Research Article 4187 Low peak bone mass and attenuated anabolic response to parathyroid hormone in mice with an osteoblast-specific deletion of connexin43 Dong Jin Chung1,2,*, Charlles H. M. Castro1,3,*, Marcus Watkins1, Joseph P. Stains1,‡, Min Young Chung2, Vera Lucia Szejnfeld3, Klaus Willecke4, Martin Theis§ and Roberto Civitelli1,¶ 1Division of Bone and Mineral Diseases, Washington University School of Medicine, 660 S. Euclid Avenue, St Louis, MO 63110, USA 2Department of Internal Medicine, Chonnam University Research Institute of Medical Sciences, Chonnam National University Medical School, Gwangju, Republic of Korea 3Universidade Federal de São Paulo-Escola Paulista de Medicina, São Paulo, Brasil 4Institut für Genetik, Universität Bonn, Germany *These two authors contributed equally to this work ‡Present address: Department of Orthopaedics, University of Maryland, Baltimore, MD, USA §Present address: Institute of Cellular Neurosciences, University of Bonn, Germany ¶Author for correspondence (e-mail: [email protected]) Accepted 6 July 2006 Journal of Cell Science 119, 4187-4198 Published by The Company of Biologists 2006 doi:10.1242/jcs.03162 Summary Connexin43 (Cx43) is involved in bone development, but its mice, with Cx43–/fl exhibiting an intermediate response. role in adult bone homeostasis remains unknown. To Attenuation of PTH anabolic action was associated with e c overcome the postnatal lethality of Cx43 null mutation, we failure to increase mineral apposition rate in response to n generated mice with selective osteoblast ablation of Cx43, PTH in ColCre;Cx43–/fl, despite an increased osteoblast e ci obtained using a Cx43flallele and a 2.3-kb fragment of the number, suggesting a functional defect in Cx43-deficient S (cid:2)(cid:2) (I) collagenpromoter to drive Crein osteoblasts (ColCre). bone-forming cells. In conclusion, lack of Cx43 in 1 ell Conditionally osteoblast-deleted ColCre;Cx43–/flmice show osteoblasts leads to suboptimal acquisition of peak bone C no malformations at birth, but develop low peak bone mass mass, and hinders the bone anabolic effect of PTH. Cx43 of and remain osteopenic with age, exhibiting reduced bone represents a potential target for modulation of bone al formation and defective osteoblast function. By both anabolism. n radiodensitometry and histology, bone mineral content ur increased rapidly and progressively in adult Cx43+/fl mice o J after subcutaneous injection of parathyroid hormone Key words: Gap junctions, Connexin43, Teriparatide, Bone anabolic (PTH), an effect significantly attenuated in ColCre;Cx43–/fl agents, Conditional gene deletion Introduction 2000), providing in vivo evidence that this gap junction protein Bone-forming cells are highly coupled by gap junctions formed is required for normal bone development and osteoblastic primarily by connexin43 (Cx43) and, to a lesser degree, differentiation (Lecanda et al., 1998; Stains et al., 2003). This connexin45 (Cx45) proteins (Civitelli et al., 1993; Donahue et notion is now further supported by findings of Cx43mutations al., 1995; Steinberg et al., 1994). Several in vitro studies have in patients with oculodentodigital dysplasia (ODDD), a rare demonstrated that Cx43 is involved in modulating the congenital disease whose phenotypic features include differentiation and function of bone-forming cells as well as craniofacial malformations and syndactyly (Paznekas et al., osteocytes (Cheng et al., 2001; Lecanda et al., 1998; Schiller 2003; Richardson et al., 2004). A similar, though not identical et al., 2001a; Schiller et al., 2001c); and work from our group phenotype has been recently reported in a mouse with a indicates that Cx43 controls osteoblast gene transcription via dominant negative Cx43 mutant allele, Gja1Jrt (Flenniken et modulation of specific signaling systems required for al., 2005). Interestingly, these animals have also generalized osteoblast gene expression (Stains et al., 2003; Stains and osteopenia, thus reinforcing the notion that functional Cx43 is Civitelli, 2005b). important for bone mass accrual and maintenance. Such a Although this work has laid the foundation for premise could be tested in a full gene-deletion