Ellis-van Creveld (EvC) syndrome is a skeletal dysplasia, characterized by short limbs, postaxial polydactyly, and dental abnormalities. EvC syndrome is also categorized as a ciliopathy because of ciliary localization of proteins encoded by the two causative genes, EVC and EVC2 (aka LIMBIN). While recent studies demonstrated important roles for EVC/EVC2 in Hedgehog signaling, there is still little known about the pathophysiological mechanisms underlying the skeletal dysplasia features of EvC patients, and in particular why limb development is affected, but not other aspects of organogenesis that also require Hedgehog signaling. In this report, we comprehensively analyze limb skeletogenesis in Evc2 mutant mice and in cell and tissue cultures derived from these mice. Both in vivo and in vitro data demonstrate elevated Fibroblast Growth Factor (FGF) signaling in Evc2 mutant growth plates, in addition to compromised but not abrogated Hedgehog-PTHrP feedback loop. Elevation of FGF signaling, mainly due to increased Fgf18 expression upon inactivation of Evc2 in the perichondrium, critically contributes to the pathogenesis of limb dwarfism. The limb dwarfism phenotype is partially rescued by inactivation of one allele of Fgf18 in the Evc2 mutant mice.

Taken together, our data uncover a novel pathogenic mechanism to understand limb dwarfism in patients with Ellis-van Creveld syndrome. Ellis-van Creveld (EvC) syndrome is a congenital skeleton disorder characterized by short limbs. Recent studies indicated that EVC and EVC2, the proteins encoded by two causative genes of EvC syndrome, play important function in transducing Hedgehog signaling, a signaling pathway critical for embryonic development. The defective Hedgehog signaling in chondrocytes is therefore the speculated reason for dwarfism in EvC patients. However, despite the apparent skeletal abnormalities observed in EvC patients, other tissues that require Hedgehog signaling are relatively normal. To understand how skeletal development is specifically affected in EvC syndrome, we analyze the limb skeletogenesis using Evc2 mutant mice.

Series: Game On, Book 1. By Kristen Callihan. Available ebook formats: epub mobi pdf rtf lrf pdb txt html. How to calculate radiometric dating.

Our data demonstrated that mutation in Evc2 only moderately affected Hedgehog-PTHrP feedback loop in the growth plate, which only partially contributes to the dwarfism. Additionally, the elevated Fibroblast Growth Factor (FGF) signaling, another signal important for embryonic development, critically contributes to the pathogenesis of limb dwarfism. We identified that loss of EVC2 function in the perichondrium, the tissue surrounding growth plate chondrocytes, is critical to develop the dwarfism in the Evc2 mutant mice. Overall, our data uncover a novel pathogenic mechanism to understand limb dwarfism in EvC patients and suggest that therapies for Achondroplasia caused by elevated FGF signaling may be applicable for relieving dwarfism found in EvC syndrome.

Introduction Ellis-van Creveld syndrome (EvC), a chondroectodermal dysplasia and mesoectodermal dysplasia, is an autosomal recessive congenital disease []. EvC patients generally bear a variety of defects such as shorter limbs and ribs, postaxial polydactyly, as well as dysplastic nails and teeth. Previous genetic studies have shown that more than two thirds of EvC patients carry mutations in either EVC or EVC2 [, ]. Interestingly, LIMBIN, originally identified as the causative gene for chondrodysplastic dwarfism in Japanese Brown cattle, was later discovered as the cattle orthologue of EVC2 []. A recent study reported spontaneous mutations in EVC2 in Tyrolean Grey cattle []. In both cases, EVC2 homozygous mutant cattle suffered from chondrodysplastic dwarfism, suggesting evolutionarily conserved functions of EVC2 during appendicular bone development. Appendicular bone development occurs through endochondral ossification, a process during which a series of locally produced factors are interacting to ensure the correct length and shape of each skeletal element [, ].

Fibroblast Growth Factor (FGF) signaling, mediated by FGF18 produced in the perichondrium, is involved in this regulatory network [, ]. In the growth plate, FGF signaling inhibits chondrocyte proliferation through STAT1-mediated p21 expression [, ] and inhibits chondrocyte differentiation through MEK/pERK-mediated signaling []. In addition to FGF signaling, Indian Hedgehog and parathyroid hormone-related protein (PTHrP) signaling plays a central role in the regulation of chondrocyte proliferation, differentiation and maturation. Indian Hedgehog is synthesized by pre-hypertrophic chondrocytes [] and stimulates PTHrP synthesis in chondrocytes at the distal end of the growth plate []. PTHrP promotes proliferation of chondrocytes and prevents them from progressing to pre-hypertrophic differentiation directly [] and through suppression of FGF receptor 3 ( Fgfr3) expression [].

