Regulation of cortical bone formation by SLIT and NTRK-like family member 1

Summary

Mutations in SLIT and NTRK-like family member 1 (SLITRK1) result in Tourette syndrome (TS). Patients with TS exhibit delayed bone maturation and increased fracture risk. To understand the role of SLITRK1 in bone homeostasis, we examined the skeletal phenotype of Slitrk1 null mice and investigated the mechanisms responsible for altered bone cell function. Slitrk1 null mice had thinner cortical bones due to decreased periosteal bone formation, while trabecular bone density remained unchanged. Slitrk1 was expressed within osteoblast-lineage cells. Consequently, deletion of Slitrk1 in osteoblasts impaired ex vivo differentiation capacity. Loss of SLITRK1 in osteoblast cells reduced levels of TAZ, a RUNX2 co-activator necessary for osteoblast differentiation. These findings provide evidence for a cell-autonomous role of SLITRK1 in periosteal osteoblasts regulating cortical bone homeostasis. Our data further demonstrate that periosteal-, endosteal-, and trabecular-bone homeostasis are controlled by different mechanisms and provide new insight into the skeletal manifestations of TS.

Graphical abstract

Highlights

• •Osteoblasts express Slitrk1, the gene related to Tourette syndrome pathogenesis
• •Slitrk1 null mouse exhibits thin cortical bone
• •Slitrk1-null osteoblasts showed reduced capacity for osteoblastogenesis
• •Slitrk1 helps stabilize the TAZ protein

Molecular physiology

Introduction

Tourette syndrome (TS) is a childhood-onset neuropsychiatric disorder that is characterized by multiple motor and vocal tics. TS is not a rare disease; its global prevalence is estimated at approximately 1%.^1,2 Patients with TS also exhibit mental disorders such as attention-deficit/hyperactivity syndrome,^3,4 obsessive-compulsive disorder,^5,6 and autism spectrum disorder.^3,7 In addition to mental disorders, TS is also linked with skeleton-related symptoms such as delayed bone maturation⁸ and occurrence of fractures.^9,10 Therefore, patients with TS may suffer from abnormal bone metabolism.

TS is associated with mutations in the SLITRK1. SLITRK1, a member of the SLITRK family, is a single-pass transmembrane protein with an extracellular SLIT-like leucine-rich repeat domain and an intracellular neurotrophin receptor-like region. SLITRK1 is primarily expressed in the nervous system and is involved in neuronal dendrite extension.¹¹ Base substitution mutations in the 3' untranslated region (UTR) of the SLITRK1 gene have been observed in patients with TS. This mutation causes downregulation of SLITRK1 expression by altering the binding affinity of miR-189 with the SLITRK1 UTR.¹² Slitrk1 global knockout mice exhibit neuropsychiatric symptoms such as anxiety and depression and may serve as an effective animal model of TS.¹³ The transcriptional coactivator with PDZ-binding motif (TAZ) is a 14-3-3 binding protein and regulator of Hippo signaling.¹⁴ Hippo signaling activation prevents TAZ translocation into the nucleus through phosphorylation and binding to 14-3-3. In the absence of Hippo signaling, TAZ translocates into the nucleus and activates the transcription of target genes. TAZ directly binds to runt-related transcription factor 2 (Runx2), a master regulator of osteoblast differentiation,¹⁵ and activates Runx2-dependent gene transcription.^16,17 Thus, TAZ regulates cell lineage commitment by promoting osteoblast differentiation and inhibiting adipocyte differentiation.¹⁸

In this study, we report the phenotypic characterization of a mouse model of TS. We show that Slitrk1 null mice exhibit thinner cortical bones due to impaired periosteal bone formation. Slitrk1 null osteoblasts have reduced differentiation capacity, whereas Slitrk1 null osteoclasts appear functionally normal. Slitrk1 regulates Taz protein stability and Runx2 transcriptional activity.

Results

SLITRK1 regulates cortical bone thickness

First, we compared the body size of Slitrk1 null mice with that of their wild-type (WT) littermates at 8 weeks of age. Both the height and weight of Slitrk1 null mice were lower than those of WT mice (Figures 1A and 1B). Furthermore, both femoral and tibial bones of Slitrk1 null mice were shorter than those of WT mice (Figures 1C and 1D). Histomorphometry revealed a significant difference in growth plate thickness between WT and Slitrk1 null mice at 3 weeks of age. (Figures 1E and 1F). The growth plate was significantly thinner in Slitrk1 null mice compared to WT mice at 3 weeks of age, then no difference was observed after 7 weeks of age (Figure S1A). However, tibia and femur lengths remained significantly shorter in Slitrk1 null mice at 7 and 12 weeks of age (Figure S1B). To further investigate the bone phenotype, we performed Micro-CT analysis. Micro-CT analysis of femur revealed that the cortical bone volume fraction did not differ between Slitrk1 null- and WT mice at 7 weeks of age (Figures 2A and 2B). Consistent with the reduction in total body size, Sitrk1 null mice also exhibited lower cortical bone volume and lower cortical tissue volume than WT mice. Interestingly, Slitrk1 null mice also showed low cortical bone thickness at 7 weeks of age (Figure 2B). This thinner cortical bone phenotype in Slitrk1 null mice could be attributed to reduction in periosteal bone perimeter rather than to endocortex (Figures 2A and 2B). Measurements of bone parameters over time showed that Slitrk1 null mice had lower cortical tissue volume at 3 and 7 weeks of age (Figure S2) but not at 12 weeks of age. Slitrk1 null mice also showed a reduction in the polar moment of inertia (pMOI) (Figure S3); three-point bending mechanical testing did not reveal a statistically significant difference in mechanical endpoints (Figure S3). Dynamic histomorphometric analysis of femurs further confirmed the skeletal phenotype of Slitrk1 null mice. Bone formation was quantified by intraperitoneal administration of calcein to mice. Periosteal mineral apposition rate (Ps. MAR) and bone formation rate per bone surface (Ps. BFR/BS) were significantly lower in Slitrk1 null mice than in WT mice, whereas endocortical (Ec) parameters such as Ec. MAR and Ec. BFR/BS were unchanged (Figures 2C and 2D). Additionally, trabecular bone parameters did not differ significantly between the Slitrk1 null and WT mice (Figures 2E, 2F, and 2G). These results suggest that the thinning of the cortical bone in the absence of SLITRK1 could be attributed to reduced bone formation at the periosteum.

