The Role And Mechanism Of FGFR3-regulated LECs In HO Development

PurposeHeterotopic ossification(HO)is the ectopic bone formation in soft tissues.Hereditary HO includes fibrodysplasia ossificans progressive(FOP)and progressive osseous heteroplasia(POH).Both genetic diseases develop progressively with no effective solution.FOP is mostly caused by heterozygous gain-of-function mutation(R206H)in ACVR1/ALK2 and is usually initiated in childhood through endochondral ossification.POH is intramembranous ectopic bone formation caused by inactivation mutations of GNAS.Acquired HO usually develops after major trauma,burn,surgery or central nervous system injury with a high morbidity.Patients with HO often suffer from chronic pain,joint contractures and limited mobility,which further incapacitates normal gait and daily activities.Currently,clinical therapy of acquired HO is limited to radiation,NSAIDs and surgery with high recurrence rates.HO development requires the coordinated effects of three essential factors:osteogenic progenitor cells,molecular signals triggering ectopic bone formation and proper microenvironment.Fibroblast growth factor receptor 3(FGFR3)plays an essential role in skeletal development.Meanwhile,FGFR3 also plays a vital role in the regulation of lymphatic formation.Lymphatic vessels play an important role in the clearance of local inflammatory cells and cytokines.In this study,we explored whether FGFR3,an important regulator of endochondral ossification and lymphatic formation,influences HO development,and we also try to explore the detailed underlying mechanisms.Methods1.Generation of various transgenic mouse strainsFGFR3^flox/flox;Col2a1-Cre ER^T2 mice,FGFR3^flox/flox;Prx1-Cre ER^T2 mice,Col2a1-Cre ER^T2;Rosa26^{td Tomato} mice,Prx1-Cre ER^T2;Rosa26^{td Tomato} mice,FGFR3^flox/flox;Col2a1-Cre ER^T2;Rosa26^{td Tomato} mice,FGFR3^flox/flox;Prx1-Cre ER^T2;Rosa26^{td Tomato} mice,Col2a1-Cre ER^T2;Rosa26^{m Tm G} mice,FGFR3^flox/flox;Prox1-Cre ER^T2 mice,FGFR3^flox/flox;BMPR1a^flox/+;Col2a1-Cre ER^T2 mice,FGFR3^flox/flox;BMPR1a^flox/+;Prox1-Cre ER^T2 mice,FGFR3^flox/flox;BMPR1a^flox/flox;Col2a1-Cre ER^T2 mice and FGFR3^flox/flox;BMPR1a^flox/flox;Prox1-Cre ER^T2 mice were generated.2.Tamoxifen injection and acquired HO induction in miceFor postnatal activation of Cre ER^T2,10-week-old male mice were injected intraperitoneally with 100 mg/kg tamoxifen(T5648-5G,Sigma-Aldrich)in corn oil(C8267,Sigma-Aldrich)once a day for 5 consecutive days.To induce acquired HO formation,Achilles tenotomy was performed on male mice after tamoxifen injection.3.X-ray and?CT imaging.4 and 8 weeks post Achilles tenotomy,X-ray images of HO specimens were obtained with the MX-20 Cabinet X-ray system(Faxitron X-Ray,Tucson,AZ,USA)according to the standard procedure.Undecalcified HO specimens were scanned by the viva CT 40?CT system(Scanco Medical,Brüttisellen,Switzerland)with the settings of 70 k V and 113 m A.The threshold of 170 was applied for 3-D reconstructions and analysis of HO bone volume.4.Histological assessment and immunostaining analysis.Fresh samples were fixed in 4%paraformaldehyde solution for 24 hours,decalcified in0.5M EDTA solution for 2 weeks and dehydrated in 30%sucrose solution for 24 hours.Samples were embedded in OCT(Thermo)and sectioned at 10?m intervals using a cryostat(Leica).Frozen sections were stained with hematoxylin and eosin(H&E)and Safranin O/Fast Green(SOFG)to examine HO progression in the tendon at 4 and 8 weeks post surgery.Histomorphometric analysis was conducted with the Osteomeasure Analysis System(Osteometrics).For immunofluorescence,relevant markers were stained for chondrocytes,osteoblasts,LECs and proinflammatory macrophages in the tendon post injury.FGFR3,BMPR1a and p Smad1/5 levels in LECs were also tested by co-immunostaining of LYVE1.For animal studies,about 30-50 serial sections throughout repaired Achilles tendons were obtained from each sample.We conducted the quantitative analysis with 5-8 random visual fields per section of 6-10 sequential sections in each sample.For quantification of chondrocytes,osteoblasts and macrophages,positive cells of relevant markers were counted manually with Image J