Once proliferating chondrocytes are far away enough from the source of PTHrP, they differentiate into pre-hypertrophic chondrocytes, start synthesizing Indian Hedgehog, and then further mature into hypertrophic cells. Recently, primary cilium has been identified as a cellular organelle essential for Hedgehog signaling transduction []. Upon Hedgehog signaling induction, Smoothened (SMO) moves into the cilium []. At the same time, a protein complex consisting of glioma-associated oncogene (GLI) and Suppressor of Fused (SUFU) moves into the cilium, where it is quickly dissociated [–]. GLI proteins subsequently traffic out of the cilium and move into the nucleus, where they function as transcriptional activators of Hedgehog responsive genes []. Recent studies have also demonstrated that EVC and EVC2 interact with each other and are mutually required for localization at the base of the primary cilium [–]. Protein structure analysis coupled with biochemical analysis demonstrated that both EVC and EVC2 are N-terminus-anchored type I transmembrane proteins [].

Despite that there is no functional domains identified in EVC or EVC2, they intracellularly localize at the base of the primary cilium through interaction with EFCAB7 and IQCE []. Upon induction of Hedgehog signaling, the EVC/EVC2 complex interacts with SMO at the base of cilium, which affects GLI2 accumulation in the ciliary tips [, ].

Consistent with in vitro mechanistic studies, diminished Hedgehog signaling was observed in the growth plates of mice lacking either Evc or Evc2 [,, ]. Since the Hedgehog-PTHrP loop has a central role in regulating chondrocyte proliferation and maturation, thereby determining the length of the appendicular skeleton, disrupted Hedgehog signaling in the growth plate is the speculated reason for dwarfism in the EvC syndrome [, ]. Previous genetic studies have indicated that Hedgehog signaling plays critical roles during many processes during mouse embryonic development. For example, a loss-of-function mutation in Shh leads to holoprosencephaly, neural tube defect, and a variety of midline defects, such as cyclopia, cleft palate and cleft lips [, ], and a single digit and other limb skeletal defects []. A loss-of-function mutation in Indian Hedgehog ( Ihh) was shown to severely impair endochondral ossification, resulting in extremely short and undermineralized limb bones [] and in early closure of the growth plates []. Despite the indispensable functions implicated by previous studies for EVC/EVC2 in transducing Hedgehog signaling, except for short limbs, the aforementioned severe defects are observed neither in EvC patients nor in Evc and Evc2 mutant mice.

This discrepancy between the function of EVC/EVC2 in transducing Hedgehog signaling and phenotypic observations in Evc or Evc2 mutant mice prompted us to ask to what extent does the loss of function of Evc/Evc2 reduce Hedgehog signaling, and why does Evc/Evc2 loss of function specifically impact limb development. To answer these questions, we investigated Hedgehog signaling and FGF signaling levels during defective endochondral ossification in Evc2 mutant mouse lines []. We demonstrate that a nonsense mutation in Evc2 that mimics mutations seen in EvC patients, leads to compromised but not abrogated Hedgehog signaling. In addition, we found that FGF signaling is significantly elevated in the Evc2 mutant growth plate due to increased expression of Fgf18 in the perichondrium.

We successfully demonstrated that inactivation of one allele of Fgf18 in the Evc2 mutant mice partially rescued the dwarfism phenotype. We conclude that both reduced Hedgehog signaling and elevated FGF signaling play a critical role in the pathogenesis of the unique form of dwarfism that characterizes Evc2 mutants. Our findings explain differences between proposed functions of EVC/EVC2 based on biochemical approaches and symptoms found in the EvC patients, and thus provide insight for better options to treat dwarfism found in EvC patients. Evc2 mutation affects chondrocyte proliferation and maturation We generated Evc2 mutant mice by introducing a premature stop codon, along with an IRES-LacZ cassette, into exon12 (equivalent to human exon 14) to mimic one of the nonsense mutations identified in human patients []. Compared with control littermates, homozygous mutant mice showed a decrease in body length as well as in appendicular bone length at 4 weeks of age (). They did not show a difference in body length or body weight at birth (), but they already had shorter limb bones [].

Evc2 mutant mice were thus born with disproportionately short limbs. Staining of heterozygous mutant tibia growth plates for β-Gal activity indicated that Evc2 was expressed in chondrocytes as well as the perichondrium (), which is consistent with RNA in situ hybridization results []. Histological analysis of humeral growth plates indicated shorter hypertrophic and proliferating chondrocyte zones, and fewer hypertrophic chondrocytes in Evc2 mutant embryos at E18.5 (). Similar characteristics were observed at E16.5 (), and E14.5 (). On the other hand, histologic analysis at E12.5, when chondrocytes were just starting to differentiate from condensed mesenchyme, showed no difference in the length of cartilage primordia ().