Figure 1.

Slitrk1 deletion causes reduced body size and mass

(A) Representative full-body photographs of 8-week-old wild-type (WT) and Slitrk1 null (Null) mice. Arrows indicate the way of body length measurement. Scale bars represent 10 mm.

(B) Total body length and body weight of 8-week-old WT and Slitrk1 null mice. WT, n = 7 mice; Null, n = 5 mice.

(C) Representative soft X-ray images of femoral and tibial bones of 7-week-old WT and Null mice. Scale bars represent 2 mm.

(D) Measured bone lengths of WT and Slitrk1 null mice. WT, n = 6 mice; Null, n = 5 mice.

(E) von-Kossa-stained proximal growth plate of 3-week-old WT and Slitrk1 null mice tibia. Arrows indicate the way of growth plate length measurement. Scale bars represent 100 μm.

(F) Growth plate thickness measured from images shown in (E). n = 4/group.

∗p < 0.05, ∗∗p < 0.01.

Figure 2.

Slitrk1 null mice have reduced cortical bone thickness

(A) Representative μCT images of distal (top) and midshaft (bottom) of femur from 7-week-old WT and Slitrk1 null mice. White and green scale bars represent 500 and 200 μm, respectively.

(B) Cortical bone volume per tissue volume (Ct. BV/TV), cortical tissue volume (Ct. TV), cortical bone volume (Ct. BV), cortical bone thickness (Ct. Th), periosteal bone perimeter (Ps. B. Pm), and endocortical bone perimeter (Ec. B. Pm) were calculated using μCT. n = 7 or 5.

(C) Representative photographs of calcein-stained periosteal (Ps) and endocortical (Ec) bone of 7-week-old WT and Slitrk1 null mice. CB, cortical bone. Scale bars represent 25 μm.

(D) Periosteal mineral apposition rate (Ps. MAR), periosteal bone formation rate per bone surface (Ps. BFR/BS), endocortical mineral apposition rate (Ec. MAR), and endocortical bone formation rate per bone surface (Ec. BFR/BS) were determined through bone histomorphometry analysis.

(E) Representative photographs of von Kossa-stained proximal tibiae of 7-week-old WT and Slitrk1 null mice. Scale bars represent 500 μm.

(F) The proximal tibiae of 7-week-old WT and Slitrk1 null mice were analyzed. Bone volume per tissue volume (BV/TV), trabecular bone thickness (Tb. Th), trabecular space (Tb. Sp), trabecular number (Tb. N), and connectivity density (Conn. Dn) were determined using μCT. WT, n = 7 mice; Null, n = 5 mice.

(G) Mineral apposition rate (MAR), bone formation rate per bone surface (BFR/BS), osteoblast surface per bone surface (Ob. S/BS), osteoblast number per bone surface (Ob. N/BS), osteoclast surface per bone surface (Oc. S/BS), and osteoclast number per bone surface (Oc. N/BS) were determined through bone histomorphometry analysis. WT, n = 7 mice; Null, n = 5 mice.

∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, in two-tailed unpaired t tests.

SLITRK1 expressed within osteoblast-lineage cells regulates bone formation

Next, we determined Slitrk1 expression across various mouse organs. Brain tissue showed high expression of Slitrk1 (Figure 3A), consistent with the fact that Slitrk1 is required for essential functions of the CNS.¹¹ High Slitrk1 expression was also observed in bone tissue (Figure 3A). Within bone tissue, Slitrk1 was predominantly expressed in osteoblast lineage cells such as osteoblasts and osteocytes (Figure 3B), whereas osteoclasts showed almost no Slitrk1 expression (Figure 3B). We confirmed the expression of Slitrk1 in bone tissue by in situ hybridization. Slitrk1 was expressed in osteoblasts in the periosteum, endosteum, and trabecular bone (Figure 3C).

Figure 3.

Slitrk1 expression is elevated during osteoblastogenesis

(A) Slitrk1 mRNA levels in different murine organs and tissues including bone marrow (BM), brain, femur, tibialis anterior muscle (TA), tongue, brown adipocyte tissue (BAT), epididymal white adipocyte tissue (eWAT), and spleen were determined using qPCR.

(B) Bone marrow stromal cells (BMSCs), chondrocytes (Chos), osteoblasts (OBs), osteocytes (OCYs), and osteoclasts (OCLs) were cultured from murine organs and tissues. Slitrk1 mRNA expression in cells were determined using qPCR.

(C) Representative photographs of in situ hybridization of tibial bone from 3-week-old WT mice to detect Slitrk1 transcripts. Arrows indicate Slitrk1-positive cells. CB, cortical bone; BM, bone marrow; TB, trabecular bone. Scale bars represent 50 μm.

