Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2016 Apr 15.
Published in final edited form as: Dev Biol. 2015 Feb 23;400(2):202–209. doi: 10.1016/j.ydbio.2015.02.011

BMP signaling mediated by constitutively active Activin type 1 receptor (ACVR1) results in ectopic bone formation localized to distal extremity joints

Shailesh Agarwal 1, Shawn J Loder 1, Cameron Brownley 1, Oluwatobi Eboda 1, Jonathan Peterson 1, Satoru Hayano 2, Bingrou Wu 3, Bin Zhao 3, Vesa Kaartinen 2, Victor C Wong 4, Yuji Mishina 2, Benjamin Levi 1
PMCID: PMC4397658  NIHMSID: NIHMS667004  PMID: 25722188

Abstract

BMP signaling mediated by ACVR1 plays a critical role for development of multiple structures including the cardiovascular and skeletal systems. While deficient ACVR1 signaling impairs normal embryonic development, hyperactive ACVR1 function (R206H in humans and Q207D mutation in mice, ca-ACVR1) results in formation of heterotopic ossification (HO). We developed a mouse line, which conditionally expresses ca-ACVR1 with Nfatc1-Cre+ transgene. Mutant mice developed ectopic cartilage and bone at the distal joints of the extremities including the interphalangeal joints and hind limb ankles as early as P4 in the absence of trauma or exogenous bone morphogenetic protein (BMP) administration. Micro-CT showed that even at later time points (up to P40), cartilage and bone development persisted at the affected joints most prominently in the ankle. Interestingly, this phenotype was not present in areas of bone outside of the joints – tibia are normal in mutants and littermate controls away from the ankle. These findings demonstrate that this model may allow for further studies of heterotopic ossification, which does not require the use of stem cells, direct trauma or activation with exogenous Cre gene administration.

Keywords: Nfatc1, ALK2, Acvr1, BMP receptor, heterotopic ossification, fibrodysplasia ossificans progressive, FOP, cartilage, bone, endochondral ossification

INTRODUCTION

The development of extremity long bones occur through a cartilage template which is replaced by bone in a process known as endochondral ossification (Kronenberg, 2003). Mesenchymal cells condense and then differentiate into chondrocytes. The cartilaginous mold is then invaded by blood vessels and replaced by bone. When dysregulated, such cartilage deposition and subsequent osteogenic differentiation can occur in extra-skeletal sites called heterotopic ossification (HO). One well-described example of HO is fibrodysplasia ossificans progressiva (FOP) which is an inherited autosomal dominant disorder characterized by a mutation in the Activin A receptor, type I (ACVR1) gene, which encodes the type I bone morphogenetic protein receptor ACVR1 which is also called ALK2 (Shore et al., 1997; Shore et al., 2006). Global knockout of ACVR1 results in early embryonic lethality during gastrulation (Komatsu et al., 2007; Mishina et al., 1999). The ACVR1 R206H mutation in patients changes the interaction of ACVR1 with the inhibitory protein FKBP12 to destabilize the protein structure making ACVR1 hyper activated (Chaikuad et al., 2012; Culbert et al., 2014; Groppe et al., 2007). This mutation activates neighboring protein kinase domains and induces downstream signal transduction by phosphorylation of canonical Smads (Smad1, Smad5, and Smad8) or components of the mitogen activated protein kinase (MAPK) pathway regulating downstream gene transcription (Derynck and Zhang, 2003). The exact cell lineage with dysregulated ACVR1 which plays a central role in heterotopic ossification, however, is still unknown. In mice, global expression of the constitutively active ACVR1 (ca-ACVR1) results in uniform in utero lethality (Fukuda et al., 2006). However studies of cells derived from these mice have confirmed elevated SMAD 1/5/8 suggestive of a pro-osteogenic phenotype (Fukuda et al., 2006; Shimono et al., 2011; Yu et al., 2008).

To confirm that mice with the ca-ACVR1 (Q207D) mutation (ca-ACVR1WT/flox) develop ectopic bone, we previously injected adenovirus-expressing Cre (Ad.Cre) into the hind limb of conditional ca-ACVR1 Q207D mutants (Yu et al., 2008). These mice develop ectopic bone limited to the injection site.

In this study, we use the nuclear factor associated T-cell c1 (Nfatc1) gene as a driver for Cre expression to direct tissue-specific expression of ca-ACVR1 in cells of mesodermal lineage. NFATc1 dysregulation in osteoblasts has been shown to alter bone and cardiac development (Maoqiang et al., 2010; Wu et al., 2012). NFATc1 is expressed in cells important for skeletogenesis. In our model, we found robust heterotopic bone formation only at joints, most notably the ankle without alteration in skeletal bone thickness. Additionally the region of HO was enriched for bone progenitor cells and cells harvested from the heterotopic bone were more osteogenic in vitro compared to cells from skeletal bone. This phenotype holds several advantages for studies into the pathogenesis and possible treatment of heterotopic ossification since it does not require additional supplementation of osteogenic cytokines or external trauma.

METHODS

Ethics statement

The experiments described in this study were approved by the University Committee on Use and Care of Animals of the University of Michigan-Ann Arbor (Protocol Number: #09944, #05716). This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health.