The same tendency was also observed in other limb skeletal elements, such as ulna, radius, femur and tibia []. These observations suggest that mutation of Evc2 leads to dwarfism by affecting growth plate chondrocyte proliferation and/or maturation, but not by affecting mesenchymal condensation or differentiation of condensed mesenchymal cells into chondrocytes.

Hedgehog signaling is compromised but not abrogated in Evc2 mutant growth plates It has been reported that EVC2 is a ciliary protein [, ]. To examine whether Evc2 mutation leads to loss of ciliary EVC2, we visualized EVC2 protein in embryonic growth plates using an antibody recognizing the N-terminus of EVC2. As expected, EVC2 was localized at the base of the cilia in control growth plates, but was undetectable in cilia of Evc2 mutant growth plates ().

Similarly, our previous work [] indicated that Evc2 mutant primary chondrocytes do not have ciliary EVC2 or EVC. These findings strongly suggest that the truncation mutation at exon12 of Evc2 leads to abrogation of ciliary localization of both EVC2 and its interaction partner EVC. Hedgehog signaling, mediated by Indian Hedgehog in the growth plate, plays an important role in directing chondrocyte proliferation and hypertrophic maturation.

Consistent with previous studies in Evc and Evc2 mutant mice [, ], we also detected decreased Hedgehog signaling in Evc2 mutant growth plates []. To evaluate the remaining Hedgehog signaling levels, we dissected out E16.5 tibia cartilage from control and Evc2 mutant growth plates for RNA isolation.

QRT-PCR for Gli1, Ptch1 and Pthrp (), which are direct targets of Hedgehog signaling, indicated that Hedgehog signaling was significantly reduced, up to 40% of its normal level, in Evc2 mutants. These observations are consistent with our previous report on the Hedgehog signaling level using Gli1-lacZ reporter mice []. On the other hand, Ihh expression in Evc2 mutant growth plates remained at the control level (). The results from in situ hybridization were consistent with expression analysis from qRT-PCR. Despite increased signal intensity of Ihh expression detected in Evc2 mutants, the expression area is more restricted in Evc2 mutant growth plate (), which is consistent with previous reports [, ]. To test whether this reduced Hedgehog signaling reflected impaired response of Evc2 mutant chondrocytes to the Hedgehog ligand, we isolated primary chondrocytes and examined their response to Hedgehog signaling induction by a smoothened agonist (SAG). Similarly to what we observed in vivo, quantification of Gli1 mRNA levels indicated that Hedgehog signaling was reduced to 60% and 40% in SAG-treated limb and rib primary chondrocytes from Evc2 mutant mice, respectively, compared to SAG-treated control cells ().

The accumulation of GLI proteins in ciliary tips is regarded as a hallmark of Hedgehog signaling induction []. To analyze the ciliary localization of GLI2, we co-stained E16.5 tibia growth plates for GLI2 with acetylated tubulin, a ciliary marker. In contrast with previous observations in cultured cells [], we found decreased GLI2 accumulation in ciliary tips in both resting and proliferating chondrocytes of Evc2 mutant growth plates (), while the percentage of cilia with GLI2 staining remained the same in control and Evc2 mutants ().

Induction of Hedgehog signaling in primary chondrocytes also resulted in diminished accumulation of GLI2 (), SUFU () and KIF7 () in ciliary tips, with no effect on ciliary accumulation of SMO (). To further confirm that tissues/cells bearing Evc2 mutation can still respond to Hedgehog signaling to some extent, we treated Evc2 mutant tibiae with the Smoothened agonist, SAG, ex vivo and observed significant increases in tibia length compared with untreated Evc2 mutant tibiae (). All aforementioned data thus concurred that the mutation of Evc2 leads to compromised but not abrogated Hedgehog signaling, likely due to impaired accumulation of Hedgehog components in ciliary tips. Evc2 mutant growth plates display elevated FGF signaling We sought to further dissect the mechanisms by which Evc2 loss of function specifically impacts limb development. The aforementioned decrease in length of the hypertrophic chondrocyte zones in Evc2 mutants is characteristic of Achondroplasia, the most common form of dwarfism in humans caused by gain of function mutations in FGF receptor 3 (FGFR3) [, ] To examine whether Evc2 mutant growth plates have altered FGF signaling, we first examined their phospho-ERK level by immunohistochemistry. In E16.5 Evc2 mutant tibiae, we detected elevated phospho-ERK in both resting and proliferating chondrocytes, but not in hypertrophic chondrocytes ().

