Gene targeting of the transcription factor Mohawk in rats causes heterotopic ossification of Achilles tendon via failed tenogenesis Hidetsugu Suzukia,b, Yoshiaki Itoa, Masahiro Shinoharaa,c, Satoshi Yamashitaa, Shizuko Ichinosed, Akio Kishidae, Takuya Oyaizub, Tomohiro Kayamaa, Ryo Nakamichia, Naoki Kodaa, Kazuyoshi Yagishitab,f, Martin K. Lotzg, Atsushi Okawab, and Hiroshi Asaharaa,g,h,1 a Department of Systems BioMedicine, Tokyo Medical and Dental University, Bunkyo-ku, Tokyo 113-8510, Japan; bDepartment of Orthopaedic Surgery, Tokyo Medical and Dental University, Bunkyo-ku, Tokyo 113-8510, Japan; cPrecursory Research for Embryonic Science and Technology, Japan Science and Technology Agency, Kawaguchi, Saitama 332-0012, Japan; dResearch Center for Medical and Dental Sciences, Tokyo Medical and Dental University, Bunkyo-ku, Tokyo 113-8510, Japan; eDepartment of Material-Based Medical Engineering, Institute of Biomaterials and Bioengineering, Tokyo Medical and Dental University, Chiyoda-ku, Tokyo 101-0062, Japan; fSports Science Organization, Tokyo Medical and Dental University, Bunkyo-ku, Tokyo 113-8510, Japan; gCore Research for the Evolutionary Science and Technology, Japan Agency for Medical Research and Development, Chiyoda-ku, Tokyo 100-0004, Japan; and hDepartment of Molecular and Experimental Medicine, The Scripps Research Institute, La Jolla, CA 92037
Edited by Dennis A. Carson, University of California at San Diego, La Jolla, CA, and approved May 25, 2016 (received for review November 8, 2015)
Cell-based or pharmacological approaches for promoting tendon repair are currently not available because the molecular mechanisms of tendon development and healing are not well understood. Although analysis of knockout mice provides many critical insights, small animals such as mice have some limitations. In particular, precise physiological examination for mechanical load and the ability to obtain a sufficient number of primary tendon cells for molecular biology studies are challenging using mice. Here, we generated Mohawk (Mkx)−/− rats by using CRISPR/Cas9, which showed not only systemic hypoplasia of tendons similar to Mkx−/− mice, but also earlier heterotopic ossification of the Achilles tendon compared with Mkx−/− mice. Analysis of tendon-derived cells (TDCs) revealed that Mkx deficiency accelerated chondrogenic and osteogenic differentiation, whereas Mkx overexpression suppressed chondrogenic, osteogenic, and adipogenic differentiation. Furthermore, mechanical stretch stimulation of Mkx−/− TDCs led to chondrogenic differentiation, whereas the same stimulation in Mkx+/+ TDCs led to formation of tenocytes. ChIP-seq of Mkx overexpressing TDCs revealed significant peaks in tenogenic-related genes, such as collagen type (Col)1a1 and Col3a1, and chondrogenic differentiation-related genes, such as SRY-box (Sox)5, Sox6, and Sox9. Our results demonstrate that Mkx has a dual role, including accelerating tendon differentiation and preventing chondrogenic/ osteogenic differentiation. This molecular network of Mkx provides a basis for tendon physiology and tissue engineering.
|
tendon development Mkx Achilles tendon ossification
(MSC) elevates tendon-related markers, and transplantation of these cells increases the diameter of collagen fibers in tendons (7, 8), suggesting the potential application of Mkx in cell therapy for tendon injury. Although the results from analysis of Mkx knockout mice have provided critical information about tendon development, the utility of mice as an animal model has some limitations. In particular, regenerative experiments for tendon repair with precise surgical interventions are challenging, and most reports of cell therapy for tendon repair used animals that were larger than mice (7, 9). For physiological experiments, such as treadmill exercise to test the effect of mechanical load on tendons/ligaments, rats are preferable for analyzing the exact responses to the stress because they are physiologically more similar to humans than mice (10). It is also difficult to obtain a sufficient number of primary tenocytes from mice for tendon/ligament research. To overcome these limitations, rats are frequently used in musculoskeletal research. However, technical challenges related to the isolation and culture of ES cells posed difficulties in generating genetically modified rats (10, 11). Recent developments in gene-editing technologies, such as zinc-finger nuclease (ZFN) (12), transcription activator-like effector nuclease (TALEN) (13), and clustered regularly interspaced short palindromic repeats/ CRISPR associated proteins (CRISPR/Cas9) facilitate the Significance
| knockout rat | CRISPR-Cas9 |
Molecular mechanisms of tendon development and homeostasis are not well understood. Generation and analysis of Mkx−/− rats revealed new functions of Mohawk (Mkx) in mediating cellular responses to mechanical stress. An Mkx-ChIP assay in rat tendon-derived cells with Mkx expression suggested that this factor may associate with both tendon- and cartilage-related genes to orchestrate tendon cell differentiation and maintenance. These findings advance our understanding of tendon physiology and pathology.