(D and E) Slitrk1 mRNA (D) and protein levels (E) were determined using qPCR and western blotting analysis, respectively, 0, 7, or 14 days after stimulation for osteoblast differentiation in murine primary cultured calvarial osteoblasts. The numbers below the bands indicate the density of each band, normalized to β-actin and compared to the value at day 0, which is set to 1. Mean ± SD; n = 3.

∗∗p < 0.01, ∗∗∗p < 0.001.

When primary calvaria cells were cultured and differentiated in vitro, Slitrk1 mRNA expression and protein levels increased with osteoblast differentiation (Figures 3D and 3E). Consistent with a cell-autonomous role for Slitrk1 within osteoblast cells, we found that osteoblast differentiation genes and alkaline phosphatase (ALP) activity were reduced in Slitrk1-null-mice-derived primary osteoblasts (POBs). Particularly, Osterix (Osx), Alp, and Osteocalcin (Ocn) mRNA levels, which are the downstream effectors of Runx2 during osteoblastogenesis, were significantly lower in Slitrk1 null cultures compared to WT controls (Figures 4A and 4B). Changes to osteoblast differentiation capacity following Slitrk1 deletion were not due to changes in cell proliferation since no differences in cell number or cell-cycle-related gene expression (Figure S4) were observed. In addition, no difference was observed between osteoclastogenesis in Slitrk1 null and WT mice (Figure S5). To confirm the in vivo role of SLITRK1 on osteogenesis, we evaluated SLITRK1 function during BMP2-induced ectopic bone formation.¹⁹ The implantation of BMP2-containing collagen pellets resulted in formation of ectopic bones with a smaller bone volume in Slitrk1 null mice than in wild-type mice (Figures 4C and 4D).

Figure 4.

Slitrk1 regulates osteogenesis

(A) Calvarial osteoblasts isolated from WT or Slitrk1 null mice were cultured in osteogenic medium. mRNA levels of Runx2, Osx, Alp, and Ocn were determined on days 0, 7, or 14 using qPCR.

(B) Cells were treated with 0 or 25 ng/mL of rhBMP4. Alp activity was determined on day 3.

(C and D) Eight-week-old WT and Slitrk1 null mice were implanted with collagen pellets containing 1 μg BMP2. The ectopic bones were harvested 4 weeks after implantation. Representative soft X-ray and H&E staining of ectopic bones. White and red scale bars represent 2 mm and 100 μm, respectively (C). Ectopic bone volume was quantified in WT and Slitrk1 null mice (D). (A, B, and D) Mean ± SD; n = 3.

∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001.

SLITRK1 regulates RUNX2 function by suppressing TAZ degradation

Deletion of Slitrk1 significantly decreased the expression of the downstream effectors of Runx2 during osteoblast differentiation (Figure 4A). TAZ is a known transcriptional coactivator of RUNX2¹⁸ that may interact with SLITRK1 via 14-3-3 proteins since both TAZ and SLITRK1 associate with 14-3-3.^20,21 Therefore, we hypothesized that SLITRK1 deficiency disrupts TAZ function, leading to impaired RUNX2 activity. To test this hypothesis, we investigated the effects of SLITRK1 on TAZ and RUNX2 expression. Taz mRNA levels increased with osteoblast differentiation in both Slitrk1 null and WT POBs. Taz mRNA levels in WT and Slitrk1 null POBs were not significantly different (Figure 5A). In contrast, TAZ protein levels were lower in Slitrk1 null POB than that in WT POB (Figure 5B). Notably, the overexpression of wild-type Slitrk1 (Slitrk1-WT) increased Taz levels. However, the expression of mutant Slitrk1 S695A (Slitrk1-A), which could not interact with 14-3-3, did not increase the Taz levels. Treatment with the proteasome inhibitor MG132 prevented Taz levels from decreasing after Slitrk1 mutation (Figure 5C). These data suggest that SLITRK1 suppresses TAZ degradation through the proteasomal pathway. We then examined the effect of Slitrk1 on Runx2-induced osteoblastogenesis using C3H10T1/2 cells. Runx2 overexpression increases Alp and Ocn mRNA levels,²² which were significantly enhanced by Slitrk1 overexpression (Figure 5D). This suggests that SLITRK1 supports RUNX2 function and osteoblast differentiation by disrupting the TAZ–14-3-3 interaction, which stabilizes TAZ (Figure 5E).

Figure 5.

Slitrk1 regulates Taz protein stability

(A and B) Calvarial osteoblasts isolated from WT or Slitrk1 null mice were cultured in osteogenic differentiation medium. mRNA (A) and protein (B) levels of TAZ were determined on days 0, 7, and 14 using qPCR and western blotting analysis, respectively. The numbers below the bands indicate the density of each band, normalized to β-actin and compared to the value at day 7 WT, which is set to 1.

(C) Cos 7 cells were co-transfected with Myc-tagged wild-type Slitrk1 (Slitrk1-WT), Myc-tagged mutated Slitrk1 (Slitrk1-A), or DDDDK-tagged Taz as indicated and then treated with proteasome inhibitor MG132. Protein levels were determined by western blotting analysis on day 2. The numbers below the bands indicate the density of each band, normalized to β-actin and compared to the value at only Taz transfected (Flag) or Taz and Slitrk1-WT transfected (Myc), with the value set to 1.

(D) C3H10T1/2 cells were transfected with Slitrk1-WT or Runx2, and then Alp and Ocn mRNA levels were determined on day 2.

(E) Model of SLITRK1’s role in osteoblastogenesis. Mean ± SD; n = 3.

∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001.

Discussion

SLITRK1 has been shown to be involved in neuronal dendrite extension within the CNS. In this study, we show that in addition to its previously known roles, SLITRK1 is highly expressed in osteoblast lineage cells and regulates osteoblast differentiation.¹¹ SLITRK5 is another member of the SLITRK family expressed in both osteoblast lineage cells and neural cells.^23,24 In contrast to SLITRK1, SLITRK5 suppresses osteoblastogenesis by regulating Hedgehog signaling.²³ Thus, SLITRK1 and SLITRK5 exert opposing effects on osteoblast differentiation. Notably, SLITRK1 and SLITRK5 exert opposing functions in nerves. SLITRK1 promotes neurite elongation,¹¹ whereas SLITRK5 inhibits it.²⁴ The major difference between SLITRK1 and other members of the SLITRK family is the presence of a TRK-like structure in the intramembrane region. Therefore, future comparative analyses are warranted between SLITRK1 and SLITRK5. These studies may enhance our understanding of the biological significance of the SLITRK family.

Peripheral nerves and related factors such as substance P,²⁵ neurokinin B,²⁶ hemokinin-1,²⁷ and Sema3A^27,28 regulate bone metabolism. Additionally, the autonomic nervous system regulates bone metabolism, whereas the sympathetic nervous system increases bone resorption and suppresses osteoblast proliferation.^29,30 Slitrk1 null mice exhibited increased levels of norepinephrine and its metabolite 3-methoxy-4-hydroxyphenylglycol.¹³ This could possibly be attributed to increased sympathetic nervous system activity in response to increased norepinephrine levels, which causes cortical bone thinning in Slitrk1 null mice. Further studies are warranted using conditional osteoblast- or neuronal-lineage specific knockout mice to test this hypothesis.

In this study, we conclude that the reduced thickness and mass of the cortical bones in Slitrk1 null mice was due to a reduction in bone periosteal formation. However, the trabecular bones did not show significant difference between Slitrk1 null and wild-type littermates. Regulation of bone metabolism varies depending on the bone type or area (such as cortical or trabecular). For example, leptin-null mice showed lower femoral bone mineral density (BMD), thinner cortical bones, and lesser mineralization compared with that of the WT mice. However, they showed higher spinal length, lumbar BMD, and trabecular bone volume.³¹ Leptin receptors are expressed on osteoblasts and are associated with increased bone mass, osteoblast number, and osteoblast activity³²; however, they additionally control bone metabolism via the sympathetic nerves.³³ Thus, leptin activity is mediated both by the CNS and through an independent bone metabolism control system in the periphery, which may lead to complex phenotypes. This indicates that SLITRK1 may similarly control bone metabolism in multiple ways. In the current study, we showed that the total body length and bone length were shorter in Slitrk1 null mice, suggesting SLITRK1 function in growth plate cartilage. In fact, Slitrk1 was expressed in chondrocytes (Figure 3B) in the proliferating zone of growth plate cartilage. Growth plate thickness was significantly thinner in Slitrk1 null mice at 3 weeks old (Figures 1E and 1F), which could subsequently lead to shortening of long bones. We also obtained data that compared to WT mice, Slitrk1 null mice have altered adipose tissue phenotypes (Figure S6). This also suggests that alteration in the metabolism of adipose tissue may affect bone metabolism in Slitrk1 null animals, because adipose tissue is deeply associated with bone.^31,32,33 In addition to SLITRK1, the activator of transcription and developmental regulator AUTS2 (AUTS2) has been implicated in several neurological diseases including TS. Notably, Nestin-expressing cell-lineage-specific Auts2 knockout mice showed severe shortening of body length.³⁴ The reason for the shortening of the body length in response to Auts2 deletion remains unknown. Therefore, a combined analysis of SLITRK1 and AUTS2 may provide deeper insights into the regulatory mechanisms of skeletal development.

In the current study, we hypothesized that SLITRK1 stabilized TAZ by regulating its association with 14-3-3 proteins (Figure 5E). TAZ protein stabilization is observed in the regulation of Canonical Wnt signaling, which is critical for osteoblastogenesis.³⁵ In the absence of canonical Wnt signaling, the phosphorylation of β-catenin causes the degradation of TAZ-bound β-catenin. In contrast, during functional Wnt signaling, β-catenin and TAZ are stabilized, and TAZ translocates to the nucleus to promote transcriptional activity. The stabilized β-catenin and TAZ induces osteoblast differentiation.^16,36 Lymphoid enhancer-binding factor 1 (LEF1) is a transcription factor that plays a fundamental role in canonical Wnt signaling. LEF1 physically binds to TAZ and RUNX2 to induce their transcriptional functions.³⁷ In other words, stabilization of TAZ leads to RUNX2 activation via the canonical Wnt signaling pathway. In contrast, non-canonical Wnt signaling suppresses Hippo signaling and promotes osteoblast differentiation by activating TAZ in a β-catenin-independent manner.³⁸ Therefore, the activation mechanism and role of TAZ in Wnt signaling and osteoblast differentiation remain unclear and require further investigation.