Quantification indicated an increase of about 50% in Evc2 mutant growth plates compared to controls (). Beside elevating ERK phosphorylation, FGF signaling is also known to slow down cell cycle progression through STAT1-mediated p21 expression [, ]. Immunofluorescence for STAT1 demonstrated less plasma membrane-associated and more nuclear STAT1 in resting and proliferating chondrocytes in Evc2 radii than in controls (). On the other hand, as a negative control, in the perichondrium, there was no nuclear STAT1 detected in Evc2 mutants, which is consistent with a previous report that only FGFR3 (expressed in resting and proliferating chondrocytes) but not FGFR2 (expressed in perichondrium) can induce STAT1 nuclear translocation []. Increased FGF signaling is detected in Evc2 mutant growth plates.

To confirm the elevation of FGF signaling in Evc2 mutants, we examined the expression of FGF signaling targets in the growth plate. Spry2, Spry3, and Spry4 were all significantly overexpressed in Evc2 mutant growth plates ().

An increase of Spry3 expression was also detected in the embryonic limbs by in situ hybridization (). It is known that in chondrocytes, STAT1 activation up-regulates p21 expression [, ]. Consistently, we detected an elevated level of p21 mRNA in embryonic growth plate cartilage in Evc2 mutants (). In conclusion, the mutation of Evc2 led to an elevation of FGF signaling that is likely to contribute to the dwarfism phenotype.

Increased expression of Fgfr3 in chondrocytes and Fgf18 in perichondrium likely contributes to the elevation of FGF signaling in Evc2 mutants To dissect the mechanism leading to elevated FGF signaling in Evc2 mutants, we first evaluated the expression levels of FGF receptors 1, 2 and 3 in the growth plate. Consistent with our observation that elevated FGF signaling was only detected in resting and proliferating chondrocytes, we detected elevated expression for Fgfr3 (), which is exclusively expressed in resting and proliferating chondrocytes [], but did not detect any expression change for Fgfr2, which is specifically expressed in perichondrium [] or Fgfr1, primarily expressed in osteoblasts [].

The increased Fgfr3 expression is also supported by the in situ hybridization of Fgfr3 in the embryonic limbs (). It was previously observed that PTH/PTHrP treatment could repress Fgfr3 expression in a chondrocyte cell line in vitro [] and in growth plate chondrocytes in vivo [].

As a direct target of Hedgehog signaling in the growth plate, Pthrp expression was decreased in Evc2 mutants (), an effect that may lead to derepression of Fgfr3 expression. To test this potential regulatory network, we evaluated the impact of PTH on the expression of Fgfr3 in wild type primary chondrocytes. PTH (1–34) binding to the PTH/PTHrP receptor can elicit downstream signaling as a substitute for PTHrP in chondrocytes []. Treatment of primary chondrocytes with PTH for 24 h led to a 75% decrease in Fgfr3 expression (). These results thus suggest that elevation of Fgfr3 expression, occurring as a secondary effect of compromised Hedgehog signaling, may contribute to the increase in FGF signaling observed in Evc2 mutant growth plates.

FGF18, an FGF ligand documented to regulate endochondral ossification, is produced by the perichondrium surrounding the growth plates [,, –]. To assess Fgf18 expression in Evc2 mutants, we isolated perichondrium and detected a two-fold increase in Fgf18 mRNA level in Evc2 mutants compared to controls (). On the other hand, we did not detect changes in the expression levels of Fgf9 or Fgf23, FGFs that also regulate skeletal development [, ]. Elevated Fgf18 expression in the perichondrium was corroborated by Fgf18 RNA in situ hybridization in tibia growth plates and immunohistochemistry of GFP in Evc2 mutant carrying an Fgf18 GFP: CreEr allele (). Taken together, our data suggest that both elevated Fgfr3 expression, as a secondary effect of compromised Hedgehog signaling, and elevated Fgf18 expression in the perichondrium contribute to elevating FGF signaling in Evc2 mutant growth plates. Tibia ex vivo culture suggests elevated FGF signaling in Evc2 mutant growth plates To examine if increased endogenous FGF signaling in Evc2 mutant growth plates is a potential cause of the dwarfism phenotype, we set up tibia ex vivo cultures and promoted bone growth by suppressing endogenous FGF signaling. Tibiae from control and Evc2 mutants grew at similar rates during a 7-day culture without FGF signaling inhibitor treatment (1.13±0.03 and 1.16±0. Download Saint Seiya Movie Sub Indo Mp4 there. 03, ).

At a low concentration of U0126 (20 μM), an inhibitor of MEK (activated by FGFR and activator of ERK), tibiae from control and mutant mice grew faster than untreated samples (1.38±0.03 and 1.34±0.03, respectively, p. Compromised Hedgehog signaling partially contributes to the pathogenesis of dwarfism in Evc2 mutants Taken together, all results presented so far suggest that in addition to compromised Hedgehog signaling, elevated FGF signaling also contributes to the shortening of Evc2 mutant growth plates. In Evc2 mutants, compromised Hedgehog signaling is possibly due to impaired ciliary accumulation of GLI2; while higher FGF signaling is likely due to (1) elevated Fgfr3 expression in the proliferative zone as a result of compromised Hedgehog signaling, and (2) elevated Fgf18 expression in the perichondrium. Thus, in a chondrocyte-specific deletion of Evc2, we would expect to exclude the impact of increased expression of Fgf18 in the perichondrium.