model, but understanding the biology of gap junction proteins in bone, unfortunately homozygous Cx43 null mice die shortly after only recent studies in human and mouse genetics have brought birth because of severe cardiovascular malformations (Reaume to the fore the biologic role of Cx43 in the skeleton. We had et al., 1995), thus precluding the use of this model to study the reported that targeted ablation of the Cx43 gene in the mouse consequences of complete lack of Cx43in the adult skeleton. leads to a skeletal phenotype characterized by retarded In vitro studies have also shown that Cx43 is critical for bone intramembranous and endochondral ossification, craniofacial cell response to a variety of stimuli and pharmacologic agents. abnormalities and osteoblast dysfunction (Lecanda et al., For example, inhibition of gap junctional communication or 4188 Journal of Cell Science 119 (20) Cx43 expression hinders osteoblast responses to fluid flow faint stain was observed in Cx43–/fl mice, which may reflect (Cherian et al., 2005; Saunders et al., 2001), or to mechanically endogenous (cid:3)-galactosidase expression (Kim et al., 2004), as induced calcium waves (Jørgensen et al., 2000). Further, the it is observed in animals lacking the lacZ reporter (Fig. 1D). action of parathyroid hormone (PTH), an important regulator In conditional Cx43-deleted mice, X-gal blue staining was of bone remodeling also seems to be dependent on gap intense in areas of more advanced ossification, such as the junctional communication. PTH increases gap junctional diaphyses of long bones, vertebral bodies, ribs, distal mandible communication between osteoblasts by modulating Cx43 and facial bones, whereas staining was not observed in the expression or function (Civitelli et al., 1998; Donahue et al., epiphyses of long bones, corresponding to cartilaginous growth 1995), and interference with Cx43-mediated gap junctional plates, nor in the skin or internal organs (Fig. 1D). Confirming communication using antisense oligonucleotides or chemical osteoblast-specific Cx43 gene deletion, X-gal stain was inhibitors disrupts both PTH-induced cAMP accumulation selectively observed in cells lining the bone surfaces of tibial (Van der Molen et al., 1996) and osteoblast differentiation cortical endosteum and trabecular surfaces of both tibia and (Schiller et al., 2001b). Based on these findings, and vertebra. As expected, most osteocytes were also X-gal considering the osteoblast dysfunction of Cx43null osteoblasts positive, whereas no stain was observed in bone marrow cells (Lecanda et al., 2000), we hypothesized that lack of Cx43 (Fig. 1E). would negatively affect skeletal responsiveness to anabolic stimuli, such as that produced by intermittent PTH Osteopenia and reduced osteoblast number in Cx43 administration, the only currently available modality for conditional knockout mice inducing new bone formation (bone anabolism) in patients with All genotypes were obtained at the expected Mendelian osteoporosis and fractures (Neer et al., 2001). frequency and were viable. Whole-mount alizarin red/alcian To determine the biologic importance of Cx43 in the adult blue staining of newborn mice did not reveal any major skeletal skeleton, we generated a conditional Cx43gene ablated mouse abnormalities in Cx43 conditional knockout mice compared model based on the Cre/loxP system (Nagy, 2000). In this with their control littermates (Fig. 2A), consistent with post- model, which overcomes the lethality of the germline Cx43 developmental deletion of Cx43. However, body weight at 1 e c null mutation, Cre expression is driven by a 2.3-kb fragment month was significantly lower in ColCre;Cx43–/flmice relative n of the (cid:2)(I) collagenpromoter, resulting in replacement of the to the other genotypes in both males and females, a difference e 1 ci entire Cx43 reading frame with the lacZ reporter cassette that persisted until at least 6 months of age (Fig. 2B). S selectively in bone-forming cells (Castro et al., 2003). With this Importantly, conditional Cx43-deficient mice exhibited ell model, Cre is expressed just before birth and in cells that are significantly lower whole-body bone density by