The Slitrk1 null mice showed a reduction in pMOI, but the mechanical endpoints based on the 3-point bending test did not result in a statistically significant difference (Figure S3). Therefore, the results of these mechanical tests do not strongly support the fracture susceptibility of bones in TS.³⁹ Perhaps the violent and atypical movements often involved in TS possibly subject the bones to special forces. Therefore, a mechanical testing system needs to be established to consider the characteristic movements of TS and clarify the relationship between the SLITRK1 function and fracture susceptibility in TS. The cortical bone phenotype in Slitrk1 null mice was more pronounced at a younger stage, and this phenotype was reduced with age (Figure S2). This is consistent with the bone features of TS since delays in bone formation and maturation have been reported as a feature of TS.⁸ In addition, the symptoms of TS are more pronounced in younger people.⁶ However, we must note that our model does not completely mimic the TS phenotype, and there are genetic differences between TS and Slitrk1 null. To understand the mechanisms of TS, it will be necessary to establish more accurate models that mimic the genetic background and pathology of TS and perform further experiments.

Conclusion

SLITRK1, which is one of the genes associated with TS pathogenesis, controls osteoblastogenesis by stabilizing TAZ. The data presented in this study may at least partly explain the skeletal phenotypes of patients with TS.

Limitations of this study

The potential limitations of this study should be considered when interpreting the results. First, we used conventional Slitrk1 knockout mice. Therefore, we could neither specifically determine the function of SLITRK1 in osteoblast lineage cells nor exclude the influence of SLITRK1 expressed in other cells during in vivo experiments. Second, Slitrk1 null mice exhibited shorter body and bone lengths, which suggest that SLITRK1 also plays a role in cartilage growth. Similar to the formation of long tubular bones through endochondral ossification, cartilage chondrogenesis may contribute to the bone phenotype of Slitrk1 null mice. Third, seven subtypes have been reported for the 14-3-3 protein; however, this study did not ascertain the 14-3-3 subtype pertaining to the results reported in this paper. Further investigations are required to determine the subtypes and functions of 14-3-3 proteins during osteoblast differentiation. Fourth, Slitrk1 expression alone does not explain why the phenotype is only observed in cortical bone. Slitrk1 was expressed in osteoblasts within both cortical and trabecular bone. Fifth, we could not clarify why the bone phenotype is more pronounced in young Slitrk1 null mice. Sixth, our model does not truly mimic the TS phenotype possibly due to genetic differences between TS patients and Slitrk1 null mice. Seventh, these results were obtained from male mice only; therefore, sex differences were not examined.

Resource availability

Lead contact

Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact, Shoichiro Kokabu (r14kokabu@fa.kyu-dent.ac.jp).

Materials availability

This study did not generate new unique reagents.

Data and code availability

• •Data: this paper does not report new dataset.
• •Code: this paper does not report original code.
• •All other items: any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request.

Acknowledgments

This work was supported by grants from the Japan Society for the Promotion of Science (21K21072, 22K17257, and 24K20063 to T.Shirakawa.; 17H04409 to T.Satou.; 18K09861 to K.N.K.; and 22KK0141 to S.K.). This study was supported by the Joint Usage/Research Center for Developmental Medicine at Kumamoto University. We would like to thank Editage (www.editage.jp) for English language editing.

Author contributions

T.Shirakawa, T.Satou, and S.K. designed the study. T.Shirakawa, T.Satou, T.M., and S.K. performed the in vitro experiments and analyzed the data. T.Shirakawa, T.Satou, K.N., T.I., M.I., H.K., C.N., and M.N. performed the in vivo experiments and analyzed the data. Q.Y. and W.N.A. helped with data interpretation. T.Shirakawa and S.K. wrote the first draft of the manuscript. T.Shirakawa, T.Satou, K.N., T.I., M.I., H.K., C.N., M.N., Q.Y., W.N.A., T.M., M.H., J.A., K.N.K., T.K., and S.K. edited the manuscript. All authors helped with literature review.

Declaration of interests

The authors declare no competing interests.