Aggrecan enhancer-driven, tetracycline-inducible Cre ( ATC) is a transgenic allele containing an A ggrecan gene enhancer with internal tetracycline regulatory elements that allows Tetracycline-dependent expression of C re recombinase specifically in growth plate and other chondrocytes, but not in the perichondrium [] (). To specifically delete Evc2 in chondrocytes we generated ATC; Evc2 floxed mice. Examination of E18.5 bone showed a significant decrease in Gli1 and Pthrp expression in growth plate chondrocytes (), indicating decreased Hedgehog signaling in these cells. At the same time, we detected no change in Spry3 expression (readout for FGF signaling) in the growth plate () and no change in Fgf18 expression in the perichondrium (), indicating that neither Fgf18 expression nor FGF signaling was altered, despite elevated Fgfr3 expression (). In situ results for Gli1 and Spry3 also support the notions of compromised Hedgehog signaling while no difference for FGF signaling in tibia at E18.5 ().

These Evc2 conditional mutants showed only a moderate decrease in tibia length () compared to Evc2 germ line mutants (). At E18.5, these mutants only displayed a 10% decrease in the total length of tibiae (), 20% decrease in the length of the hypertrophic chondrocyte zone (), 20% decrease in the number of hypertrophic chondrocytes (), and a 13% decrease in the length of the proliferating chondrocyte zone (), but no difference in the total length of the growth plate (). Elevated FGF signaling critically contributes to the pathogenesis of dwarfism in Evc2 mutants To further demonstrate that Evc2 loss of function in the perichondrium is critically involved in the elevation of FGF signaling and pathogenesis of dwarfism, we used the Dermo1(Twist2) Cre allele [] that demonstrates recombination in both chondrocytes and the perichondrium (). In Evc2 floxed mice carrying Dermo1 Cre, we detected a significant decrease in Gli1 and Pthrp expression () and an increase in Fgfr3 expression () in the growth plate, as expected. At the same time, we detected increased Spry3 expression in the growth plate () and increased Fgf18 expression in the perichondrium (), indicating elevated FGF signaling.

We also detected a dramatic decrease of tibia length (), which was similar to that observed in Evc2 germ-line mutants (). More specifically, at E18.5, we detected a 21% decrease in the overall length of tibiae (), 52% decrease in the length of the hypertrophic chondrocyte zone (), 44% decrease in the number of hypertrophic chondrocytes (), 36% decrease in the length of the proliferating chondrocyte zone (), and a 21% decrease in the total length of the growth plate (). Taken together, these data demonstrate that compromised Hedgehog signaling mediated by Evc2 mutation only partially contributes to the dwarfism in Evc2 mutants; additionally, elevated FGF signaling, mediated by Evc2 mutation in the perichondrium, plays a critical role in the pathogenesis of dwarfism in Evc2 mutants.

Dwarfism in Evc2 mutant mice is partially rescued by loss of one copy of Fgf18 To further demonstrate that elevated FGF signaling plays critical roles during the pathogenesis of dwarfism in Evc2 mutant mice, we inactivated one allele of Fgf18 in Evc2 mutant mice ( Evc2 ex12/ex12; Fgf18 LacZ/+). Compared to Evc2 ex12/ex12 mutants, removal of one copy of Fgf18 allele partially rescued the dwarfism in the Evc2 ex12/ex12; Fgf18 LacZ/+mutant (). More specifically, the length of Evc2 mutant tibia is 75% of control, while the length of Evc2 ex12/ex12; Fgf18 LacZ/+ tibia is about 82% of control (, n = 6, p. Discussion The EvC syndrome is categorized as a ciliopathy due to the ciliary localization of the proteins encoded by two causative genes, EVC and EVC2. The identification of a function for EVC/EVC2 in transducing Hedgehog signaling has classified the pathogenic reason for EvC syndrome as defective Hedgehog signaling through aberrant Hedgehog-PTHrP feedback loop in the growth plate. Here, through genetic approaches, we demonstrate that compromised Hedgehog-PTHrP feedback loop mediated by Evc2 mutations only partially contributes to the dwarfism; compromised Hedgehog-PTHrP feedback loop is not sufficient in itself to impact growth plate function as severely as observed in Evc2 germ-line knockout mice. We have additionally demonstrated that elevation of FGF signaling in growth plates due to loss of Evc2 function in the perichondrium plays a critical role in the pathogenesis of dwarfism in Evc2 mutants.