Mice

Mice carrying the conditional constitutively active allele of ACVR1 (ACVR1 carrying Q207D mutation, ca-ACVR1) and Nfatc1-Cre transgenic mice were used and genotyped as previously described (Fukuda et al., 2006; Thomas et al., 2014; Wu et al., 2012). Resulting pups carrying both transgenes (ca-ACVR1fx/WT/Nfatc1-Cre+) were used as experimental. Either Cre−, or ca-ACVR1WT/WT/Nfatc1-Cre+ littermates were used as controls. To assess the effect of ca-ACVR1 expression on HO formation after trauma, we used a tamoxifen-inducible Cre transgenic mouse line under control of ubiquitin promoter with the ca-ACVR1 mutation (Ub.Cre-ERT/ca-ACVR1) and performed an Achilles tenotomy given our previous experience with this model (Peterson et al., 2014). Weights and lengths were measured for wild type and ca-ACVR1fx/WT/Nfatc1-Cre mice at the same timepoints after birth. Length and weight were obtained for mice at P4 (n=3), P7 (n=3), and P22 (n=4).

Computed Tomography (CT) analysis

Mouse hind limbs were imaged with µCT at various times from P4 to P40 (µCT: Siemens Inveon, using 80kVp, 80mA and 1100 ms exposure). Images were reconstructed and ectopic bone volume formation was calculated as the difference between the lower hind limb bone volume of ca-ACVR1fx/WT/Nfatc1-Cre and littermate control mice of the same age over the same length. Whole-body µCT and nano-CT (GE Nanotom S) was performed at P20 and P11 respectively in ca-ACVR1fx/WT/Nfatc1-Cre and wild type mice. All scans were analyzed with threshold Hounsfield units (HU) of 800, 1250, or 1800 as indicated. Tibial cortical thickness was measured in cross-sections above the level of the tibia-fibula fuse point in both ca-ACVR1fx/WT/Nfatc1-Cre and littermate controls to avoid regions of obvious ectopic bone formation. Bone density was obtained at P4 (n=2 mutant hindlimbs, 2 littermate control hindlimbs), P20 (n=6 mutant hindlimbs, 4 littermate control hindlimbs), and P40 (n=2 mutant hindlimbs, 2 littermate control hindlimbs). Finally, for bone volume quantification at the ankle, one mutant and one littermate control mouse were used to compare the right and the left sides.

Ankle range-of-motion (ROM) analysis

At P4, P22, and P40, mutant mice and littermate controls were briefly anesthetized and assessed for ROM at the tenotomy site by extending the ankle with a 75-g weight attached to the hindpaw to ensure full, uniform extension. Photographs were taken with the extended ankle centered on a disk of standardized size and distance. The angle of maximum extension was then assessed by two independent, blinded observers using the ruler tool in Adobe Photoshop (Peterson et al., 2014). ROM was obtained at P7 (n=2 mutant hindlimbs, 2 littermate control hindlimbs), P11 (n=2 mutant hindlimbs, 2 littermate control hindlimbs), and P22 (n=4 mutant hindlimbs, 3 littermate control hindlimbs).

Whole mount staining

Alcian blue/alizarin red and alcian blue staining alone were performed on P11 and P20 whole mount ca-ACVR1flox/WT/Nfatc1-Cre+ and littermate control mice respectively to assess for cartilage and bony distribution throughout the skeleton (Ovchinnikov et al., 2006). Images were taken using a Leica MXSL3 Stereo/Dissecting Microscope with Olympus DP70 digital camera. Images were digitally edited using Adobe Photoshop to remove remnant peripheral fascia.

Histology and immunohistochemistry

Histologic evaluation of ca-ACVR1flox/WT/Nfatc1-Cre+ and littermate control mice were euthanized at various times from P4 to P40. Hind limbs were fixed overnight in formalin overnight and subsequently decalcified in 19% EDTA solution for 7 weeks in 4C. Hind limbs were embedded in paraffin, and 5–7 µm sections were cut and mounted on Superfrost plus slides (Fisher) and stored at room temperature. Haematoxylin & eosin and Movat’s pentachrome staining were performed of the ankle region. Immunohistochemistry staining was performed on rehydrated wax sections with the following primary antibodies: anti -SOX-9, anti-Osteocalcin, anti-HIF1alpha, and anti-Ki67 of the ankle region. Appropriate dilutions were determined prior to achieving final images. The appropriate secondary biotinylated antibody (Vector Laboratories, Burlingame, CA) was applied and visualized with diaminiobenzidine (Zymed Laboratories, South San Francisco, CA).

Flow cytometry

Flow cytometry was performed of the ankle, knee, and areas of heterotopic bone/cartilage present in mutants. The same amount of tissue was harvested from corresponding anatomic regions of mutant and littermate controls based on anatomic landmarks (e.g. calcaneus to the fuse point of the tibia and fibula). All samples were taken from the distal hindlimb of mutant and littermate controls. Tissue was digested for 120 minutes in 0.75% Collagenase 2 (Sigma-Aldrich) in Hanks Balanced Salt Solution (HBSS) at 37C under constant agitation. Samples were filtered using a 70-micron sterile strainer and centrifuged at 800 rpm for 5 minutes before removing the supernatant and washing in HBSS. This process was repeated three times before incubation with fluorescently labeled antibodies to determine a bone progenitor subpopulation defined as AlphaV+/CD105+/Tie2−/CD45−/CD90−/BP1-(Chan et al., 2013). Antibodies used included: AlphaV-PE (12-0512-83, eBioscience), Tie2-AlexaFluor 488 (334208, BioLegend), CD45-APC (17-0451-83, eBioscience), CD105-eFluor450 (48-1057-42, eBioscience), CD90-PE-Cyanine7 (25-0902-82, eBioscience), and BP1-Biotin (13-5891-85, eBioscience) conjugated with Streptavidin-PerCP-Cyanine5.5 (45-4317-82, eBioscience). Values were normalized to total number of viable cells to control for any possible differences in the amount of tissue harvested. Following thirty minutes of incubation at 4 degrees C, sample were washed three times as described before and filtered through a 45 micron mesh filter before being run on a FACSAria II (BD Biosciences) Cell Sorter at the University of Michigan Flow Cytometry Core. Data was then analyzed using the FlowJo software.