STAR★Methods

Key resources table

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Antibodies
anti-Slitrk1	Novas Biological	NBP1-49541
anti-TAZ	Cell Signaling Technology	#28287
Anti-CyclinD1 antibody	Santa Cruz	Cat#sc-450; RRID:AB_627342
Anti-Ki67 antibody	Abcam	Cat#ab15580; RRID:AB_443209
anti-Myc	MEDICAL & BIOLOGICAL LABORATORIES CO., LTD.	Cat#M192-3; RRID:AB_11160947
Anti-DDDDK-tag	MEDICAL & BIOLOGICAL LABORATORIES CO., LTD.	Cat#M185-3L; RRID:AB_11123930
Anti-β-actin	MEDICAL & BIOLOGICAL LABORATORIES CO., LTD.	#M177-3; RRID:AB_10697039
anti-rabbit HRP	Jackson ImmunoResearch Laboratories	Cat#111-035-045; RRID:AB_2337938
anti-mouse HRP	Jackson ImmunoResearch Laboratories	Cat#115-035-062; RRID:AB_2338504
Alexa Fluor 488-conjugated anti-rabbit IgG	Thermo Fisher Scientific	Cat#A-32731; RRID:AB_2633280
Alexa Fluor 594-conjugated anti-mouse IgG	Thermo Fisher Scientific	Cat#A-32742; RRID:AB_2762825
Bacterial and virus strains
5-alpha Competent E. coli	New England Biolabs	Cat#C2987H
Chemicals, peptides, and recombinant proteins
Ascorbic acid	Sigma aldrich	Cat#A5960
β-glycerophosphate	Sigma aldrich	Cat#G5422
CollaCote	Zimmer Dental	Cat#0101
Collagenase	Fujifilm Wako	Cat#034-22363
DISPASE II	Godo Shusei Co., Ltd.	Cat#383-02281
D-MEM (High Glucose) with L-Glutamine and Phenol Red	Fujifilm Wako	Cat#044-29765
Immobilon ECL Ultra Western HRP Substrate	Merck	Cat#WBULS0100
Lipofectamine 2000 Reagent	Thermo Fisher Scientific	Cat#11668-019
MEMα with L-Glutamine and Phenol Red	Fujifilm Wako	Cat#135-15175
Methyl-methacrylate and 2-hydroxyethyl methacrylate mixture	Fujifilm wako	Cat#086-04385
MG132	Selleck	Cat#S2619
Penicillin-Streptomycin Solution	Fujifilm Wako	Cat#168-23191
PhosSTOPTM	Merck	Cat#4906845001
Pierce™ BCA Protein Assay Kits	Thermo Fisher Scientific	Cat# 23225
Power Up SYBRTM Green Master Mix	Thermo Fisher Scientific	Cat#A25742
Protease Inhibitor Cocktail SetⅠ	Fujifilm Wako	Cat#165-26021
Recombinant Human BMP-2	R&D Systems	Cat#355-BEC
Recombinant Human BMP-4	R&D Systems	Cat#314-BP-010
TRIzolTM Reagent	Thermo Fisher Scientific	Cat#15596026
4′,6-diamidino-2-phenylindole, dihydrochloride (DAPI)	Dojindo	Cat#D212
Critical commercial assays
Cell Counting Kit-8 (CCK8)	Dojindo	Cat#CK04
Fast GeneTM RNA Basic Kit	Nippon Gene	Cat#FG-80050
High-Capacity cDNA Reverse Transcription Kit	Thermo Fisher Scientific	Cat#4368814
PrimeSTAR® HS DNA Polymerase	Takara Bio	Cat#R010A
Experimental models: Cell lines
Cos7 cells	JCRB	Cat# JCRB9127; RRID:CVCL_0224
C3H10T1/2 cells	JCRB	Cat# JCRB9080; RRID:CVCL_0190
Experimental models: Organisms/strains
Slitrk1 conventional knockout/C57BL/6J	Katayama et al. 13	https://doi.org/10.1038/mp.2008.97
Oligonucleotides
Primer: Slitrk1 Forward: TTGGACCTCAGGTGGCTCTACA	This paper	N/A
Primer: Slitrk1 Reverse: GGAGAATGAGCTGAATCGCGTTG	This paper	N/A
Primer: Taz Forward: TTGACGCATCCTAATCCT	This paper	N/A
Primer: Taz Reverse: CTCGTCACAAGCAGGGTTAAG	This paper	N/A
Primer: Runx2 Forward: TTCAACGATCTGAGATTTGTGGG	Rojasawasthien et al.40	https://doi.org/10.1096/fba.2022-00093
Primer: Runx2 Reverse: GGATGAGGAATGCGCCCTA	Rojasawasthien et al.40	https://doi.org/10.1096/fba.2022-00093
Primer: Sp7 Forward: AGAGATCTGAGCTGGGTAGAGG	Rojasawasthien et al.40	https://doi.org/10.1096/fba.2022-00093
Primer: Sp7 Reverse: AAGAGAGCCTGGCAAGAGG	Rojasawasthien et al.40	https://doi.org/10.1096/fba.2022-00093
Primer: Alp Forward: CGGGACTGGTACTCGGATAA	Rojasawasthien et al.40	https://doi.org/10.1096/fba.2022-00093
Primer: Alp Reverse: ATTCCACGTCGGTTCTGTTC	Rojasawasthien et al.40	https://doi.org/10.1096/fba.2022-00093
Primer: Bglap Forward: AGACTCCGGCGCTACCTT	Rojasawasthien et al.40	https://doi.org/10.1096/fba.2022-00093
Primer: Bglap Reverse: CTCGTCACAAGCAGGGTTAAG	Rojasawasthien et al.40	https://doi.org/10.1096/fba.2022-00093
Primer: β-actin Forward: AAGGCCAACCGTGAAAAGAT	Inoue et al.41	https://doi.org/10.1002/2211-5463.12918
Primer: β-actin Reverse: GTGGTACGACCAGAGGCATAC	Inoue et al.41	https://doi.org/10.1002/2211-5463.12918
Recombinant DNA
pcDNA3.1	Thermo Fisher Scientific	Cat#V79020
Software and algorithms
Olympus cellSense software	Olympus	https://evidentscientific.com/en/products/software/cellsens
Histometry RT camera software	System Supply	https://www.system-supply.co.jp/wordpress/wp-content/uploads/2025/05/%E9%AA%A8%E5%BD%A2%E6%85%8B%E8%A8%88%E6%B8%AC%E3%82%B7%E3%82%B9%E3%83%86%E3%83%A0-%E3%82%AB%E3%82%BF%E3%83%AD%E3%82%B0.pdf
Other
Slitrk1 hybridized digoxigenin-labeled riboprobes	Aruga et al.11	https://doi.org/10.1016/s1044-7431(03)00129-5
MycoAlertTM Mycoplasma Detection Kit	Lonza	Cat#LT07-118

Experimental model and study participant details

Mouse experiments and isolation of primary cells

Mice were used as approved by the Experimental Animal Care and Use Committee of the Kyushu Dental University (approval number: 18-01). Slitrk1 null mice were obtained from Dr. Jun Aruga (Nagasaki University).¹³ Calvarial osteoblasts were isolated from 2 to 3 days old Slitrk1 null and WT littermates by digestion with 0.1% collagenase (Fujifilm Wako Pure Chemical Corporation, Osaka, Japan) and 0.2% dispase (Godo Shusei, Tokyo, Japan). Bone marrow stromal cells (BMSCs) were collected from femurs and tibial bones of 8-week-old Slitrk1 null and WT mice. BMSCs were flushed with α-minimal essential medium (α-MEM) (Fujifilm Wako) after cutting the edge of the bones. Spleen cells were isolated from mice 7-week-old Slitrk1 null mice or WT littermates.