Overall, our developmental studies provide direct in vivo evidence for the function of EVC/EVC2 in transducing Hedgehog signaling; and more importantly, our genetic approaches to investigate the dwarfism in Evc2 mutant mice uncover a novel pathogenic mechanism to understand limb dwarfism in patients with Ellis-van Creveld syndrome. Evc2 deletion of exon12 abrogates the ciliary function of EVC2 In this work, Evc2 mutant mice were generated that mimic one of nonsense mutations identified in EvC human patients.

Although the mutation is different from that of previously reported Evc2 mutant mouse line [], in which Evc2 was deleted from exon 1, our strategy too resulted in abrogation of EVC2 protein in cilia. Using an antibody recognizing the N-terminus of EVC2, we could not detect any ciliary EVC2 () [], and as a result, EVC also lost its ciliary localization []. Le Petit Prolexis 3 Crack on this page. This observation is consistent with previous studies that deletion of 83 amino acids or more in the C terminus of EVC2 leads to complete loss of ciliary localization [, ]. Thus, in our Evc2 mutant, the ciliary function of the EVC/EVC2 complex is completely abolished, just as it is when Evc2 is deleted from exon1. Consistently, the length of each appendicular bone is reduced in our Evc2 mutant mice as much as in the previously reported mutants [].

Compromised but not abrogated Hedgehog signaling is consistent with the clinical features of EvC and the phenotypes in Evc2 mutant mice Previous in vitro biochemical and molecular biological studies indicated that the EVC/EVC2 complex is essential during Hedgehog signaling [, ]. However, congenital defects due to abnormally diminished Hedgehog signaling, such as neural tube and midline defects, are not present in EvC patients or in Evc/ Evc2 mutant mice. On the other hand, our in vivo and in vitro studies suggest that an inactivating mutation of Evc2 leads to compromised but not abrogated Hedgehog signaling. Therefore, we favor the idea that compromised but not abrogated Hedgehog signaling in Evc2 mutant embryos leads to limited aspects of developmental abnormalities in the processes that require Hedgehog signaling. Similarly, although decreased Hedgehog signaling in the limb bud was detected using a Ptch1-lacZ reporter, there was no digit pattern defect in Evc2 mutant mice []. During limb development, loss of function of Indian Hedgehog leads to extremely short limbs and distorted growth plates [], which are apparently more severe than the phenotypes found in Evc [] and Evc2 mutant mice ([] and this work).

Chondrocyte-specific postnatal knockout of Indian hedgehog leads to early closure of growth plates, as a result of diminished Pthrp expression from as early as postnatal day 15 []. Similarly, diminished Hedgehog signaling caused by ablation of primary cilia specifically in cartilage also leads to early closure of growth plate by P15 [], which is observed in neither Evc nor Evc2 mutant mice for up to 6 weeks []. All these facts strongly suggest that in Evc and Evc2 mutant mice there is still a substantial level of Hedgehog signaling remaining. This notion is supported by previous observations that there is still substantial level of Gli1 expression detectable in Evc or Evc2 mutant growth plates [, ]. These phenotypic observations coincide with our finding that Evc2 mutant primary chondrocytes from embryonic limbs and ribs still retain 40% to 60% of wild type levels of Hedgehog signaling (). Elevated FGF signaling is mainly due to elevated Fgf18 expression in the perichondrium Since FGF signaling was shown to regulate ciliary length [], no studies have yet addressed how primary cilia regulate cellular responses to FGF ligands. In primary chondrocytes and MEFs, we did not detect differential responses of control and Evc2 mutant cells to FGF ligands.

On the other hand, we detected both elevated Fgfr3 expression () in the growth plate as well as elevated Fgf18 expression in the perichondrium of Evc2 mutants (). Previous reports [, ] and our current work suggest that PTHrP negatively regulates Fgfr3 expression (). Therefore, elevated Fgfr3 expression in the growth plate is possibly due to decreased Pthrp expression caused by compromised Hedgehog signaling.

However, elevated Fgfr3 expression itself appears insufficient to elevate FGF signaling, since the chondrocyte-specific deletion of Evc2 that we achieved using ATC resulted in compromised Hedgehog signaling and elevated Fgfr3 expression (), but neither an elevation of Fgf18 expression in the perichondrium () nor in an increase in FGF signaling (). Therefore, elevated Fgf18 expression in the perichondrium likely plays a major role in upregulating FGF signaling in the Evc2 mutant growth plate. It is still possible that elevated Fgfr3 has an additive impact in the presence of elevated Fgf18 expression on the final outcome of FGF signaling.