T
endons play a critical role in the musculoskeletal system by connecting muscle to bone to transmit mechanical loads and enable movement. Tendon injuries and damage are repaired slowly and incompletely because of poor intrinsic healing capacity, which in part results from tissue hypocellularity and hypovascularity (1). Even after surgical tendon repair, a standard treatment for tendon rupture, clinical outcomes are not satisfactory because of recurrent rupture or adhesions (2). To develop cell-based or pharmacological approaches for promoting tendon repair, the molecular mechanism of tendon development and regeneration must be determined; however, the key genome network for tendon differentiation and homeostasis has not been well characterized. We, along with other researchers, recently reported the tendon-specific expression and functions of the transcription factor Mohawk (Mkx), which regulates tendon-related gene expression (3, 4). Mkx knockout mice showed general tendon hypoplasia (5, 6), suggesting that Mkx plays an important role during tendon development. Moreover, overexpression of Mkx in mesenchymal stem cells
7840–7845 | PNAS | July 12, 2016 | vol. 113 | no. 28
Author contributions: H.S. and H.A. designed research; H.S., M.S., S.Y., S.I., and T.O. performed research; H.S., S.Y., S.I., A.K., T.O., and T.K. contributed new reagents/analytic tools; H.S., Y.I., M.S., S.Y., S.I., A.K., T.O., T.K., R.N., N.K., K.Y., M.K.L., A.O., and H.A. analyzed data; and H.S., T.K., M.K.L., and H.A. wrote the paper. The authors declare no conflict of interest. This article is a PNAS Direct Submission. Data deposition: The ChIP-seq data have been deposited in the DNA Database in Japan, www.ddbj.nig.ac.jp/ (accession no. DRA004354). 1
To whom correspondence should be addressed. Email:
[email protected].
This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. 1073/pnas.1522054113/-/DCSupplemental.
www.pnas.org/cgi/doi/10.1073/pnas.1522054113
generation of genetically modified rats in a one-step injection (14). The objective of this study was to investigate the function of Mkx in vitro and in vivo, including developmental phenotype and molecular targets, through the generation of Mkx knockout rats. Results Embryonic and Adult Rat Tendons Express Mkx. Mouse tendons express Mkx in the embryonic and adult stages (3, 5, 15). To confirm that the same is true in rats, the expression of Mkx was analyzed at several developmental stages using whole-mount in situ hybridization. Embryonic day (E) 11.5 embryos showed expression of Mkx mRNA in the dermomyotome dorsomedial lip, forelimb, and hind limb (Fig. S1A). In E15.5 embryos, the expression of Mkx was identical to the tendon tract (15). In the postnatal stage, quantitative RT-PCR (RT-qPCR) revealed that tendon showed significantly higher expression of Mkx than other tissues (Fig. S1B). Mkx Knockout Rats Show “Wavy Tails” and Systemic Hypoplasia of Tendons. Expression of Mkx correlating with tendon development
suggested that Mkx is a tenogenesis-related transcription factor not only in mice but also in rats. To evaluate the function of Mkx, we generated Mkx knockout rats using a CRISPR/Cas9 system. The mRNA of hCas9 and guide RNA (gRNA), which targets the second exon of the Mkx gene, were injected into 96 rat zygotes (Fig. 1A) (14). Twenty-six rats were obtained; direct sequencing of genomic DNA revealed that nine rats (34%) contained
B
ATG
Mkx Target
PAM
Hindlimb extensor tendon
AGAAGA GGCTCG
C
Mkx
+/+
Mkx
Forelimb extensor tendon
-/-
Tail tendon
HE
Relative expression
100um
60 40
***
Mkx-/16 ***
tensile strength (MPa)
100um
Mkx+/+ tensile strength (N)
Picro -sirius red
D
1mm
1mm
10 8 6 4 2 0
Heterotopic Ossification in Achilles Tendon of Mkx−/− Rats. Mkx−/−
Forelimb flexor tendon
↓ frameshift
E
Mkx-/-
Hindlimb flexor tendon
ACTGAGAAGA TACTCTTGGCTCTA GGCTCGC
14
Mkx+/+
12 8 4 0
20 0 ***
Mkx+/+ Mkx-/***
*** ***
*** ***
**
***
***
***
***
***
***
***
Mkx Scx Tnmd Col1a1 Fmod Dcn Tnxb Col2a1 Runx2 Alpl Osx Opn Ibsp Acan
Fig. 1. Generation of Mkx−/− rats. (A) Target site of gRNA and result of direct sequencing. (B) Tendons of Mkx+/+ or Mkx−/− rats (8-wk-old). (C) H&E staining (Upper) and Picrosirius red staining (Lower) of the patellar tendons (black arrowhead) in 2-wk-old Mkx+/+ or Mkx−/− rats. (D) Tensile strength of the patellar tendon in Mkx+/+ or Mkx−/− rats. Absolute value of tensile strength (Left) and tensile strength per unit area (Right). Error bars, SEM (n = 3). (E) RT-qPCR analysis in the patellar tendon of 3-wk-old Mkx+/+ or Mkx−/− rats. GAPDH was used as an internal control. Error bars, SEM (n = 3). **P < 0.01; ***P < 0.005.
Suzuki et al.