Osteogenic differentiation

Cells from control and mutant tibia and from the heterotopic bone/cartilage were harvested as described above and cultured in DMEM with 10% fetal bovine serum. Once p0 cells were confluent (2 weeks), they were lifted and split into an experiment. Cells were then plated at a density of 100,000 cells/well in 6-well plates and cultured in osteogenic differentiation medium (ODM: DMEM supplemented with 10% FBS, 1% penicillin/streptomycin, 10mM β-glycerophosphate, 100 µg/mL ascorbic acid. Invitrogen, Grand Island, NY). Cells were cultured in ODM for 6 days prior to alkaline phosphatase staining to assess osteogenic differentiation.

Statistical tests

Two-tailed Student’s t-test with unequal variance was used to compare mutant and littermate control measures including bone density, tibial thickness, and range-of-motion. All statistical analyses were performed using SPSS v 19. Statistical significance was defined prior to experimental analysis as p < 0.05.

RESULTS

Gross morphologic differences between ca-ACVR1flox/WT/Nfatc1-Cre+ mice and controls

Previously, we have shown that mice which globally express ca-ACVR1 die in utero. We found that mice expressing ca-ACVR1 in Nfatc1-Cre+ cells (mutant mice hereafter) survive at least 40 days after birth (P40). On initial observation, we found altered body morphology in mutant mice when compared with littermate controls. Mutant mice were smaller and had shorter tails (Fig 1A,B). At P22, mutant mice were shorter in the cranial-caudal direction (11.7 cm v. 8.75 cm, p<0.05) (Fig 1A, B, K). Mutant mice also had lower weights when compared with wild type mice at P22 (9.5 g v. 6.5 g, p<0.05) (Fig 1L).

Figure 1.

Figure 1

Open in a new tab

Anatomic morphology of ca-ACVR1W/flox/Nfatc1-Cre+ mutant and control mice. (A,B) Control (A) and mutant (B) mouse photos at P22, (C,D) whole body µCT images of control (C) and mutant (D) at P22 (E,F) Nano-CT images of control (E) and mutant mice (F) at P11, (G,H) whole mount alcian blue stain with wrist and ankle magnified (black arrows point to alcian blue stain) of control (G) and mutant (H) mice, and (I,J) Alizarin red/Alcian blue images of anatomically corresponding regions of control (I) and mutant (J). (K) At P22, mutant mice were significantly shorter in the cranio-caudal direction (p<0.05). (L) Similarly, at P22, mutant mice weighed significantly less (p<0.05). Length and weight were obtained for mice at P4 (n=3), P7 (n=3), and P22 (n=4). Px = day ‘x’ after birth. * indicates p value <0.05. µCT = micro-computed tomography.

Whole body micro-CT, at 800 HU mutant mice showed increased opacity specifically along the ankle region, suggesting that mutant mice had increased presence of lower density bone and cartilage at the ankle (Fig 1C,D). Nano-CT was next used to analyze the specific joints. Distal joints including the interphalangeal joints of the forelimb and hindlimbs, wrists, and ankles all had significant ectopic bone and cartilage presence (Fig. 1E,F). However, this was diminished at more proximal joints including the knee, elbows, hips (Fig. 1E,F).

Whole mount alcian blue staining was performed at P20 in mutant and control mice. We found more robust staining in mutant mice, most notably at the upper and lower extremity joint spaces and regions of cartilage formation (Fig 1G,H). Alizarin red/alcian blue staining (P11), and alcian blue staining alone (P20) were compared at each of the major joints with increased bone most notable at the ankle (Fig. I, J). Additionally, we did not see any ectopic cartilage or bone at non-joint regions of bone (Fig. 1C–J).

Hindlimb Morphology

At P4, the mutants and wild type mice could be differentiated based on the presence of clinodactyly, or internal rotation of the digits (Fig 2A,B). MicroCT scans of the hindlimb confirmed the presence of ectopic bone at the ankle of mutant mice (Fig 2C,D). Range of motion evaluation performed at the ankle showed decreased ROM in the mutant mice when compared with wild type mice at each time (p<0.05; Fig 2E–G).

Figure 2.