Our work also indicated that elevated expression of Fgf18 in the perichondrium was a consequence of the Evc2 mutation in the perichondrium, since elevation of Fgf18 expression was dependent upon Evc2 deletion in these cells (). The molecular mechanism of how loss of Evc2 in the perichondrium leads to elevated Fgf18 expression is currently under investigation. Elevated FGF signaling is critically important for the pathogenesis of dwarfism in Evc2 mutants ATC-dependent Evc2 conditional mutants allowed us to exclude the impact of FGF signaling and thereby to evaluate how a compromised Hedgehog-PTHrP feedback loop results in dwarfism. Despite a mild decrease in the length of their tibiae (), these mutants had growth plate defects quite different from those of Evc2 germ-line mutants (). These results clearly demonstrate that the affected Hedgehog-PTHrP feedback loop mediated by Evc2 mutation in chondrocytes is not sufficient to impact the length of proliferating chondrocyte zone, hypertrophic chondrocyte zone and growth plate as severely as observed in Evc2 germ-line mutants (). In support of this idea, Dermo1 Cre-dependent Evc2 conditional mutants exhibited a more dramatic decrease in tibia length () and their growth plate phenotype was similar to that of Evc2 germ-line mutants. Therefore, both elevated FGF signaling and a compromised Hedgehog-PTHrP feedback loop likely contribute critically to the pathogenesis of dwarfism in Evc2 mutants.

Consistent with our findings, decreased proliferating and hypertrophic chondrocyte zones were also observed in Evc mutant mice [], suggesting a shared mechanism leading to dwarfism in Evc and Evc2 mutant mice. In conclusion, the results obtained from in vitro and in vivo suggest a model wherein Evc2 inactivating mutations as well as EvC syndrome causing mutations partially compromise but do not abrogate Hedgehog signaling. The resulting compromised Hedgehog-PTHrP feedback loop only partially contributes to the dwarfism. In addition, our new findings have exposed a novel regulatory mechanism in which Evc/ Evc2 inactivating mutations in the perichondrium leads to a significant elevation of FGF signaling. Both effects, i.e., compromised Hedgehog-PTHrP feedback loop and elevated FGF signaling, likely synergize to render the severe dwarfism that characterizes the EvC syndrome. This model thus also suggests that the available therapeutic solutions being tested for Achondroplasia could be used to relieve, at least partially, the severity of dwarfism in EvC patients.

Animals The generation of Evc2 mutant mice and Evc2 floxed mice was reported elsewhere []. Fgf18 LacZ mutant mice were reported previously [].

Fgf18 GFP: CreER mice contain a splice acceptor (SA) GFP:CreERT2 insertion into the first intron of Fgf18 (D.M.O., I.H.H. To obtain Evc2 homozygous mutant embryos, timed mating between two heterozygous Evc2 germ-line knockout mice was carried out. Noon of the date when the vaginal plug was observed was designated embryonic day 0.5 (E0.5). Evc2 floxed mice were bred with mice carrying doxycycline inducible Aggrecan enhancer-driven, tetracycline-inducible Cre ( ATC)[] or Dermo1 Cre [] mice to generate conditional deletions of Evc2. For chondrocyte-specific deletion mediated by ATC, doxycycline-supplemented chow diet (Harlan TD01306) was provided to pregnant females from E9.5. All mouse experiments were performed in accordance with University of Michigan guidelines and federal laws covering the humane care and use of animals in research.

All animal procedures used in this study were approved by the Institutional Animal Care and Use Committee (IACUC) at the University of Michigan (Protocol #PRO00005716). Histology, skeletal staining, and immunohistochemistry Limbs were dissected out from embryos, fixed in 4% paraformaldehyde (PFA), embedded in paraffin, sectioned and stained with hematoxylin and eosin (H&E) according to standard procedures. For skeletal staining, dissected limbs were skinned and stained with alcian blue and alizarin red according to [, ]. For immunohistochemistry, limbs were fixed in 4% PFA overnight at 4°C and cryo-protected in 30% sucrose in PBS solution before embedding in OCT. Specimens were cut into 10-μm sections and incubated overnight at 4°C with antibody against EVC2 (Y20, 1:50, Santa Cruz Biotechnology, Santa Cruz, CA, USA), GLI2 (H300, 1:50, Santa Cruz Biotechnology, Santa Cruz, CA, USA), SMO (N19, 1:50, Santa Cruz Biotechnology, Santa Cruz, CA, USA), SUFU (H300, 1:50, Santa Cruz Biotechnology, Santa Cruz, CA, USA), EVC (HPA008703, 1:50, Sigma, St. Louis, MO, USA), KIF7 (Ab95884, 1:100, Cambridge, MA, USA), acetylated tubulin (T6793, 1:1000, Sigma, St.

Louis, MO, USA), gamma tubulin (T5326, 1:1000, Sigma, St. Louis, MO), pERK (4695, 1:50, Cell signaling, Danvers, MA 01923), or STAT1 (Ab3987, 1:100, Cambridge, MA, USA).