rats showed elevated osteogenic and chondrogenic markers in the patellar tendon, suggesting heterotopic ossification in the tendon. To check general ossification of Mkx−/− rats, microcomputed tomography (microCT) was performed, which revealed early heterotopic ossification in the Achilles tendon of Mkx−/− rats (Fig. S4A). Heterotopic ossification occurred in 20% of 3-wk-old Mkx−/− rats and in all 5-wk-old Mkx−/− rats (Fig. S4B). The crosssectional area of the ossifications gradually increased until the age of 15 weeks in Mkx−/− rats (Fig. S4C). No heterotopic ossification in other tendons or ligaments was observed in 15-wk-old Mkx−/− rats (Fig. S5). Histochemistry of the Achilles tendon revealed that postnatal day 9 (P0) Mkx−/− rats already had chondral lesions in their Achilles tendon (Fig. 2A). Endochondral ossification occurred from the middle of the chondral lesion in 3- to 4-wk-old animals. To compare rats and mice, Venus knockin homozygous mutant mice (5) were also analyzed. P0 and 4-wk-old Mkx−/− mice had no chondral lesion or ossification in the Achilles tendon (Fig. S4D). RT-qPCR on RNA from the Achilles tendon revealed that expression of chondrogenic genes increased in the 2-wk-old Mkx−/− rat Achilles tendon (Fig. 2B). The mRNA levels of osteogenicrelated genes increased in Mkx−/− rats from 2 to 4 wk of age. The expression of bone morphogenetic protein (BMP) pathwayrelated genes, such as BMPr1a, BMPr2, Smad1, and Smad5 was also elevated in Mkx−/− rats. Immunohistochemistry also showed expression of chondrogenic/osteogenic genes at the site of heterotopic ossification in Mkx−/− rats (Fig. S4E). To evaluate the influence of heterotopic ossification on the physiological function, gait analysis was performed, which revealed a significant decrease in the maximum ankle plantar flexion of Mkx−/− rats (Fig. S6 and Movies S1 and S2). PNAS | July 12, 2016 | vol. 113 | no. 28 | 7841
DEVELOPMENTAL BIOLOGY
A
mutations at the target site. Of these nine chimeric rats (F0), three were crossed with Wistar rats and germ-line transmission was confirmed. Three lines of F1 rats were obtained. Putative off-target sites (16) of CRISPR/Cas9 were evaluated by direct sequencing, which showed no mutations (Table S1). Two primer pairs were designed to analyze the mRNA expression of Mkx in the patellar tendon. RT-qPCR with primers that flanked the target site showed significantly decreased Mkx expression. When one primer was designed precisely to be on the target site, the Mkx expression was reduced in Mkx−/− rats, and these primer pairs were used for subsequent experiments (Fig. S1 C and D). Mkx−/− mice are known to have a “wavy tail” phenotype, which becomes more apparent when the mice are running (6), but Mkx−/− rats have a more profound wavy tail phenotype without running (Fig. S2). Mkx−/− rats manifest general hypoplasia of tendons, such as the flexor and extensor tendons of limbs, tail tendons, and patellar tendons while maintaining collagen orientation (Fig. 1 B and C). The tensile strength of the patellar tendons was decreased in the Mkx−/− rats even after normalization with their cross-sectional area (Fig. 1D). These wavy tail phenotypes and general hypoplasia of the tendon were confirmed in all three knockout lines (Fig. S2). One of these lines (14-deletion) was used for subsequent experiments. mRNA levels of tendon-related markers, such as tenomodulin (Tnmd), collagen type 1 α 1 (Col1a1), fibromodulin (Fmod), decorin (Dcn), and tenascin XB (Tnxb), were decreased in the patellar tendons of Mkx−/− rats (Fig. 1E). Scleraxis, known as a tenogenesis-related transcription factor (17), was also decreased in Mkx−/− rats but the difference was smaller compared with the other tendon markers. Osteogenesis- and chondrogenesis-related genes, such as Col2a1, Acan, Runx2, Alpl, and IBSP, were elevated in Mkx−/− rats. Transmission electron microscopy (TEM) showed that the collagen fibril diameter in the tail tendon of Mkx−/− rats was uniformly smaller than that in Mkx+/+ rats (Fig. S3). These phenotypes of Mkx−/− rats were similar to those of Mkx−/− mice (5, 6), but were severe and more readily observed because of their larger sizes.
Mkx-/-
Alizarin red Mkx+/+
Mkx-/-
P0
Mkx+/+
SafraninO -fast green Mkx+/+ Mkx-/-
B
Col2a1 20
***
16
1.6
12
1.2
8
0.8
4
n.s. 0.4
0 2wk 4wk
3-week
relative expression
200um
relative expression
4-week 500um
0
Sox9
Sox6
** n.s.
2wk 4wk
Runx2
Alpl 3 n.s. n.s.
7 6 5 4 3 2 1 0
***
Acan 3
**
2
n.s. 2wk 4wk
Opn p
1 0
n.s.
2wk 4wk
Ibsp
30 14 * * 12 25 10 20 8 15 *** 6 10 4 ** 5 ** 2 0 0 2wk 4wk 2wk 4wk 2wk 4wk
**
2 1 0
500um
2
relative expression
HE
relative expression
A
2wk 4wk
Bmpr2 Smad1 Smad5 Bmpr1a 1.8 n.s. 3.5 4 3 * ** n.s. 3 1.5 1.2 0.9 0.6 0.3 0
2.5 2.5 ** 2 n.s. 2 2 1.5 1.5 1 1 1 0.5 0.5 0 0 0 2wk 4wk 2wk 4wk 2wk 4wk 2wk 4wk 3
*
*
Fig. 2. Heterotopic ossification in the Achilles tendon of Mkx−/− rats. (A) H&E staining, Safranin O-Fast green staining, and Alizarin red staining of the Achilles tendons in P0, 3-wk-old, and 4-wk-old Mkx+/+ or Mkx−/− rats. (B) RT-qPCR analysis in the Achilles tendon of 2-wk-old and 4-wk-old Mkx+/+ or Mkx−/− rats. Gapdh was used as an internal control. Error bars, SEM (n = 3); ns, not significant. *P < 0.05; **P < 0.01; ***P < 0.005.