software.For quantification of lymphatic vessels,LYVE1⁺cells were assessed by Angio Tool according to the standardized procedure as previously described.5.RNA extraction and quantitative real-time PCRTotal RNA was extracted from mouse Achilles tendon tissues.Quantitative real-time PCR was performed for chondrogenic and osteogenic markers to examine HO progression in the tendon.LEC markers were tested to evaluate lymphangiogenesis in the Achilles tendon post injury.6.Protein extraction and western blot analysisAchilles tendon tissues of mice post surgery were obtained and grinded with glass pestles in RIPA buffer on ice before ultrasonication for tissue protein extraction.Chondrogenic and osteogenic markers were examined by western blot analysis to evaluate HO development in the tendon.7.Isolation and culture of primary mouse tendon cells and immunofluorescent stainingRepaired Achilles tendons of mice post surgery were obtained,cut into pieces and digested with 3 mg/ml type I collagenase to isolate primary cells in the tendons.The cells were cultured in endothelial cell medium(ECM,1001,Scien Cell).Immunofluorescent stainings were performed to identify LECs in the primary cells.8.Indocyanine green(ICG)near infrared(NIR)lymphangiography of miceLocal lymphatic drainage in the repaired Achilles tendon was determined by ICG NIR lymphatic imaging as previously described.10?l ICG solution was injected intradermally into the mouse footpad about 2 mm distal to the calcaneus.ICG imaging of mouse footpads and legs was collected by a Fusion FX Edge system(Vilber Lourmat).The signal intensity in the footpad was recorded immediately after ICG injection as the initial signal intensity.ICG imaging was collected again 24 hours after ICG injection.Conditions including exposure time,focus and position of the mouse hindlimbs need to be consistent for all imaging and overexposure needs to be avoided.The images were analyzed using Evolution-Capt v18.02software.ICG clearance was quantified as the percentage of reduced ICG signal intensity in the footpad 24 hours post injection relative to the initial ICG signal intensity.9.Human HO specimen collectionAcquired HO specimens were obtained from patients undergoing surgeries in the Department of Trauma Surgery of Daping Hospital.HO specimens were collected from male patients who had previously sustained elbow fractures that were treated with internal fixation and returned for surgical treatment of acquired HO.All subjects with elbow fractures were previously healthy,nonsmoking males aged between 25 and 45.Osteogenesis stage(3-6months after fixation surgery)and maturation stage(12-18 months after fixation surgery)were defined by the period since their fixation surgery according to the reference.10.Cell culture and FGFR3 knockdown.Murine LECs(4×10⁵/well)were seeded in 12-well plates.For si RNA-mediated FGFR3knockdown,small interfering RNAs(si RNAs)for mouse FGFR3 and for a negative control(Ribo Bio)were transfected with Lipofectamine RNAi MAX Transfection Reagent(13778100,Invitrogen)at 100n M according to the manufacturer's instructions.36 hours after transfection,mLECs were collected in RIPA buffer for protein extraction.FGFR3,BMPR1a and p Smad1/5levels were examined by western blot.11.Local application of recombinant human FGF9 or VEGF-c in Matrigel5?l mixed solution of FGF9(0,0.01,0.1 or 1?g)and Matrigel was dripped on the Achilles tendon of male mice after surgery and the skin was gently sutured after the mixture was gelled.Similarly,5?l mixed solution of recombinant human VEGF-c(100-20C,Peprotech)(0 or 0.1?g)and Matrigel was applied locally in the Achilles tendon of male mice after tenotomy to examine the effect of VEGF-c on HO development in the tendon.Results1.FGFR3 deficiency in Col2⁺cells promotes acquired HO developmentThe 10-week-old FGFR3^Col2 mice and FGFR3^f/f mice were injected with tamoxifen for 5consecutive days,followed by Achilles tenotomy.The incidence of HO in the Achilles tendon of FGFR3^f/f mice at 2,4,6 and 8 weeks after surgery was 4.3%(1/23),13.0%(3/23),47.8%(11/23)and 78.3%(18/23),which was respectively increased to 18.2%(4/22),22.7%(5/22),68.2%(15/22)and 95.5%(21/22)in FGFR3^Col2 mice.Micro-CT