dual-energy X- C already partially differentiated into osteoblasts, thus providing ray absorptiometry (DEXA) compared with Cx43+/florCx43–/fl of an osteoblast-specific and postnatal gene ablation model littermates by two-way ANOVA (Fig. 2C). This relative al (Dacquin et al., 2002). We find that these animals are viable, osteopenia was significant as early as 1 month of age and n but develop a low peak bone mass and remain osteopenic persisted with age at least up to 6 months (P<0.05 and P<0.01, r u throughout their adult life, the result of a reduced ability of respectively). Bone mineral content (BMC) very closely o J bone-forming cells to fully differentiate. They also exhibit a resembled the bone density data, with approximately 5% lower dramatically attenuated response to the anabolic effect of bone mass in ColCre;Cx43–/flmice relative to the wild-type and intermittent PTH administration. Thus, Cx43 is important not heterozygous equivalent mice. only for normal skeletal development, but also for peak bone Histomorphometric analysis evidenced a markedly more mass accrual and adult bone homeostasis. Pharmacologic reduced trabecular bone mass in ColCre;Cx43–/fl mice, with stimulation of gap junctional communication may enhance the approximately 40% reduction in bone volume/total volume and effect of osteoanabolic agents, such as PTH. more than 50% reduction in osteoblast number relative to wild- type littermates (Fig. 3A-C). Trabecular thickness in the Results conditional Cx43 ablated mice was likewise reduced by ~30%, Cre-mediated Cx43 gene deletion in osteoblasts without differences in trabecular number (Fig. 3D,E). Mineral Specific osteoblast Cx43 gene deletion was demonstrated in apposition rate was reduced by ~17%, although not statistically ColCre;Cx43–/flmice by different approaches. PCR of genomic significantly, relative to wild-type and heterozygous equivalent DNA extracted from bone revealed the expected 670-kb band littermates (Fig. 3F). By contrast, there were no statistical corresponding to the Cx43 deleted allele, a band that was differences in osteoclast number among the different genotypes absent from extracts of tail soft tissue (Fig. 1A). Accordingly, (Fig. 3G). The apparent discrepancy in the degree of osteopenia Cx43 immunoreactive bands were barely detectable in Western between DEXA and histomorphometric measurement is not blots of bone tissue extracts from conditionally deleted mice, uncommon (Castro et al., 2004), and reflects both a lower contrasting with strong bands in wild-type equivalent sensitivity of DEXA and different skeletal sites measured. littermates and fainter bands in heterozygous equivalent mice (Fig. 1B), the latter reflecting both the loss of one Cx43allele Delayed differentiation of Cx43-deficient osteoblasts and haploinsufficiency of the ‘floxed’ allele (Theis et al., To gain further insights into the pathobiologic mechanism of 2001). Substantial amounts of mRNA transcripts for the Cre this osteopenic phenotype, we studied calvaria cells isolated transgene were detected only in bone extracts from from genetically modified animals. Demonstrating osteoblast- ColCre;Cx43–/fl mice but not in either Cx43–/fl or Cx43+/fl specific and differentiation-dependent Cx43gene replacement, extracts (Fig. 1C). Whole-mount preparations of newborn X-gal staining was negative in ColCre;Cx43–/fl calvaria cells animals revealed strong X-gal stain in areas corresponding to upon reaching confluence, but it became progressively stronger mineralized skeleton of ColCre;Cx43–/fl mice, whereas very 1 week post-confluence onward (Fig. 4A). Progressively Connexin43 and PTH in vivo 4189 e c n e ci S ell C Fig. 1.Cre-mediated Cx43gene deletion in osteoblasts. (A) PCR of genomic DNA extracted from bone (b) or tail (t) tissue revealed the of presence of a 670-kb band corresponding to the Cx43-deleted allele selectively in the bone tissue extract. (B) Western blot of bone tissue al extracts from mice of the different genotypes showing barely detectable Cx43 immunoreactive bands in Western blots of ColCre;Cx43–/fltissue, n contrasting with strong expression of Cx43 protein in wild-type equivalent littermates, and lower but