Mkx Deficiency Accelerates Chondrogenic and Osteogenic Differentiation of Tendon-Derived Cells. Endochondral ossification involves mes-
enchymal stem cell condensation and chondrocytic differentiation (18). Endochondral ossification in the Achilles tendon of Mkx−/− rats prompted us to test whether Mkx is critical for restricting mesenchymal cell differentiation into tenocytes and for preventing chondrocytic differentiation. It was previously reported that a stem cell population, known as tendon stem/progenitor cells (TSPCs), were enriched in tendon tissues (19). We isolated and cultured tendon-derived cells (TDCs) from the patellar tendons of 3-wk-old Mkx+/+ and Mkx−/− rats (Fig. 3A). There were no differences in the morphology of Mkx+/+ and Mkx−/− TDCs (Fig. S7A) and FACS analysis revealed that these cell preparations did not contain hematopoietic stem cells (Fig. S7B). These TDCs were capable of osteogenic, chondrogenic, and adipogenic differentiation, supporting the idea that the TDCs contained stem/progenitor populations. Under chondrogenic differentiation conditions, the pellets of Mkx−/− TDCs were larger than those of Mkx+/+ TDCs (Fig. 3 B and C). RT-qPCR revealed that the expression of chondrogenic markers was higher in Mkx−/− TDCs after 1 wk of chondrogenic differentiation (Fig. S7C). Under osteogenic differentiation, Alizarin red staining revealed greater ossification in Mkx−/− TDCs than in Mkx+/+ TDCs at 14 d (Fig. 3 D and E). The difference in ossification between Mkx+/+ and Mkx−/− decreased at 21 d, but remained significant. RT-qPCR showed higher expression of osteogenic genes in Mkx−/− TDCs (Fig. S7D) and there were no differences in adipogenic differentiation between Mkx+/+ and Mkx−/− TDCs (Fig. S7 E and F). These data suggest that Mkx deficiency leads to enhanced osteogenic and chondrogenic differentiation of TSPCs. Mkx Overexpression Suppressed Chondrogenic, Osteogenic, and Adipogenic Differentiation of Mkx−/− TDCs. To rescue the loss of
Mkx, Mkx−/− TDCs were retrovirally transduced with the Mkx coding sequence containing a FLAG tag. Retrovirus-encoding Venus protein was also used as a control. Induction of FLAGtagged Mkx was confirmed by RT-qPCR, Western blotting, and immunocytochemistry (Fig. S7 G–I). Mkx-transduced Mkx−/− TDCs showed lower expression of chondrogenic markers after chondrogenic differentiation (Fig. S7J). 7842 | www.pnas.org/cgi/doi/10.1073/pnas.1522054113
In the osteogenic differentiation condition, Mkx-transduced TDCs had fewer calcium deposits (Fig. S7 K and L) and reduced osteogenic markers (Fig. S7M). Mkx-transduced TDCs lost the ability to differentiate into adipocytes (Fig. S7 N and O). Mechanical Stretch Stimulation of Mkx−/− TDCs Leads to Chondrogenic Differentiation. Tendons respond to appropriate mechanical strains
by increasing collagen production in tenocytes (20) and mechanical strains promote MSC differentiation into tenocytes (21). As shown above, Mkx−/− TDCs showed a strong ability to differentiate into osteocytes and chondrocytes, suggesting that the loss of Mkx affects the response to mechanical stress of TDCs. To investigate this theory, TDCs were subjected to mechanical stretch stimulation (Fig. 4). After 4% monoaxial cyclic elongation for 6 h, Mkx+/+ TDCs showed elevated levels of tendon-related genes, such as Mkx, Col1a1, and Col3a1, indicating the tenogenic differentiation. However, the same mechanical stimulation of Mkx−/− tendon-derived cells increased chondrogenic markers, such as SRY-box (Sox)6, Sox9, and Acan, rather than tendon-related genes. Both Tendon-Related and Chondrogenic Differentiation-Related Genes Are Putative Targets of Mkx. Although Mkx appears to be a crit-
ical transcription factor for tendon development and homeostasis, a genome-wide approach to identify the direct targets of Mkx in tenocytes has not yet been reported, partly because of the difficulty of assembling a sufficient number of samples in mice. In the experiments described above (Fig. 3 and Fig. S7 G–O), we successfully rescued the Mkx−/− TDC phenotype by overexpression of Mkx. Because a ChIP-grade antibody for mice and rats is not yet available, we attempted to perform ChIP-sequencing using the Mkx−/− TDCs overexpressing tagged Mkx. The hemagglutinin (HA) tag-fused Mkx binding region in Mkx−/− TDCs was analyzed by ChIP-seq in the next-generation sequencer MiSeq (Fig. 5A). We obtained 6,356,463 sequence reads with the anti-HA antibody ChIP sample and 7,541,415 sequence reads from input DNA. The sequencing data were aligned with the rat genome (rn6) using bowtie software (22), resulting in 4,177,212 read maps of ChIP and 4,961,690 read maps of input samples. By using the mapped sequence reads, Mkx binding regions were detected by model-based analysis of ChIP-seq (MACS) (23) using the default parameters. The following analysis revealed 6,000 peaks of putative Mkx Suzuki et al.
Mkx+/+
Mkx-/-
E
10 8 6 4 2
*
Mkx+/+ Mkx-/-
**
Mkx+/+ Mkx-/-
***
0 14 2 1d d ay
s
ay
s
Fig. 3. Mkx regulates chondrogenic, osteogenic, and adipogenic differentiation in TDCs. (A) Protocol for isolation of TDCs. (B) Appearance of pellets of Mkx+/+ or Mkx−/− TDCs after chondrogenic differentiation (Alcian blue staining). (C) Diameter of the pellets of Mkx+/+ or Mkx−/− TDCs. Error bars, SEM (n = 3). (D) Appearance of the wells of Mkx+/+ or Mkx−/− TDCs after osteogenic differentiation (Alizarin red staining). (E) Relative absorbance at 450 nm of Alizarin red dye elution. Error bars, SEM (n = 3). *P < 0.05; **P < 0.01; ***P < 0.005.
