts associated with solid tumors. Exp Hematol. 2006;34:208–218. doi: 10.1016/j.exphem.2005.10.009. [DOI] [PubMed] [Google Scholar]
  • 19.Kalajzic I., Kalajzic Z., Kaliterna M., Gronowicz G., Clark S.H., Lichtler A.C., Rowe D. Use of type I collagen green fluorescent protein transgenes to identify subpopulations of cells at different stages of the osteoblast lineage. J Bone Miner Res. 2002;17:15–25. doi: 10.1359/jbmr.2002.17.1.15. [DOI] [PubMed] [Google Scholar]
  • 20.Boban I., Jacquin C., Prior K., Barisic-Dujmovic T., Maye P., Clark S.H., Aguila H.L. The 3.6 kb DNA fragment from the rat Col1a1 gene promoter drives the expression of genes in both osteoblast and osteoclast lineage cells. Bone. 2006;39:1302–1312. doi: 10.1016/j.bone.2006.06.025. [DOI] [PubMed] [Google Scholar]
  • 21.Jurecic R., Van N.T., Belmont J.W. Enrichment and functional characterization of Sca-1+WGA+, Lin-WGA+, Lin-Sca-1+, and Lin-Sca-1+WGA+ bone marrow cells from mice with an Ly-6a haplotype. Blood. 1993;82:2673–2683. [PubMed] [Google Scholar]
  • 22.Lacey D.C., Simmons P.J., Graves S.E., Hamilton J.A. Proinflammatory cytokines inhibit osteogenic differentiation from stem cells: implications for bone repair during inflammation. Osteoarthritis Cartilage. 2009;17:735–742. doi: 10.1016/j.joca.2008.11.011. [DOI] [PubMed] [Google Scholar]
  • 23.Tang Y., Wu X., Lei W., Pang L., Wan C., Shi Z., Zhao L., Nagy T.R., Peng X., Hu J., Feng X., Van Hul W., Wan M., Cao X. TGF-beta1-induced migration of bone mesenchymal stem cells couples bone resorption with formation. Nat Med. 2009;15:757–765. doi: 10.1038/nm.1979. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Gensure R.C., Gardella T.J., Juppner H. Parathyroid hormone and parathyroid hormone-related peptide, and their receptors. Biochem Biophys Res Commun. 2005;328:666–678. doi: 10.1016/j.bbrc.2004.11.069. [DOI] [PubMed] [Google Scholar]
  • 25.Mendez-Ferrer S., Michurina T.V., Ferraro F., Mazloom A.R., Macarthur B.D., Lira S.A., Scadden D.T., Ma'ayan A., Enikolopov G.N., Frenette P.S. Mesenchymal and haematopoietic stem cells form a unique bone marrow niche. Nature. 2011;466:829–834. doi: 10.1038/nature09262. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Takayanagi H. New immune connections in osteoclast formation. Ann N Y Acad Sci. 2011;1192:117–123. doi: 10.1111/j.1749-6632.2009.05303.x. [DOI] [PubMed] [Google Scholar]
  • 27.Metz C.N. Fibrocytes: a unique cell population implicated in wound healing. Cell Mol Life Sci. 2003;60:1342–1350. doi: 10.1007/s00018-003-2328-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Phillips R.J., Burdick M.D., Hong K., Lutz M.A., Murray L.A., Xue Y.Y., Belperio J.A., Keane M.P., Strieter R.M. Circulating fibrocytes traffic to the lungs in response to CXCL12 and mediate fibrosis. J Clin Invest. 2004;114:438–446. doi: 10.1172/JCI20997. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Strieter R.M., Keeley E.C., Hughes M.A., Burdick M.D., Mehrad B. The role of circulating mesenchymal progenitor cells (fibrocytes) in the pathogenesis of pulmonary fibrosis. J Leukoc Biol. 2009;86:1111–1118. doi: 10.1189/jlb.0309132. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Suda R.K., Billings P.C., Egan K.P., Kim J.H., McCarrick-Walmsley R., Glaser D.L., Porter D.L., Shore E.M., Pignolo R.J. Circulating osteogenic precursor cells in heterotopic bone formation. Stem Cells. 2009;27:2209–2219. doi: 10.1002/stem.150. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Lotinum S., Sibonga J.D., Turner R.T. Evidence that the cells responsible for marrow fibrosis in a rat model for hyperparathyroidism are preosteoblasts. Endocrinology. 2005;146:4074–4081. doi: 10.1210/en.2005-0480. [DOI] [PubMed] [Google Scholar]
  • 32.Kuznetsov S.A., Riminucci M., Ziran N., Tsutsui T.W., Corsi A., Calvi L., Kronenberg H.M., Schipani E., Robey P.G., Bianco P. The interplay of osteogenesis and hematopoiesis: expression of a constitutively active PTH/PTHrP receptor in osteogenic cells perturbs the establishment of hematopoiesis in bone and of skeletal stem cells in the bone marrow. J Cell Biol. 2004;167:1113–1122. doi: 10.1083/jcb.200408079. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental Figure S1

The expression of ScaI and c-Kit in the lineage-negative CD45+CD11b+ cells (middle panel). The expression of F4/80 and Col 1 is also shown in lin-CD45+CD11b+ScaI+c-Kit+ (top right panel) and lin-CD45+CD11b+ScaI−cKit− (bottom right panel) cells.

mmc1.pdf (119.8KB, pdf)

Articles from The American Journal of Pathology are provided here courtesy of American Society for Investigative Pathology

RESOURCES