detectable expression in heterozygous ur equivalent mice. (C) Quantitation of mRNA for Cre by real-time PCR, showing the presence of the transgene only in extracts of ColCre;Cx43–/fl o bone (one femur). (D) Whole mounts of newborn mice revealing strong X-gal stain in areas corresponding to mineralized skeleton of J ColCre;Cx43–/flmice, whereas only very faint stain was observed in Cx43–/flmice. (E) X-gal stained sections of the mid-shaft (upper left) and metaphysis (lower left) of the tibia, and of one lumbar vertebra (right) of ColCre;Cx43–/flmice, showing blue stain in cells lining the bone surface in both skeletal sites, but not in bone marrow cells. Non-stained surfaces correspond to areas where the cell layer was artifactually detached from bone. F, female; M, male; Ob, osteoblasts; Ocy, osteocytes. Fig. 2.Low bone mass phenotype in conditionally Cx43deleted mice. (A) Whole mount of alizarin red and alcian blue staining of newborn mice did not reveal any major skeletal abnormalities in ColCre;Cx43–/flmice compared with their control littermates. (B) Lower body weight in ColCre;Cx43–/fl(n=46) relative to Cx43+/fl(n=69) and Cx43–/flgroups (n=42) at 6 months of age. (C) Whole-body BMD measured by DEXA monitored in vivo revealed significantly lower bone mass in ColCre;Cx43–/flrelative to wild-type littermates (P<0.05, for genotype effect; two- way ANOVA for repeated measures). 4190 Journal of Cell Science 119 (20) e c n e ci S ell C f o Fig. 3.Histomorphometric characterization of osteopenia in conditionally osteoblast Cx43-deleted mice. (A) Trichrome Masson stain of 4 (cid:4)m al n undecalcified sections of proximal tibiae from mice of the three different genotypes at 6 months. (B-G) Quantitative histomorphometry showing r significantly lower bone volume/total volume, osteoblast number and trabecular thickness in the ColCre;Cx43–/fl, with non-statistically different u o trabecular number, mineral apposition rate and osteoclast number among the three genotype groups. *P<0.05 versus Cx43+/fl(one-way J ANOVA); n=6 per genotype group. increased X-gal stain during osteoblast differentiation is mineralization after 2 weeks (Fig. 4F,G). These in vitro data entirely consistent with the expression pattern of the promoter strengthen the notion that Cx43 expression is necessary for full used to drive Cre (Dacquin et al., 2002). Accordingly, barely elaboration of the osteoblast phenotype. detectable Cx43 immunoreactive bands were detected in lysates of conditionally deleted cells, with reduced abundance Attenuated bone anabolic response to intermittent PTH of Cx43 in heterozygous equivalent cells (Fig. 4B). Likewise, in osteoblast Cx43-deficient mice Cx43 mRNA abundance, assessed by real-time PCR, was We next tested the ability of conditional Cx43-deficient mice to reduced by ~90% and ~50% in conditionally Cx43-deleted and respond to the anabolic stimulus provided by intermittent PTH heterozygous equivalent calvaria cells after 3 weeks in culture injections. In a first study, we tested 4 doses of PTH in 5- to 6- (Fig. 4C). month-old mice treated 5 days a week for 4 weeks. Because of Development of alkaline phosphatase activity, a marker of the lower bone mass in the conditionally deleted mice relative osteogenic differentiation, was significantly reduced in to the other genotypes (Figs 2, 3), in these studies we monitored calvarial cells derived from ColCre;Cx43–/flmice 2 weeks post- whole-BMC rather than bone density, to assess the absolute confluence, when it usually reaches a peak, as it occurred in amount of bone gained in each group. In the wild-type equivalent the other two genotypes (Fig. 4D). Furthermore, after 2 weeks Cx43+/fl group, PTH treatment induced rapid and dose-related in culture the abundance of mRNA transcripts for other increments in whole-body BMC, with significant increases over osteoblast-specific genes, namely osteocalcin, (cid:2)(I) collagen, baseline at 4 weeks with all doses, except the lowest one. 1 osteopontinand Cbfa1/Runx2 was reduced by more than 50%, Maximal increases were 13.1% and 13.4% in Cx43+/fl and measured by real-time PCR, relative to wild-type equivalent Cx43–/flmice, respectively, with significant bone gain