binding sites. Within these peaks, protein-coding genes were selected and functional annotation was performed using DAVID v6.7 (https://david.ncifcrf.gov/home.jsp). Among the functional annotation chart, “skeletal system development (GO: 0001501)” was listed with a P value 7.6E-11, which included tendon-related genes, such as Col1a1, Col3a1, and Mkx. Interestingly, chondrogenic differentiation-related genes, such as Sox5, Sox6, and Sox9 were also included in the ontology (Fig. 5B). Moreover, other tendon-related genes, such as tenascin C (Tnc) and Fmod, were also included within these peaks. De novo motif discovery of these peaks using MEME-ChIP software (24) revealed three binding motifs, and two of them contained A-C-A, which is the putative binding site of mouse Mkx (25) (Fig. 5C). To support the physical interaction between Mkx and promoters, peaks in the promoter regions of Mkx, Col1a1, and Col3a1 were inserted into a thymidine kinase (TK) minimal promoter luciferase vector, and luciferase assays were performed after coexpression of Mkx tagged with the VP16 effector. As a result, higher luciferase activity was observed with VP16-Mkx expression compared with the control (Fig. S7P). This result indicates that these promoter regions may interact with Mkx either directly or indirectly. Discussion Although the rat is a preferable experimental animal compared with the mouse in several medical and biological research fields, including that of the musculoskeletal system, limited information is available regarding the use of knockout rats. This limitation is because of the technical difficulty in manipulating rat ES cells (26) and germ-line stem cells (27) for targeted genome deletion compared with that in the widely used mouse ES cells (11). In this regard, recent genome-editing technologies, such as ZFN, TALEN, and CRISPR/Cas9, are powerful strategies by which gene knockouts can be generated in various species of animals without using ES cells and homologous recombination (10). Here, we successfully generated genetically modified rats with deletion of Mkx with CRISPR/Cas9. The analysis of Mkx−/− rats not only confirms but also extends our knowledge on Mkx-dependent tendon differentiation and regulation, by allowing us to observe a more severe phenotype of heterotopic ossification in the Achilles tendon, to perform the physiological assessment of the ankle joint angle during ambulation, and to collect sufficient amounts of primary tenocytes from knockout rats for mechanistic analyses and ChIP-seq. Suzuki et al.
In human, heterotopic ossification is a substantial medical problem because it is associated with pain and dysfunction (28). In systemic heterotopic ossification, a fibrodysplasia ossificans progressiva-like phenotype (29) and ossification of the posterior longitudinal ligament of the spine (30) are important hereditary diseases, although these ossifications occur only several years after birth. BMP4 transgenic mice (31) and Npps−/− mice, referred to as tip-toe-walking mice (32), have been considered as animal models of these human diseases, with a fibrodysplasia ossificans progressiva-like phenotype and posterior longitudinal ligament of the spine, respectively. Biglycan (Bgn)- and Fmoddeficient mice showed heterotopic ossification in not only the Achilles tendon, but also around the knee joint (33). These Bgn/ Fmod-deficient mice also showed decreased diameter of collagen fibrils, similar to that observed in Mkx knockout mice and rats. Our data also indicate that Fmod expression in the patellar tendons of Mkx−/− rats was decreased. Furthermore, ChIP-seq of Mkx showed significant peaks around Fmod, which suggest an interaction between Fmod and Mkx. Although many patients with Achilles tendon heterotopic ossification have a history of trauma or surgery, hereditary factors are thought to be involved in the etiology of this disorder (34). Excessive stress on the Achilles tendon, as well as surgical intervention or injection of growth factors (35–37) have been shown to cause heterotopic ossification of the tendon. Here, we observed Achilles tendon ectopic ossification in neonatal Mkx−/− rats. This ectopic ossification was related to an endochondral ossification program; at birth, chondrogenesis occurs in the center of the Achilles tendon and then cartilage tissues are replaced by bone tissues. The precise molecular mechanisms of this heterotopic ossification of the tendon are not fully understood; however, our data support that idea that mesenchymal cells (i.e., TSPCs), which should differentiate into tenocytes, may lose their fate without Mkx and can differentiate into chondrocytes during embryogenesis. Several factors, such as Indian hedgehog, insulin-like growth factors, and BMPs, regulate the behavior of chondrocytes during endochondral ossification (38, 39). The upregulation of BMP pathway-related genes in the Achilles tendon of Mkx−/− rats suggests that Mkx regulates the BMP pathway. Mechanical stress affects tendon development before and after birth (40), and excessive mechanical stress can lead to ossification of the tendon (28, 41). It is also reported that mechanical strain promotes the differentiation of MSCs into tenocytes in vitro (21, 42). In this regard, we recently found that Gtf2ird1 translocates into the nucleus in response to mechanical strain and activates the Mkx promoter through chromatin regulation (43). Here, our mechanical stretch experiments with TDCs
A
B3
silicon chamber with collagen coating ↓
mono-axial mechanical stretch for 6 h ↓
RNA isolation
***
2 1
***
0
-
Col1a1 2 1
+
Sox6
*
3
0
1.8 1.2
1
0.6 -
+
stretch
-
+
stretch
2
0
***
3
stretch
incubate TDCs for 12 h ↓
Mkx
0
Sox9
*
-
+
stretch
Col3a1 4 3 2 1 0
***
-
Mkx+/+ Mkx-/-
+
stretch
Acan
2.0 1.6 1.2 0.8 0.4 0 -
**
+
stretch
Fig. 4. Mechanical stretch stimulation of Mkx−/− TDCs leads to chondrogenic differentiation. (A) Protocol of mechanical stretch stimulation. (B) Real-time PCR analysis in Mkx+/+ or Mkx−/− TDCs after mechanical stretch stimulation. GAPDH was used as an internal control. Error bars, SEM (n = 3). *P < 0.05; **P < 0.01; ***P < 0.005.