analysis showed that HO in FGFR3^Col2 mice was aggravated compared with that in control group 4 weeks after Achilles tenotomy.SOFG staining showed that the ectopic cartilage in FGFR3^Col2 mice was significantly increased compared with that in the control group.Immunohistochemistry,q PCR,WB analysis showed that the cartilage in the tendon of FGFR3^Col2 mice and chondrogenic/osteogenic markers such as SOX9,ACAN,OSX and Runx2 were significantly increased compared with that in FGFR3^f/f mice.HO was also aggravated in mice at 8 weeks post surgery,as shown by micro-CT analysis,SOFG staining,IHC,q PCR and WB.2.Col2⁺cells adopt LEC fate during acquired HO formationImmunofluorescent staining showed that tomato⁺cells in the injured Achilles tendon were not stained with SOX9 or Runx2 in Col2^tomato mice and FGFR3^f/f-Col2^tomato mice at 8weeks after tenotomy,suggesting that Col2⁺lineage cells were not directly involved in the formation of HO.CD31 immunofluorescent staining showed that Col2⁺lineage cells were not involved in angiogenesis.Further studies showed that tomato⁺cells in the Achilles tendon of Col2^tomato mice were co-stained by LEC marker LYVE1.Col2⁺lineage cells in connective tissue and peritendineum of normal Achilles tendon were co-stained with LYVE1 at 8 weeks post TM injection.Col2⁺lineage cells were co-stained with LEC markers including LYVE1,PROX1 and PDPN in the tendon of Col2^{m Tm G} mice at 8 weeks after surgery.8 weeks after tamoxifen injection,tomato⁺cells were observed in the bone marrow of Col2^tomato mice and contributed to the muscles in the footpad of mice.Tomato⁺cells were also co-stained with the marker of mesenchymal stem cells?-SMA in the injured Achilles tendon.Primary GFP⁺cells from repaired Achilles tendon of Col2^{m Tm G} mice were immunostained by LYVE1,PROX1 and VEGFR3.Tomato⁺cells in Col2^tomato mice and FGFR3^f/f-Col2^tomato mice were co-stained with LYVE1 instead of F4/80 in the injured Achilles tendon 8 weeks after surgery.Tomato⁺cells of Prx1^tomato and FGFR3^f/f-Prx1^tomato mice were co-stained with SOX9instead of LYVE1 in the tendon at 8 weeks post surgery.No significant difference was found in the number of LYVE1⁺LECs in injured Achilles tendon of Prx1^tomato mice compared with FGFR3^f/f-Prx1^tomato mice.3.FGFR3 deficiency in LECs aggravates acquired HO formationX-ray and micro-CT analysis showed that HO was significantly increased in the tendon of FGFR3^Prox1 mice compared with control mice at 4 and 8 weeks after tenotomy.Pathological staining showed that the ectopic cartilage in the Achilles tendon of FGFR3^Prox1mice was significantly increased compared with the control group at 4 weeks after surgery.IHC showed that the SOX9 and OSX levels in the Achilles tendon of FGFR3^Prox1 mice were significantly increased compared with those of FGFR3^f/f mice.SOFG staining showed that HO in the Achilles tendon of FGFR3^Prox1 mice was significantly increased compared with control mice at 8 weeks after surgery.Runx2 and OC levels in the injured Achilles tendon of FGFR3^Prox1 mice were significantly increased compared with control mice.4.FGFR3 deficiency in LECs inhibits local lymphatic formation and enhances inflammation in the Achilles tendon after traumaLYVE1 immunostaining demonstrated that the number of LECs derived from Col2⁺cells was significantly reduced with FGFR3 deletion in FGFR3^f/f-Col2^tomato mice compared with Col2^tomato mice at 4 and 8 weeks after surgery.Results of q PCR analysis demonstrated that LEC markers in the injured Achilles tendon in FGFR3^Col2 mice were significantly reduced compared with the control group.Immunofluorescent staining showed that the amount of LYVE1⁺LECs in Achilles tendon of FGFR3^Prox1 mice was significantly reduced compared with control mice at 8 weeks after injury.ICG NIR analysis showed that lymphatic drainage was significantly reduced in the repaired tendon of FGFR3^Col2 and FGFR3^Prox1 mice.Co-staining of F4/80 and i NOS showed that the number of proinflammatory macrophages in the injured Achilles tendon of FGFR3 knockout mice was significantly higher than that in control mice at 4 and 8 weeks post tenotomy.Immunofluorescent staining showed that FGFR3 