Sections were then incubated with corresponding Alexa Fluor 488 conjugated secondary antibody for 1 h at room temperature before mounting with ProLong Gold Anti-fade Reagent with DAPI (, Life Technologies, Grand Island, NY, USA). All fluorescence images were acquired at room temperature by confocal microscopy (Nikon C1) through Nikon EZ-C1 3.91 and processed by Adobe Photoshop CS6. Isolation of primary chondrocytes and immunocytochemistry For primary chondrocyte isolation, rib or long bone cartilage was dissected from E18.5 embryos and digested with collagenase A (Roche, Indianapolis, IN, USA). Chondrocytes released will be subsequently cultured in DMEM (Life Technology, Grand Island, NY, USA) with 10% FBS (Atlanta Biologicals, Flowery Branch, GA, USA). Experiment will be carried out using cells within 5 passages.

For immunocytochemistry, cultured primary chondrocytes were starved in 0.5% serum for 36 h before treatment with 100 nmol of SAG (Chemicon, Billerica, MA, USA) for 4 h. Cells were then fixed in 4% PFA and permeabilized in PBS with 0.1% Triton X-100 (Sigma, St. Louis, MO, USA) before incubation with primary antibody at 4°C for overnight and with fluorescent secondary antibody. Mounting was done with ProLong Gold Anti-fade Reagent containing DAPI. RNA isolation and quantitative real-time PCR RNA was isolated from primary chondrocytes using TRIzol (Life Technologies, Grand Island, NY, USA) according to manufacturer’s instructions. For RNA isolation from embryonic growth plates, long bones were dissected out at E16.5. Growth plates were collected from tibiae and placed into TRIzol for homogenization according to manufacturer’s instructions.

For isolation of perichondrium cells, growth plates were dissected out from embryonic tibiae and digested with 1 unit /ml Dispase []. For reverse transcription, 1 μg of total RNA was reverse-transcribed using SuperScript Reverse Transcriptase (Life Technologies, Grand Island, NY, USA). Quantitative real-time PCR was performed using Applied Biosystems ViiA7, with the following taqman probes: Mm00494645_m1 for Gli1, Mm99999915_g1 for Gapdh, Mm00439612_m1 for Ihh, Mm00436026_m1 for Ptch1, Mm00436057_m1 for Pthrp, Mm00433294_m1 for Fgfr3, Mm00438941_m1 for Fgfr2, Mm00438932_m1 for Fgfr1, Mm00432448_m1 for Cdkn1a (P21), and Mm00433286_m1 for Fgf18. RNA in situ hybridization RNA in situ hybridization was carried out as previously described [] using a digoxygenin-labeled Fgf18 probe [, ]. Briefly, embryonic tissues were immediately fixed in 4% PFA, before cryo-protected in 30% sucrose in PBS. Then, 20 μm sections were treated with proteinase K, post-fixed with 4% PFA, before treated with acetic anhydride solution (Sigma, St.

Louis, MO, USA). Sectioned tissues were hybridized with RNA probe in hybridization solution containing 5X SSC, 50% formamide, 1mg/mg tRNA (Sigma, St. Louis, MO, USA), 0.1mg/ml Heparin (Sigma, St. Louis, MO, USA) at 65°C. Sectioned tissues were then digested with RNase A (Roche) and washed in post-hybridization washing solution containing 0.2X SSC, before incubation with alkaline phosphatase conjugated mouse anti-digoxygenin for overnight. Purple color for positive signal was developed through incubation sections with BM Purple for AP substrate precipitating (Roche).

S1 Fig Generation of Evc2 mutant mice. Diagram showing that the Evc2 mutant allele was obtained by inserting a premature stop codon and IRES-lacZ cassette into exon 12. Lateral views of an Evc2 mutant and a littermate control at 4 weeks of age. Quantification of body length (C) and body weight (D) at E18.5. Data are presented as percentages of controls (n = 10, p>0.2). X-gal staining of the tibia proximal growth plate of Evc2 wild-type ( Evc2 WT) and heterozygous mutant ( Evc2 +/ex12) littermates indicates Evc2 expression in growth plate chondrocytes, perichondrocytes and other neighbor tissue cells. Scale bars: 200 μm for (D).

S7 Fig Verification of the recombination activity of the Dermo1-Cre and ATC transgenes. X-gal staining of proximal tibiae from E17.5 or E18.5 embryos carrying Dermo1 Cre (A) or ATC (B) and a Cre-dependent ROSA26R LacZ allele demonstrates that ROSA26R LacZ is efficiently recombined in both chondrocytes and perichondrium of embryos carrying Dermo1 Cre, but that it is recombined only in chondrocytes in embryos carrying ATC.

Diagram showing the generation of Evc2 floxed mutant mice. LoxP sites were inserted into mouse Evc2 to flank exon13 and exon14. Scale bars: 200 μm for (A) and (B).