PNAS | July 12, 2016 | vol. 113 | no. 28 | 7843
DEVELOPMENTAL BIOLOGY
Analysis (P3-P5)
14 days
Incubation
D
21 days
Filtration
100um
(um) 500 450 400 350 300 250 200 150
relative expression
Digestion with Collagenase
C
relative expression
Mkx+/+ Mkx-/-
diameter of cartilage spheloid
B Dissection of Patella tendon (3-week-old)
relative absorbance
A
A
B Mkx-/- TDCs
9kb
2.5kb
Mkx 1kb 13kb
Retrovirus encoding HA-Mkx
16kb
Col1a1
HA -Mkx
1kb
350 bp
Col3a1
HA-Mkx induced TDCs
1kb
53kb 52 kb
Fmod 293 kb
Y
anti-HA antibody
52kb
HA
Mkx
ChIP
1kb
Sox5 10kb 157kb 186kb
10kb 51kb
14kb
Sox9 1kb
C
Mkx binding motif
1
1
1
(E-value 2.9e-013)
0
1 2 3 4 5 6 7 8
2
1 2 3 4 5 6 7 8
2
0
Generation of Mkx Knockout Rats and Genotyping. All animal experiments were approved by the Institutional Animal Care and Use Committee at the Tokyo Medical and Dental University. The gRNA and hCas9 mixture RNA was microinjected into the cytoplasm of Wistar rat zygotes by the UNITECH Corporation. The resulting chimeric offspring were crossed with Wistar rats and germ-line transmission was confirmed by sequencing.
bits
2
(E-value 8.7e-002)
0
Preparation of hCas9 and gRNA. The preparation of hCas9 and gRNA has been described previously (45). Briefly, the gRNA, the target of which was designed to the second exon of the Mkx gene, was constructed by inverse PCR. In vitro RNA synthesis and purification were performed.
1 2 3 4 5 6 7 8
bits
bits
Materials and Methods Detailed materials and methods were described in SI Materials and Methods. The list of primer sequences for RT-PCR is shown in Table S2.
433kb
Sox6 Next-Generation Sequencer analysis
Therefore, we show in Mkx knockout rats that Mkx plays a critical role in TSPC differentiation to tenocytes and development of tendon tissues. These findings indicate that Mkx can be applied as a therapeutic target for tendon repair or tissue engineering. In addition, the Mkx knockout rats represent a powerful animal model for further research on musculoskeletal tissues and diseases.
(E-value 8.2e-035)
Fig. 5. ChIP-seq of HA-Mkx–overexpressing TDCs revealed that Sox5, Sox6, and Sox9 are putative targets of Mkx. (A) Protocols of ChIP-seq in HA-Mkx– overexpressing TDCs. (B) ChIP-seq peaks of Mkx. Red bars indicate the peak. The distance from the transcription start site is shown above each peak. (C) De novo motif analysis of ChIP-seq.
support the idea that mechanical stimulation causes Mkx−/− TDCs but not wild-type cells to undergo chondrogenic differentiation. The reason why rats showed a more severe phenotype than mice with Achilles tendon ossification may be explained as follows. Rats are larger than mice, which may increase mechanical stimulation to the Achilles tendon during embryogenesis and more readily stimulate Mkx−/− TSPCs to undergo chondrogenic differentiation. In Mkx−/− rats, ectopic chondrogenesis in the Achilles tendon was observed in the embryonic and neonatal stages; however, increased chondrogenic marker gene expression was terminated by 4 wk, suggesting that additional chondrogenic changes may not occur in the Achilles tendon after birth. The discrepancy of cartilaginous change phenotypes between the embryonic stage and after birth may reflect the pluripotency of TSPCs in embryos and adults. In this regard, it is of great interest to examine whether there would be a difference in TSPCs from Mkx−/− rat embryos and adults with mechanical loading. In addition, whether excessive exercise promotes ectopic ossification in other tendons in mature Mkx−/− rats should be examined in the future. Overexpression of Mkx has been shown to promote tendonrelated gene expression and to repress gene expression characteristic of other cell lineages (7, 8, 25, 44). Consistent with these previous reports, our study also showed that osteogenic and chondrogenic differentiation occurs more readily in TDCs from an Mkx−/− background than in those from an Mkx+/+ background (Fig. 3 B–E). Regulation of Mkx expression was shown to affect expression of essential extracellular matrix genes of tendon tissues, such as Col1a1, decorin, and Tnc, and chondrogenesis master genes. These in vitro observations, as well as the in vivo ectopic ossification phenotype, indicate that the potential function of Mkx is to regulate cell fate of TSPCs via repressing chondrogenenic factors. Our ChIP-seq data revealed that Mkx interacts with both extracellular matrix genes and chondrogenic genes. Further detailed experiments are needed to clarify the precise molecular mechanisms of how Mkx coordinates expression of this diverse set of genes. 7844 | www.pnas.org/cgi/doi/10.1073/pnas.1522054113
Tensile Testing. Patellar tendons from 6-mo-old wild-type or Mkx−/− rats (n = 3) were pulled at a constant strain rate of 0.05 mm/s by a uniaxial materials testing system (5). Isolation of Rat TDCs. Patellar tendons of 3-wk-old wild-type or Mkx−/− rats were dissected. The samples were cut into small pieces and digested with collagenase (Sigma). After filtration with a nylon filter, digested cells were cultured. All experiments were performed until passage 5. Adipogenic/Osteogenic Differentiation. TDCs were plated into 24-well plates at 37 °C. After 24 h, the medium was changed to Adipogenesis Induction Medium (Lonza) and incubated for 7 d for adipogenic differentiation. The medium was changed to Osteogenesis Induction Medium (Lonza) and incubated for 14 d, and Alizarin red staining was performed for osteogenic differentiation. Chondrogenic Differentiation and Alcian Blue Staining. TDCs were suspended in Chondrogenic Incomplete Medium (Lonza). After centrifuging and changing the medium to Chondrogenic Complete Medium (Lonza), the pellets were incubated for 21 d and Alcian blue staining was performed. Retrovirus Infection. Venus, FLAG-Mkx, or HA-Mkx was inserted to the MIGR vector (Addgene). The plasmids were transfected into PLAT-E cells. After 24 h, the filtered supernatant was used to infect TDCs (P1) derived from an Mkx−/− rat. Mechanical Stretch Stimulation. Cells were seeded into the elastic silicon rubber chambers 12 h before stretching. The chambers were set on a monoaxial stretching device (STB-140, Strex) and monoaxial cyclic strain was applied for 6 h. ChIP. After fixation, the cells were washed with cell lysis buffer containing protease inhibitors, and resuspended in nuclear lysis buffer containing protease inhibitors. Chromatin was fragmented to 100–400 base pairs using sonication. The solution was then incubated with HA or normal rabbit IgG antibodies bound to beads. The immunoprecipitates were eluted from the beads, incubated to reverse the cross-linking, and purified for DNA analysis. ChIP-seq Library Preparation and Data Analysis. DNA libraries for next-generation sequencing were prepared using the TruSeq ChIP Sample Preparation kit (Illumina) from 3-ng ChIP DNA or input DNA and sequenced on a MiSeq (Illumina). ChIP DNA-enriched regions were detected by MACS v1.4.2 with default parameters. De novo motif discovery was performed with MEME-ChIP (24) using the default parameter. Luciferase Reporter Assays. HEK293FT cells were seeded in 96-well plates at 30% confluence and were transfected with pcDNA-VP16-Mkx or control pcDNA-Venus along with firefly and Renilla luciferase reporters. Thirty-six hours after transfection, luciferase activity was measured. The results were normalized to Renilla luciferase activity.