Figure 2

Open in a new tab

Morphology, CT imaging and range of motion of ca-ACVR1W/flox/Nfatc1-Cre+ mutant and control mice. (A,B) Hindlimb paw morphology of littermate control (A) and mutant (B) at P7. (C,D) Representative ankle micro-CT of littermate control (C) and mutant (D) at P20. (E,F) Range of motion measurement for littermate control (E) and mutant mice (F) at P40. (G) Range of motion measurements from P7 to P22 of littermate control (white) and mutants (blue) showed significantly less ROM in mutant hindlimbs (p<0.05). (H,I) Serial micro-CT scans of the hindlimb for littermate control (H) and mutant (I) mice from P4 to P40. (J,K) Representative tibial cross-sections at the fusepoint and endpoint of littermate control (J) and mutant mice (K). (L) Mean tibial thickness of littermate control (white) and mutant (blue) mice at P20 and P40 were similar. (M) Symmetry of heterotopic bone at left and right ankles at P40. (N,O) Quantification of high (N) and low (O) density bone at the ankle. Bone density and tibial thickness were obtained at P4 (n=2 mutant hindlimbs, 2 littermate control hindlimbs), P20 (n=6 mutant hindlimbs, 4 littermate control hindlimbs), and P40 (n=2 mutant hindlimbs, 2 littermate control hindlimbs). * indicates p<0.05. Px = day ‘x’ after birth. L = left, R = right.

Serial microCT scans were taken at P4, P20, and P40 showing progression of ectopic bone growth (Fig 2H,I). In ca-ACVR1flox/WT/Nfatc1-Cre+ mice, over time there was a gradual increase in the amount of ectopic cartilage and low and high density bone present specifically at the ankle joint and interphalangeal joints of the hind limb (Fig 2N,O). This differing bone density likely represents the early cartilage precursor prior to the presence of mature heterotopic osteoid. Among wild type mice, the volume of high-density bone volume (HU 1,800+) increased over time while low-density bone identified with Hounsfield Units from 800–1,800 remained relatively even over time (Fig 2N,O). There was no ectopic bone (extending beyond the outer cortex) noted along the middle third of the tibia from any of the mutant mice. Mean tibial bone thickness was assessed over a 2-cm distance starting from the fuse-point of the tibia and fibula moving cranially; mean cortical thickness was similar for mutant and wild type P20 mice (0.10 mm v. 0.12 mm) and for mutant and wild type P40 mice (0.24 mm v. 0.24 mm) (Fig 2J–L). This finding suggests that the cells responsible for the observed HO phenotype in this model do not alter skeletal bone development. Thus, the pathologic bone formation appears isolated to joints and does not extend into the native appendicular skeleton. Mutants showed different results, as both high- and low-density bone increased over time. Ectopic bone and cartilage formation was relatively symmetric across the right and left sides in mutant mice (Fig 2M).

Histologic evaluation confirms presence of ectopic cartilage and bone

To assess if cartilage preceded the heterotopic bone deposition, pentachrome staining was performed on the hind limbs of mutant and corresponding control mice at the ankle. By P13, we noted that the ankle of the mutant mice showed evidence of significant cartilage deposition whereas areas of non-heterotopic bone in control mice lacked chondrogenesis (Fig 3A). Immunostaining with SOX9 confirmed that these areas were chondrogenic (Fig 3B). Ki67 staining confirmed that the chondrocytes are proliferative (Fig 3C). Thus, the heterotopic ossification seen in this mutant model has a proliferative chondrogenic precursor. Immunostaining with Osteocalcin confirmed that these regions undergo osteogenic differentiation as the majority of the cells express OCN (Fig 3D).

Figure 3.

Figure 3

Open in a new tab

Histology through the ankle of ca-ACVR1W/flox/Nfatc1-Cre+ mutant and control mice (20× magnification). (A) Pentachrome stain of normal cartilage and bone from littermate control and heterotopic ossification (HO) from mutant mice. Areas of ectopic bone and cartilage in mutant mice demonstrate strong blue stain at P3 and P7, and evidence of ossification at P13. (B) Sox-9 stain (cartilage marker) of normal cartilage and bone from littermate control and HO from mutant mice. Robust staining in the ectopic bone/cartilage of mutant mice. (C) Ki67 stain (proliferation marker) of normal cartilage and bone from littermate control and HO from mutant mice. Robust Ki67 staining in the ectopic bone/cartilage of mutant mice. (D) Osteocalcin stain of normal cartilage and bone from littermate control and HO from mutant mice. Robust staining in the ectopic bone/cartilage of mutant mice becomes more prominent at later time points. HO = heterotopic ossification. Px = day ‘x’ after birth.

Composition of osteoblasts in mutant and control bone

To further characterize the cells present at the sites of heterotopic ossification, flow cytometry performed on the normal ankle and knee of control and mutant mice, as well as the ectopic bone/cartilage of mutant mice (Fig 4A). While joints of the distal hindlimb in both control and wild type mice had similar presence of non-hematopoetic bone progenitor cells (AlphaV+/Tie2−/CD45−/CD105+/BP1−/CD90−), ectopic bone/cartilage of mutant mice showed nearly 9-fold enrichment of these cells (Fig 4B, C). Thus, there is a clear enrichment of cells with this flow profile which have been shown to represent skeletal progenitor (bone, cartilage and stromal progenitor) cells.(Chan et al., 2013)

Figure 4.