Cell lines

Cos7 cells (JCRB9127, JCRB) and C3H10T1/2 cells (JCRB9080, JCRB) were purchased from Riken. The cell lines were tested for mycoplasma contamination using a MycoAlert Mycoplasma Detection Kit. Cos7 cells and C3H10T1/2 cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) (Wako) supplemented with 10% FBS and 1% penicillin/streptomycin at 37°C in an atmosphere of 5% CO₂.

Method details

Radiography and microcomputed tomography (μCT)

Radiographic images of the femur and tibia from 3, 7 and 12-week-old Slitrk1 null (n = 5) and their Wt littermates (n = 7) male mice were obtained using a μFX-1000 (Fujifilm) at 25 kV for 10 s. For μCT, the femurs were kept in 70% ethanol after collection and fixed with 4% paraformaldehyde. Micro CT analyses were performed using a Skyscan 1272 (Bruker, Kontich, Belgium). Bones were imaged using an X-ray tube at 55 kV and 181 μA. For the trabecular bone, the region of interest (ROI) was defined as 200 slices starting 300 μm above the peak of the growth plate to the proximal in the femur. For cortical bone, the ROI was defined as 200 slices starting 4,000 μm above the peak of the growth plate to the proximal in the femur. Cross-sectional images were reconstructed and analyzed.⁴²

Bone histomorphometry analysis

Eight (12-week-old), 6 (7-week-old) or 4 (3-week-old) day and 2 days prior to dissection, 20 mg/kg calcein (Sigma) was administered subcutaneously. Tibiae and femurs were collected from Slitrk1 null (n = 5) and WT (n = 7) male mice, fixed in 4% paraformaldehyde, and stored in 70% ethanol. After dehydration with a series of graded ethanol concentrations, the bones were embedded in a mixture of methyl methacrylate and 2-hydroxyethyl methacrylate (Fujifilm Wako Pure Chemical, Osaka, Japan) without decalcification.⁴³ Four-μm-thick sagittal sections from tibiae were stained with Von Kossa stain for image acquisition using Olympus cellSense software (Olympus). Growth plate thickness was measured using Histometry RT camera software (System Supply). Consecutive sections were stained with 0.05% toluidine blue solution (pH 7.0), and we evaluated the cellular parameters of trabecular bone in a 0.75 mm high × 0.7 mm wide region located 300-μm apart from the growth plate of proximal tibia. Dynamic parameters of cortical bone were evaluated using unstained sections from midshaft located approximately 4 to 5-mm apart from the growth plate of distal femur. Histomorphometric parameters were expressed according to standardized nomenclature.⁴⁴

Ectopic bone formation assay

Bovine collagen sheets (CollaCote; Zimmer Dental) were used to prepare 6-mm diameter discs. The collagen disks were soaked with 1 μg of recombinant human BMP-2 (R&D Systems) and lyophilized. All procedures were performed under sterile conditions. 8-week-old male Slitrk1 null (n = 3) and WT littermates (n = 3) were anesthetized using three mixed anesthesia (0.3 mg/kg Medetomidine; Domitol, Meiji Seika Pharma Co., Ltd, Tokyo, Japan/4.0 mg/kg Midazolam; WAKO/5.0 mg/kg Butorphanol; Vetorphale, Meiji Seika Pharma Co., Ltd.).⁴⁵ Pellets were surgically implanted under the fascia of the back muscle. The mice were subcutaneously administered meloxicam (0.2 mg/kg) for analgesia immediately and 1 day after surgery.⁴⁰ The ectopic bone was measured using a caliper, and paraffin sections were prepared. The sections were stained with hematoxylin and eosin and observed.

Osteoblast differentiation and alkaline phosphatase assay

Primary osteoblasts and Bone marrow stromal cells were maintained in α-minimal essential medium (α-MEM) (Fujifilm Wako) supplemented with 10% FBS (Sigma-Aldrich, St. Louis, MO) and 1% penicillin/streptomycin (Fujifilm Wako) at 37°C under 5% CO₂. Differentiation of confluent cells was initiated by treatment with 50 mM β-glycerophosphate (Sigma aldrich) and 50 μg/mL ascorbic acid (Sigma aldrich).

Alkaline phosphatase (ALP) activity of primary osteoblasts was observed after 3 days using recombinant human (rh) BMP-4 (R&D Systems). The cell layers were washed with phosphate-buffered saline (PBS) and fixed with an acetone/ethanol mixture (50:50, v/v). The cell lysate was incubated at 37°C in the buffer containing a substrate solution comprising 0.1 M diethanolamine, 1 mM MgCl₂, and 1 mg/mL p-nitrophenyl phosphate (WAKO). The reaction was stopped by adding 3 M NaOH. Absorbance was measured at 405 nm using an iMark Microplate Absorbance Reader (BioRad).⁴⁰

Osteoclast differentiation and TRAP staining

Spleen macrophages were cultured under 5% CO₂ at 37°C in α-MEM (Fujifilm WAKO) with 10% FBS, and 1% penicillin/streptomycin (Fujifilm WAKO). The spleen macrophages were subsequently cultured with 100 ng/mL sRANKL (Oriental Yeast Co., Ltd.) and 30 ng/mL M-CSF (Oriental Yeast Co., Ltd.) on 48-well culture plates for 7 days. The medium was changed every 2 days.