Suzuki et al.
ACKNOWLEDGMENTS. We thank Dr. Tomoki Chiba, Dr. Masaki Mori, Dr. Masafumi Inui, Dr. Masashi Naito, Dr. Yusuke Mochizuki, and all other laboratory members for the helpful discussions; Dr. Mari Uomizu for providing helpful advice for tensile testing; Dr. Takaaki Kubota and Dr. Zhang Yongwei for their help with tensile testing; and Dr. Mitsuhiro Enomoto for
providing helpful advice for gait analysis. This work was supported by the Core Research for the Evolutionary Science and Technology funding from the Japan Agency for Medical Research and Development; JSPS KAKENHI (Grants 26113008, 15H02560, and 15K15544); grants from the NIH (AR050631, AR065379, and AG007996); the Takeda science foundation; a Bristol-Myers K.K. RA Clinical Investigation grant (to H.A.); the Japan Aerospace Exploration Agency (Grant 14YPTK-005512); and “Creation of Life Innovation Materials for Interdisciplinary and International Researcher Development” project, Ministry of Education.
1. Sharma P, Maffulli N (2005) Tendon injury and tendinopathy: Healing and repair. J Bone Joint Surg Am 87(1):187–202. 2. Griffin M, Hindocha S, Jordan D, Saleh M, Khan W (2012) An overview of the management of flexor tendon injuries. Open Orthop J 6:28–35. 3. Yokoyama S, et al. (2009) A systems approach reveals that the myogenesis genome network is regulated by the transcriptional repressor RP58. Dev Cell 17(6):836–848. 4. Shimizu H, et al. (2013) The AERO system: A 3D-like approach for recording gene expression patterns in the whole mouse embryo. PLoS One 8(10):e75754. 5. Ito Y, et al. (2010) The Mohawk homeobox gene is a critical regulator of tendon differentiation. Proc Natl Acad Sci USA 107(23):10538–10542. 6. Liu W, et al. (2010) The atypical homeodomain transcription factor Mohawk controls tendon morphogenesis. Mol Cell Biol 30(20):4797–4807. 7. Otabe K, et al. (2015) Transcription factor Mohawk controls tenogenic differentiation of bone marrow mesenchymal stem cells in vitro and in vivo. J Orthop Res 33(1):1–8. 8. Liu H, et al. (2015) Mohawk promotes the tenogenesis of mesenchymal stem cells through activation of the TGFβ signaling pathway. Stem Cells 33(2):443–455. 9. Ho JOY, Sawadkar P, Mudera V (2014) A review on the use of cell therapy in the treatment of tendon disease and injuries. J Tissue Eng 5:1–18. 10. Shao Y, et al. (2014) CRISPR/Cas-mediated genome editing in the rat via direct injection of one-cell embryos. Nat Protoc 9(10):2493–2512. 11. Kawamata M, Ochiya T (2010) Generation of genetically modified rats from embryonic stem cells. Proc Natl Acad Sci USA 107(32):14223–14228. 12. Geurts AM, et al. (2009) Knockout rats via embryo microinjection of zinc-finger nucleases. Science 325(5939):433. 13. Sung YH, et al. (2013) Knockout mice created by TALEN-mediated gene targeting. Nat Biotechnol 31(1):23–24. 14. Mali P, et al. (2013) RNA-guided human genome engineering via Cas9. Science 339(6121):823–826. 15. Anderson DM, et al. (2006) Mohawk is a novel homeobox gene expressed in the developing mouse embryo. Dev Dyn 235(3):792–801. 16. Fu Y, et al. (2013) High-frequency off-target mutagenesis induced by CRISPR-Cas nucleases in human cells. Nat Biotechnol 31(9):822–826. 17. Murchison ND, et al. (2007) Regulation of tendon differentiation by scleraxis distinguishes force-transmitting tendons from muscle-anchoring tendons. Development 134(14):2697–2708. 18. Karsenty G, Wagner EF (2002) Reaching a genetic and molecular understanding of skeletal development. Dev Cell 2(4):389–406. 19. Bi Y, et al. (2007) Identification of tendon stem/progenitor cells and the role of the extracellular matrix in their niche. Nat Med 13(10):1219–1227. 20. Yang G, Crawford RC, Wang JH (2004) Proliferation and collagen production of human patellar tendon fibroblasts in response to cyclic uniaxial stretching in serum-free conditions. J Biomech 37(10):1543–1550. 21. Zhang L, Kahn CJ, Chen HQ, Tran N, Wang X (2008) Effect of uniaxial stretching on rat bone mesenchymal stem cell: Orientation and expressions of collagen types I and III and tenascin-C. Cell Biol Int 32(3):344–352. 22. Langmead B, Trapnell C, Pop M, Salzberg SL (2009) Ultrafast and memory-efficient alignment of short DNA sequences to the human genome. Genome Biol 10(3):R25. 23. Zhang Y, et al. (2008) Model-based analysis of ChIP-Seq (MACS). Genome Biol 9(9):R137. 24. Machanick P, Bailey TL (2011) MEME-ChIP: Motif analysis of large DNA datasets. Bioinformatics 27(12):1696–1697. 25. Anderson DM, et al. (2012) Characterization of the DNA-binding properties of the Mohawk homeobox transcription factor. J Biol Chem 287(42):35351–35359. 26. Kawaharada K, Kawamata M, Ochiya T (2015) Rat embryonic stem cells create new era in development of genetically manipulated rat models. World J Stem Cells 7(7):1054–1063. 27. Kanatsu-Shinohara M, et al. (2006) Production of knockout mice by random or targeted mutagenesis in spermatogonial stem cells. Proc Natl Acad Sci USA 103(21):8018–8023. 28. O’Brien EJO, Frank CB, Shrive NG, Hallgrímsson B, Hart DA (2012) Heterotopic mineralization (ossification or calcification) in tendinopathy or following surgical tendon trauma. Int J Exp Pathol 93(5):319–331.