Figure 4

Open in a new tab

Cell analysis of the ankle in ca-ACVR1W/flox/Nfatc1-Cre+ and wild type mice using flow cytometry and osteogenic differentiation assays. (A) Sites of tissue harvest for flow cytometry from the ankle and knee of littermate controls and mutant mice, and the ectopic bone/cartilage of mutant mice. (B) Identification of AlphaV+/CD105+ population. Sub-population analysis for CD45−/Tie2- cells. Further sub-population analysis for CD90−/BP1−. (C) Enrichment of bone progenitor cells (AlphaV+/CD105+/CD45−/Tie2−/CD90−/BP1−) in the ectopic bone/cartilage of mutant mice. Bone progenitor population is similar in mutant mice in the region of non-heterotopic bone (mid-tibia) and in the same region of littermate control mice. (D) Alkaline phosphatase staining of osteoblasts derived from anatomically normal tibia of littermate control and mutant mice, and osteoblasts derived from ectopic bone/cartilage (HO) of mutant mice. (E) Osteoblasts from anatomically normal bone of mutant mice and from heterotopic bone had significantly more alkaline phosphatase staining than osteoblasts derived from littermate control mice (p<0.05). (F) Expression of ACVR1 (top row) within osteoblasts from littermate control mice, anatomically normal bone of mutant mice, and heterotopic bone were similar while pSmad 1/5 (middle row) was elevated in the osteoblasts of anatomically normal bone from mutant mice and heterotopic bone. (G) ACVR1 and pSmad 1/5 expression are similar in the osteoblasts (black arrows) and periosteum (red arrows) of anatomically normal bone from mutant and littermate control mice. Heterotopic bone (HO) (black arrows) displays evidence of ACVR1 expression and pSmad1/5 expression as well suggesting that ACVR1 signaling is likely responsible for the observed phenotype. (H) The ligament of littermate control and mutant mice have similar expression of ACVR1. However ligament in mutant mice have significantly more pSmad 1/5 staining suggesting a key difference between mutant and control mice. A+K = Ankle and knee. HO = heterotopic ossification. ALP = alkaline phosphatase.

Differences in osteogenic potential of cells isolated from the heterotopic bone

Though we have shown a group of cells that were enriched within our heterotopic bone, we next set out to better understand the actual characteristics of cells isolated from heterotopic bone compared to cells from skeletal bone in vitro. Normal osteoblasts harvested from the tibia of Nfatc1-Cre+ mutant mice showed more alkaline phosphatase staining after culture in osteogenic differentiation medium when compared with control osteoblasts. Furthermore, cells harvested from the site of ectopic bone/cartilage at the ankle exhibited even more alkaline phosphatase staining than either control or mutant osteoblasts (p<0.05, Fig 4D,E). Thus, the cells from this heterotopic bone remain more osteogenic even when cultured ex vivo.

We next studied the levels of ACVR1 and pSmad 1/5 in vitro using cells isolated and cultured from the regions of heterotopic ossification and normal bone osteoblasts in mutant mice and normal bone osteoblasts in littermate controls. Western blot showed similar levels of ACVR1 protein in all three groups of cells after in vitro culture (Fig. 4F). However, pSmad 1/5 levels were elevated in cells from heterotopic bone and elevated to a lesser degree in normal bone osteoblasts from mutant mice, when compared with normal bone osteoblasts from littermate controls (Fig. 4F).

In vivo HO formation and levels of ACVR1 and pSmad 1/5

Following this, we evaluated levels of ACVR1 and pSmad 1/5 in vivo using immunohistochemistry staining. In mutant mice, we noted that anatomically similar regions of normal bone and heterotopic bone in mutant mice and normal bone in littermate controls had similar levels of ACVR1 (Fig. 4G). Downstream ACVR1 signaling as examined by pSmad 1/5, was present in the anatomically normal bone of mutant and littermate control mice, as well as the HO region of the mutant mice (Fig. 4G). Interestingly, we noted pSmad 1/5 staining within ligaments and along periosteum of the mutant mice, suggesting that these cells were most influenced by the mutation when compared with control mice (Fig. 4G,H). We did not see major differences in ACVR1 staining in the ligaments of mutant or littermate control mice however (Fig. 4G,H).

We next set out to assess the effect of trauma on mice with a caALK-2 mutation using our proven burn/tenotomy model.(Peterson et al., 2013) Mice with tamoxifen-inducible ca-ACVR1 expression (Ub.Cre-ERT/ca-ACVRflox/WT) which have been administered tamoxifen (TM+) exhibit increased HO formation at the site of Achilles’ tendon transection 9 weeks after injury when compared with (Ub.Cre-ERT/ca-ACVRWT/WT) mice treated tamoxifen tamoxifen (TM−). (Supp. Fig. 1) This injury could not be performed in the ca-ACVR1flox/WT/Nfatc1-Cre+ mouse line as mice at the appropriate age had excessive HO making tenotomy technically impossible.