level in LYVE1⁺LECs in mature HO(12-18 months post-trauma)was significantly lower than that in osteogenic stage(3-6 months post-trauma),and the number of LECs was also significantly reduced in mature HO.Meanwhile,the number of F4/80⁺i NOS⁺proinflammatory macrophages was significantly increased in mature stage of HO.The number of LYVE1⁺LECs in the injured Achilles tendon of wild-type mice at 12weeks post surgery was significantly lower than that at 8 weeks after surgery,and the number of proinflammatory macrophages was significantly increased in the tendon at 12 weeks after tenotomy.5.FGFR3 deficiency in LECs upregulates BMPR1a-p Smad1/5 signalingImmunofluorescent staining showed that the expression of FGFR3 was significantly decreased,while the expressions of BMPR1a and p Smad1/5 were significantly increased in LYVE1⁺LECs of injured Achilles tendon in FGFR3^f/f-Col2^tomato mice compared with Col2^tomato mice at 8 weeks after surgery.WB analysis showed that the levels of BMPR1a and p Smad1/5 were significantly increased after FGFR3 knockdown in mouse LEC cell line,while the expression of Smad1/5was not significantly changed.Meanwhile,the expression of FGFR3 in LYVE1⁺LECs of injured Achilles tendon in FGFR3^Prox1 mice was significantly lower than that in FGFR3^f/f mice,while the expressions of BMPR1a and p Smad1/5 were significantly increased.6.BMPR1a deletion in FGFR3-deficient mice promotes local lymphangiogenesis and inhibits HO formationX-ray analysis revealed that HO incidence in FGFR3^Col2;BMPR1a^f/+and FGFR3^Col2;BMPR1a^f/f mice at 2,4,6 and 8 weeks post tenotomy was significantly lower than that in FGFR3^Col2 mice.The HO volume of FGFR3^Col2 mice at 4 and 8 weeks after tenotomy was significantly reduced assessed by micro-CT analysis.H&E and SOFG staining demonstrated that HO formation was significantly decreased in FGFR3^Col2;BMPR1a^f/+and FGFR3^Col2;BMPR1a^f/f mice compared with FGFR3^Col2 mice.BMPR1a deficiency significantly reduced expressions of chondrogenic and osteogenic markers in Achilles tendon 4 and 8 weeks after injury.Immunofluorescent staining revealed that the number of LYVE1⁺LECs in the injured Achilles tendon of FGFR3 and BMPR1a double knockout mice was significantly increased compared with that of FGFR3 knockout mice,while the number of F4/80⁺i NOS⁺inflammatory macrophages in the Achilles tendon was significantly decreased.7.FGF9 inhibits acquired HO formation dependent on FGFR3 in LECsWT mice were locally treated with 0.01,0.1 and 1?g doses of FGF9 after Achilles tenotomy.8 weeks after tenotomy,micro-CT and SOFG staining showed that HO volume in the tendon was significantly reduced and the expression of Sox9 and OC in HO were significantly decreased.LYVE1 immunofluorescent staining suggested that lymphangiogenesis in the injured Achilles tendon was significantly increased after FGF9treatment.Lymphatic drainage in the Achilles tendon was also enhanced and the number of local inflammatory macrophages was significantly reduced in mice treated with FGF9.However,FGFR3^Prox1 mice treated with FGF9 showed no significant change in the volume of HO or the number of LECs in injured Achilles tendon.8.VEGF-c promotes lymphangiogenesis in the tendon post surgery and inhibits HO formationAfter local VEGF-c treatment,the volume of HO in the Achilles tendon of WT mice was significantly reduced by micro-CT analysis.SOFG staining also showed that HO formation in the Achilles tendon was significantly inhibited in mice with VEGF-c treatment.Immunofluorescent staining showed that VEGF-c treatment significantly increased the number of LYVE1⁺LECs in the injured Achilles tendon,while the number of local F4/80⁺i NOS⁺inflammatory macrophages was significantly reduced.Conclusions1.Col2⁺cells act as LEC precursors to participate in lymphangiogenesis in injured Achilles tendon;2.FGFR3 knockout in LEC promotes the formation of acquired HO by inhibiting lymphangiogenesis in the injured Achilles tendon and aggravating local inflammation post injury;3.FGFR3 deficiency in LECs inhibits lymphatic formation after injury by upregulating the BMPR1a-p Smad1/5 signaling;4.Targeted activation of FGFR3 in LECs can inhibit the formation of acquired HO.