29. Shore EM, et al. (2006) A recurrent mutation in the BMP type I receptor ACVR1 causes inherited and sporadic fibrodysplasia ossificans progressiva. Nat Genet 38(5): 525–527. 30. Nakajima M, et al.; Genetic Study Group of Investigation Committee on Ossification of the Spinal Ligaments (2014) A genome-wide association study identifies susceptibility loci for ossification of the posterior longitudinal ligament of the spine. Nat Genet 46(9):1012–1016. 31. Kan L, Hu M, Gomes WA, Kessler JA (2004) Transgenic mice overexpressing BMP4 develop a fibrodysplasia ossificans progressiva (FOP)-like phenotype. Am J Pathol 165(4):1107–1115. 32. Okawa A, et al. (1998) Mutation in Npps in a mouse model of ossification of the posterior longitudinal ligament of the spine. Nat Genet 19(3):271–273. 33. Ameye L, et al. (2002) Abnormal collagen fibrils in tendons of biglycan/fibromodulindeficient mice lead to gait impairment, ectopic ossification, and osteoarthritis. FASEB J 16(7):673–680. 34. Ghormley JW (1938) Ossification of the tendo Achillis. J Bone Joint Surg Am 20(1): 153–160. 35. Lui PP, Chan LS, Cheuk YC, Lee YW, Chan KM (2009) Expression of bone morphogenetic protein-2 in the chondrogenic and ossifying sites of calcific tendinopathy and traumatic tendon injury rat models. J Orthop Surg 4:27. 36. Lounev VY, et al. (2009) Identification of progenitor cells that contribute to heterotopic skeletogenesis. J Bone Joint Surg Am 91(3):652–663. 37. Le Nihouannen D, et al. (2005) Ectopic bone formation by microporous calcium phosphate ceramic particles in sheep muscles. Bone 36(6):1086–1093. 38. Mackie EJ, Ahmed YA, Tatarczuch L, Chen KS, Mirams M (2008) Endochondral ossification: How cartilage is converted into bone in the developing skeleton. Int J Biochem Cell Biol 40(1):46–62. 39. Yu YY, Lieu S, Lu C, Colnot C (2010) Bone morphogenetic protein 2 stimulates endochondral ossification by regulating periosteal cell fate during bone repair. Bone 47(1):65–73. 40. Marturano JE, Arena JD, Schiller ZA, Georgakoudi I, Kuo CK (2013) Characterization of mechanical and biochemical properties of developing embryonic tendon. Proc Natl Acad Sci USA 110(16):6370–6375. 41. Shi Y, et al. (2012) Uniaxial mechanical tension promoted osteogenic differentiation of rat tendon-derived stem cells (rTDSCs) via the Wnt5a-RhoA pathway. J Cell Biochem 113(10):3133–3142. 42. Butler DL, et al. (2009) Using functional tissue engineering and bioreactors to mechanically stimulate tissue-engineered constructs. Tissue Eng Part A 15(4): 741–749. 43. Kayama T, et al. (2016) Gtf2ird1-dependent Mohawk (Mkx) expression regulates mechanosensing properties of tendon. Mol Cell Biol 36(8):1297–1309. 44. Chuang HN, Hsiao KM, Chang HY, Wu CC, Pan H (2014) The homeobox transcription factor Irxl1 negatively regulates MyoD expression and myoblast differentiation. FEBS J 281(13):2990–3003. 45. Inui M, et al. (2014) Rapid generation of mouse models with defined point mutations by the CRISPR/Cas9 system. Sci Rep 4:5396. 46. Kawamoto T (2003) Use of a new adhesive film for the preparation of multi-purpose fresh-frozen sections from hard tissues, whole-animals, insects and plants. Arch Histol Cytol 66(2):123–143. 47. Iwata A, Fuchioka S, Hiraoka K, Masuhara M, Kami K (2010) Characteristics of locomotion, muscle strength, and muscle tissue in regenerating rat skeletal muscles. Muscle Nerve 41(5):694–701. 48. Kim K-H, Hwangbo G, Kim S-G (2015) The effect of weight-bearing exercise and nonweight-bearing exercise on gait in rats with sciatic nerve crush injury. J Phys Ther Sci 27(4):1177–1179. 49. Ichinose S, et al. (2010) Morphological differences during in vitro chondrogenesis of bone marrow-, synovium-MSCs, and chondrocytes. Lab Invest 90(2):210–221. 50. Morita S, Kojima T, Kitamura T (2000) Plat-E: An efficient and stable system for transient packaging of retroviruses. Gene Ther 7(12):1063–1066.
Suzuki et al.
PNAS | July 12, 2016 | vol. 113 | no. 28 | 7845
DEVELOPMENTAL BIOLOGY
Statistical Analysis. The two-tailed independent Student’s t test was used to calculate the P values.