DISCUSSION

In this study, we describe a novel model of ectopic bone and cartilage development. In our model, expression of the mutated ACVR1 gene (ca-ACVR1) in Nfatc1-Cre expressing cells is not embryonically lethal (Fukuda et al., 2006). Initial studies examining the role of ca-ACVR1 were done in mice generated by inserting human caACVR1 cDNA (Q207D) into the cloning site.(Fukuda et al., 2006) When the transgene is expressed globally during development, mice are arrested during gestation. Subsequent studies circumvented this lethality by inducing overexpression of ACVR1 postnatally through injection of Ad.Cre and the induction of trauma.(Yu et al., 2008) Interestingly, we found that ca-ACVR1W/flox/Nfatc1-Cre+ mice surivive and develop ectopic bone which is more pronounced in the distal joints such as the ankle and distal radioulnar joints. Mutant mice develop ectopic cartilage and bone in a symmetric pattern across both hind limbs. This heterotopic bone has a cartilage precursor and the sites of heterotopic ossification are enriched for bone/cartilage/stromal progenitor cells. Many previous studies have described “HO progenitor cells” however, these cells have varied widely based on the model and species examined.(Downey et al., 2015; Kan and Kessler, 2014; Medici et al., 2010) Rather than identify “the” progenitor, we have chosen to look at the enrichment of a cell population that has been shown to represent a non-hematopoetic skeletal bone progenitor in endochondral ossification.(Chan et al., 2013) In the model described here the heterotopic bone/cartilage developed without any external trauma. Previous studies of mice with the Q207D mutation have required external trauma and developed intramuscular bone. In this model, the heterotopic bone does not occur within the muscle which might offer clues as to the true niche necessary for HO formation. (Yu et al., 2008) However, the mice were allowed to ambulate ad libitum, which may introduce local mechanical stress responsible for more heterotopic bone/cartilage at the interphalangeal, wrist, and ankle joints. These findings are also supported by trauma models in which Achilles’ tendon injury results in robust HO formation (Lin et al., 2010; McClure, 1983; Peterson et al., 2014). Although the role of ACVR1 remains unclear in HO after trauma, tamoxifen-inducible ca-ACVR1-expressing mouse (Ub.Cre-ERT/ca-ACVR1flox/WT) treated with tamoxefin develop more HO at the site of Achilles’ tendon transection compared with Ub.Cre-ERT/ca-ACVRWT/WT mice treated with tamoxefin (Supp Fig. 1).

Interestingly, we noted that the mutant osteoblasts from the heterotopic bone were more osteogenic than mutant osteoblasts cells from anatomically normal bone. It is possible that the mutant osteoblasts from anatomically normal bone exhibit more alkaline phosphatase stain because of the increased potency of these cells in the mutant model. Although there are fewer osteoprogenitor cells in the non-heterotopic bone regions of our mutant mice, these cells still likely retain increased potency conferred by the original mutation. We noted that the in vitro expression levels of ACVR1 protein are similar among cells cultured from anatomically normal bone or heterotopic bone from mutant mice and from anatomically normal bone from littermate controls. Therefore, it is less likely a difference in expression level of ACVR1 which results in the observed phenotype. However it is possible that cells expressing ca-ACVR1 may be more prevalent in heterotopic bone than skeletal bone. Our finding of increased pSmad 1/5 in heterotopic bone cells, and to a lesser extent in mutant osteoblasts from anatomically normal bone, suggest that the mutation results in hyperactive ACVR1 as intended. Recent studies of trauma induced heterotopic ossification have also demonstrated increased BMP ligand in human HO. (Evans et al., 2014) Furthermore, studies of overexpression of BMP-2 and BMP-4 have been used to study heterotopic ossification. (Dilling et al., 2010; Kan et al., 2004) None of these studies, however, have targeted the BMP pathway only in cells expressing Nfatc1 which may explain the unique phenotype.

Finally, we found that in vivo expression levels of ACVR1 were similar in regions such as normal bone, periosteum, and ligaments. Although osteoblasts and periosteum of littermate control and mutant mice both had similar pSmad1/5 levels, we found almost no pSmad 1/5 expression within the ligaments of littermate control mice. This was in stark contrast to mutant mice which showed heavy staining of ligaments, suggesting that these regions were strongly influenced by the mutation, and providing a possible explanation for our observed phenotype. In particular, ligaments are ostensibly located within the joints including ankle, and to a lesser extent in the wrist, knees, and elbows. Therefore, our finding of significant heterotopic bone formation within all of these regions supports the idea that ligaments may be contributing to our observed phenotype; that the ankle generally has more ligamentous attachments than other locations could explain why the ankle shows the most heterotopic bone. These findings coupled with the model described here suggest that it is the downstream effects of ACVR1 expression within ligaments which may be responsible for the phenotype observed in our model.

This model may serve as an appropriate model to study the early changes associated with heterotopic ossification and the stimuli which initiate ectopic bone formation. This model may also allow studies of treatment strategies to prevent heterotopic ossification.

Supplementary Material

Supplmental Figure 1

ca-ACVR1 expression potentiates HO formation 9 weeks after a dorsal burn injury and Achilles tenotomy. (A) MicroCT images of mice 9 weeks after a dorsal burn injury and Achilles tenotomy without tamoxifen-inducible ca-ACVR1 expression (Ub.Cre-ERT/ca-ACVRWT/WT) treated tamoxifen. (B) MicroCT images of mice 9 weeks after a dorsal burn injury and Achilles tenotomy with tamoxifen-inducible ca-ACVR1 expression (Ub.Cre-ERT/ca-ACVRflox/WT) which have been administered tamoxifen. TM= tamoxifen.

Research Highlights.

  • ca-ACVR1fx/WT/Nfatc1-Cre+ mice develop heterotopic ossification at joins without trauma.

  • Heterotopic bone in a ca-ACVR1fx/WT/Nfatc1-Cre+ mice occur by endochondral ossification.

  • ca-ACVR1fx/WT/Nfatc1-Cre+ mouse skeletal bones have similar thickness to littermate controls.