Prior to staining cells were fixed with 10% formaldehyde for 10 min, followed by addition of ethanol–acetone (1:1) mixture for 1 min. Cells were then stained using a leukocyte acid phosphatase (TRAP) staining kit (Sigma Aldrich) according to the manufacturer’s protocol. TRAP-positive multinuclear cells (>3 nuclei) were considered osteoclasts.⁴⁶ Quantification was performed independently by two researchers.

Reagent for cell culture, plasmids, transfection

Transfections were performed using Lipofectamine 2000 (Thermo fisher Scientific). When indicated, cells were treated with 20 μM of MG132 (Selleck) for 30 min.

The Slitrk1 expression plasmid (accession number NM_199065) was generated by PCR cloning from a mouse cDNA library. The Slitrk1 S695A mutant, which cannot bind 14-3-3,²¹ was generated by standard site-directed PCR mutagenesis using PrimeSTAR HS DNA polymerase (TaKaRa, Ohtsu, Japan). Destination backbone for PCR products was pcDNA3.1 (Thermo Fisher Scientific).

Immunocytochemistry analysis

Primary osteoblast cells were permeabilized and blocked with PBS containing 0.3% Triton X-100 and 5% goat serum for 1 h at room temperature. Cells were then incubated overnight with primary antibodies at 4°C. The primary antibodies used were anti-Ki-67 rabbit polyclonal antibody (ab15580, Abcam) and anti-CyclinD1 mouse monoclonal antibody (72-13G, Santa Cruz). The target proteins were visualized using an Alexa 594- or Alexa 488-conjugated secondary antibody (Thermo Fisher Scientific). Nuclei were counterstained with DAPI (Dojindo, Kumamoto, Japan) and observed using a BZ-9000 microscope (Keyence, Osaka, Japan).

RNA isolation and quantitative real-time PCR (qPCR)

Total RNA was isolated from cells or tissues using the FastGene RNA Basic Kit (Nippon Genetics, Tokyo, Japan), and cDNA was synthesized using the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems). Quantitative Real-Time PCR was performed using PowerUp SYBR Green Master Mix (Thermo Fisher Scientific) on a QuantStudio 3 Real-Time PCR System (Thermo Fisher Scientific). mRNA expression levels were normalized to β-actin using the 2^−ΔΔCt method. Primer information is provided in Table S1 in Supplemental Information.

Western blotting analysis

The cells were lysed in a lysis buffer containing phosphatase (Roche) Protease Inhibitor Cocktail Set I (Fujifilm Wako) and then sonicated (TOMY Seiko). Protein concentrations were determined using the Pierce BCA Protein Assay Kits (Thermo Fisher Scientific). The supernatants were boiled in sample buffer containing 0.125 M Tris-HCl (pH 6.8), 40% glycerol, 4% sodium dodecyl sulfate (SDS), 0.2 M dithiothreitol, and 0.01% bromophenol blue and subjected to SDS-polyacrylamide gel electrophoresis. Later, the blots were transferred to a nitrocellulose membrane, which was probed with a primary antibody, followed by incubation with the appropriate horseradish peroxidase-conjugated secondary antibody. Using the Immobilon ECL Ultra Western HRP Substrate (Merck), the bands were detected through chemiluminescence and imaged using an LAS-4000 Imaging System (Fujifilm). The following antibodies were used: rabbit anti-Slitrk1 (NBP1-49541, Novas Biological), Rabbit anti-TAZ (#28287, Cell Signaling Technology), CyclinD1 mouse monoclonal antibody (72-13G, Santa Cruz), Mouse anti-Myc (M192-3, MBL), Mouse anti-DDDDK(M185-3L, MBL), Anti-β-actin (#M177-3, Medical and biological laboratories co. Ltd.), goat anti-rabbit HRP and -mouse HRP (Jackson ImmunoResearch Laboratories).

In situ hybridization

Samples were fixed in 4% paraformaldehyde in phosphate-buffered saline (PBS) and were decalcified with Morse’s solution. The samples embedded in paraffin after they were dehydrated by ethanol and xylene. Blocks were cut into 4 μm thickness. ISH was done using the digoxigenin-labeled RNA probes for Slitrk1.¹¹

Mechanical testing

Femurs were obtained from 8-week-old Slitrk1 null (n = 8) male mice or wild-type littermates (n = 9). pMOI in WT and Slitrk1 null mice was calculated using μCT analysis. Mean ± SD; n = 7 or 5. (b) Maximum load (N), stiffness (N/mm), fracture energy (N.mm), and breaking strain (mm) were determined in WT and Slitrk1 null mice using a three-point bending analysis. Mean ± SD; n = 8 or 9.

A three-point bending analysis was performed using a KUREHA SPECIAL LABORATORY CO., LTD. (Iwaki Fukushima).

Cell proliferation assay

Proliferation of primary osteoblasts from Slitrk1 null and WT littermates was assessed using a Cell Counting kit-8 (Dojindo, Kumamoto, Japan), according to the manufacturer’s protocol.⁴⁷

Quantification and statistical analysis

Comparisons were performed using analysis of variance with a Tukey–Kramer post-hoc test or two-tailed unpaired t-tests. Statistical significance is indicated as follows: ∗p < 0.05, ∗∗p < 0.01, and ∗∗∗p < 0.001. The bar graphs are presented as the mean ± SD. In the box-and-whisker plot, the box represents the interquartile range (IQR), from the 25th to the 75th percentile, and the whiskers indicate the minimum and maximum values.

Published: May 30, 2025