  • Cells from heterotopic bone are more osteogenic in vitro compared to cells from skeletal bone.

  • Heterotopic bone in a ca-ACVR1fx/WT/Nfatc1-Cre+ mice are enriched with bone progenitor cells.

Acknowledgments

Funding: BL funded by 1K08GM109105-01, and Plastic Surgery Foundation National Endowment Award. SA funded by the NIH LRP and Coller Society. SJL funded by HHMI. YM funded by R01DE020843. VK funded by R01DE013085

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

REFERENCES

  1. Chaikuad A, Alfano I, Kerr G, Sanvitale CE, Boergermann JH, Triffitt JT, von Delft F, Knapp S, Knaus P, Bullock AN. Structure of the bone morphogenetic protein receptor ALK2 and implications for fibrodysplasia ossificans progressiva. The Journal of biological chemistry. 2012;287:36990–36998. doi: 10.1074/jbc.M112.365932. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Chan CK, Lindau P, Jiang W, Chen JY, Zhang LF, Chen CC, Seita J, Sahoo D, Kim JB, Lee A, Park S, Nag D, Gong Y, Kulkarni S, Luppen CA, Theologis AA, Wan DC, DeBoer A, Seo EY, Vincent-Tompkins JD, Loh K, Walmsley GG, Kraft DL, Wu JC, Longaker MT, Weissman IL. Clonal precursor of bone, cartilage, and hematopoietic niche stromal cells. Proceedings of the National Academy of Sciences of the United States of America. 2013;110:12643–12648. doi: 10.1073/pnas.1310212110. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Culbert AL, Chakkalakal SA, Theosmy EG, Brennan TA, Kaplan FS, Shore EM. Alk2 regulates early chondrogenic fate in fibrodysplasia ossificans progressiva heterotopic endochondral ossification. Stem cells. 2014;32:1289–1300. doi: 10.1002/stem.1633. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Derynck R, Zhang YE. Smad-dependent and Smad-independent pathways in TGF-beta family signalling. Nature. 2003;425:577–584. doi: 10.1038/nature02006. [DOI] [PubMed] [Google Scholar]
  5. Dilling CF, Wada AM, Lazard ZW, Salisbury EA, Gannon FH, Vadakkan TJ, Gao L, Hirschi K, Dickinson ME, Davis AR, Olmsted-Davis EA. Vessel formation is induced prior to the appearance of cartilage in BMP-2-mediated heterotopic ossification. Journal of bone and mineral research : the official journal of the American Society for Bone and Mineral Research. 2010;25:1147–1156. doi: 10.1359/jbmr.091031. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Downey J, Lauzier D, Kloen P, Klarskov K, Richter M, Hamdy R, Faucheux N, Scime A, Balg F, Grenier G. Prospective heterotopic ossification progenitors in adult human skeletal muscle. Bone. 2015;71:164–170. doi: 10.1016/j.bone.2014.10.020. [DOI] [PubMed] [Google Scholar]
  7. Evans KN, Potter BK, Brown TS, Davis TA, Elster EA, Forsberg JA. Osteogenic gene expression correlates with development of heterotopic ossification in war wounds. Clinical orthopaedics and related research. 2014;472:396–404. doi: 10.1007/s11999-013-3325-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Fukuda T, Scott G, Komatsu Y, Araya R, Kawano M, Ray MK, Yamada M, Mishina Y. Generation of a mouse with conditionally activated signaling through the BMP receptor, ALK2. Genesis. 2006;44:159–167. doi: 10.1002/dvg.20201. [DOI] [PubMed] [Google Scholar]
  9. Groppe JC, Shore EM, Kaplan FS. Functional modeling of the ACVR1 (R206H) mutation in FOP. Clinical orthopaedics and related research. 2007;462:87–92. doi: 10.1097/BLO.0b013e318126c049. [DOI] [PubMed] [Google Scholar]
  10. Kan L, Hu M, Gomes WA, Kessler JA. Transgenic mice overexpressing BMP4 develop a fibrodysplasia ossificans progressiva (FOP)-like phenotype. The American journal of pathology. 2004;165:1107–1115. doi: 10.1016/S0002-9440(10)63372-X. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Kan L, Kessler JA. Evaluation of the cellular origins of heterotopic ossification. Orthopedics. 2014;37:329–340. doi: 10.3928/01477447-20140430-07. [DOI] [PubMed] [Google Scholar]
  12. Komatsu Y, Scott G, Nagy A, Kaartinen V, Mishina Y. BMP type I receptor ALK2 is essential for proper patterning at late gastrulation during mouse embryogenesis. Developmental dynamics : an official publication of the American Association of Anatomists. 2007;236:512–517. doi: 10.1002/dvdy.21021. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Kronenberg HM. Developmental regulation of the growth plate. Nature. 2003;423:332–336. doi: 10.1038/nature01657. [DOI] [PubMed] [Google Scholar]
  14. Lin L, Shen Q, Xue T, Yu C. Heterotopic ossification induced by Achilles tenotomy via endochondral bone formation: expression of bone and cartilage related genes. Bone. 2010;46:425–431. doi: 10.1016/j.bone.2009.08.057. [DOI] [PubMed] [Google Scholar]
  15. Maoqiang L, Zhenan Z, Fengxiang L, Gang W, Yuanqing M, Ming L, Xin Z, Tingting T. Enhancement of osteoblast differentiation that is inhibited by titanium particles through inactivation of NFATc1 by VIVIT peptide. Journal of biomedical materials research. Part A. 2010;95:727–734. doi: 10.1002/jbm.a.32891. [DOI] [PubMed] [Google Scholar]
  16. McClure J. The effect of diphosphonates on heterotopic ossification in regenerating Achilles tendon of the mouse. J Pathol. 1983;139:419–430. doi: 10.1002/path.1711390403. [DOI] [PubMed] [Google Scholar]
  17. Medici D, Shore EM, Lounev VY, Kaplan FS, Kalluri R, Olsen BR. Conversion of vascular endothelial cells into multipotent stem-like cells. Nature medicine. 2010;16:1400–1406. doi: 10.1038/nm.2252. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Mishina Y, Crombie R, Bradley A, Behringer RR. Multiple roles for activin-like kinase-2 signaling during mouse embryogenesis. Developmental biology. 1999;213:314–326. doi: 10.1006/dbio.1999.9378. [DOI] [PubMed] [Google Scholar]
  19. Ovchinnikov DA, Selever J, Wang Y, Chen YT, Mishina Y, Martin JF, Behringer RR. BMP receptor type IA in limb bud mesenchyme regulates distal outgrowth and patterning. Developmental biology. 2006;295:103–115. doi: 10.1016/j.ydbio.2006.03.013. [DOI] [PubMed] [Google Scholar]
  20. Peterson JR, De La Rosa S, Eboda O, Cilwa KE, Agarwal S, Buchman SR, Cederna PS, Xi C, Morris MD, Herndon DN, Xiao W, Tompkins RG, Krebsbach PH, Wang SC, Levi B. Treatment of heterotopic ossification through remote ATP hydrolysis. Science translational medicine. 2014;6:255ra132. doi: 10.1126/scitranslmed.3008810. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Peterson JR, Okagbare PI, De La Rosa S, Cilwa KE, Perosky JE, Eboda ON, Donneys A, Su GL, Buchman SR, Cederna PS, Wang SC, Kozloff KM, Morris MD, Levi B. Early Detection of Burn Induced Heterotopic Ossification using Transcutaneous Raman Spectroscopy. Bone. 2013 doi: 10.1016/j.bone.2013.01.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Shimono K, Tung WE, Macolino C, Chi AH, Didizian JH, Mundy C, Chandraratna RA, Mishina Y, Enomoto-Iwamoto M, Pacifici M, Iwamoto M. Potent inhibition of heterotopic ossification by nuclear retinoic acid receptor-gamma agonists. Nature medicine. 2011;17:454–460. doi: 10.1038/nm.2334. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Shore EM, Gannon FH, Kaplan FS. Fibrodysplasia ossificans progressiva: why do some people have two skeletons? Rev Rhum Engl Ed. 1997;64:92S–97S. [PubMed] [Google Scholar]
  24. Shore EM, Xu M, Feldman GJ, Fenstermacher DA, Cho TJ, Choi IH, Connor JM, Delai P, Glaser DL, LeMerrer M, Morhart R, Rogers JG, Smith R, Triffitt JT, Urtizberea JA, Zasloff M, Brown MA, Kaplan FS. A recurrent mutation in the BMP type I receptor ACVR1 causes inherited and sporadic fibrodysplasia ossificans progressiva. Nature genetics. 2006;38:525–527. doi: 10.1038/ng1783. [DOI] [PubMed] [Google Scholar]
  25. Thomas PS, Rajderkar S, Lane J, Mishina Y, Kaartinen V. AcvR1-mediated BMP signaling in second heart field is required for arterial pole development: implications for myocardial differentiation and regional identity. Developmental biology. 2014;390:191–207. doi: 10.1016/j.ydbio.2014.03.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Wu B, Zhang Z, Lui W, Chen X, Wang Y, Chamberlain AA, Moreno-Rodriguez RA, Markwald RR, O'Rourke BP, Sharp DJ, Zheng D, Lenz J, Baldwin HS, Chang CP, Zhou B. Endocardial cells form the coronary arteries by angiogenesis through myocardial-endocardial VEGF signaling. Cell. 2012;151:1083–1096. doi: 10.1016/j.cell.2012.10.023. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Yu PB, Deng DY, Lai CS, Hong CC, Cuny GD, Bouxsein ML, Hong DW, McManus PM, Katagiri T, Sachidanandan C, Kamiya N, Fukuda T, Mishina Y, Peterson RT, Bloch KD. BMP type I receptor inhibition reduces heterotopic [corrected] ossification. Nature medicine. 2008;14:1363–1369. doi: 10.1038/nm.1888. [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

Supplmental Figure 1

ca-ACVR1 expression potentiates HO formation 9 weeks after a dorsal burn injury and Achilles tenotomy. (A) MicroCT images of mice 9 weeks after a dorsal burn injury and Achilles tenotomy without tamoxifen-inducible ca-ACVR1 expression (Ub.Cre-ERT/ca-ACVRWT/WT) treated tamoxifen. (B) MicroCT images of mice 9 weeks after a dorsal burn injury and Achilles tenotomy with tamoxifen-inducible ca-ACVR1 expression (Ub.Cre-ERT/ca-ACVRflox/WT) which have been administered tamoxifen. TM= tamoxifen.

NIHMS667004-supplement.tif (1.9MB, tif)

RESOURCES