as early as cells (Fig. 4E). By contrast, neither Cx45 nor N-cadherin after 2 weeks of treatment (Fig. 5A,B). However, in the mRNA were significantly different among the three genotypes conditionally Cx43-deleted ColCre;Cx43–/fl mice, only two (Fig. 4E). Importantly, ColCre;Cx43–/flcalvaria cells were not doses of PTH resulted in statistically significant increments in able to produce mineralized matrix until 3 weeks in culture, bone mass, and the maximal effect obtained (9.8%) was ~30% whereas Cx43+/fl and Cx43–/fl cells were able to start lower than that observed in the other two genotypes (Fig. 5C). Connexin43 and PTH in vivo 4191 e c n e ci S ell C f o al n r u Jo Fig. 4.Delayed differentiation of Cx43-deficient osteoblasts. (A) X-gal staining of post-confluent ColCre;Cx43–/flcalvaria cells grown in mineralizing medium, showing blue staining becoming stronger with time in culture. (B) Western analysis demonstrated barely detectable Cx43-specific bands in lysates of conditionally deleted cells, and reduced abundance of Cx43 in heterozygous equivalent cells. (C) Abundance of Cx43 mRNA, assessed by real-time PCR, was reduced in ColCre;Cx43–/flcalvarial cells relative to wild-type equivalent cells after 3 weeks in culture (n=3). (D) Development of alkaline phosphatase activity was significantly reduced in calvarial cells derived from ColCre;Cx43–/flmice 2 weeks post-confluence (n=4). (E) The abundance of mRNA transcripts for other osteoblast-specific genes was reduced by more than 50%, as measured by real-time PCR, relative to wild-type equivalent cells (*P<0.05 versus Cx43+/fl; one-way ANOVA; n=3). (F) Representative results of von Kossa stain of post-confluent calvaria cells and (G) quantitative data on three different preparations showing delayed in vitro matrix mineralization by ColCre;Cx43–/flcalvaria cell cultures (*P<0.05 versus Cx43+/fl; Tukey test; n=3). However, taking into account the increase occurring in untreated were just slightly higher but not statistically different than the wild-type animals (4.8%), the difference in response amplitude changes observed in a group of wild-type equivalent mice would be more than 40%. treated with vehicle (3.4±3.9%; n=9) (Fig. 6A). Region- Rather surprisingly, even in vehicle-treated groups we specific analysis on BMC changes by DEXA also revealed that detected a basal increase in BMC, presumably reflecting PTH significantly increased bone mass (>12%) at the lumbar continuous bone growth in these 5-6-month-old animals. spine only in the wild-type equivalent group, whereas no Because this may confound the extent of bone gain obtained changes occurred at this site in the conditionally Cx43-deleted with PTH, we repeated a similar study in older mice (7.4- to mice (Fig. 6B). Conversely, the anabolic effect of PTH on 9.6-months-old), whose bone mass should be stable. In this femur BMC was not affected by genotype, exhibiting an case, we used 40 (cid:4)g/kg PTH, a dose that induced maximal anabolic response of almost equal magnitude for each group effects in all genotypes in the younger animals. Again, 4-week (Fig. 6C). treatment with 40 (cid:4)g/kg PTH induced significant changes of whole-body bone mass in Cx43+/fl (12.5±4.7% from baseline; Attenuated stimulation of bone formation after PTH n=15) and Cx43–/fl (9.3±4.6%; n=11) mice, whereas the treatment in osteoblast Cx43-deficient mice anabolic effect of PTH was reduced by 47% in the Bone histomorphometric analysis was fully consistent with the ColCre;Cx43–/flgroup (6.7±5.3%; n=10). The changes in bone DEXA results. After a 4-week treatment with 40 (cid:4)g/kg PTH, mass induced by PTH in the conditionally deleted animals bone volume (BV)/total volume (TV) was increased almost 4192 Journal of Cell Science 119 (20) Fig. 5.Attenuated BMC response to PTH in osteoblast Cx43-deficient mice. Percent change from baseline of whole body BMC after a 4-week treatment with different doses of teriparatide (PTH), showing a dose-related increase of bone mass in wild-type Cx43+/flmice (A). Effects of lesser magnitude were observed in heterozygous equivalent Cx43–/flmice at the intermediate doses (B), whereas the effect of PTH treatment was uniformly attenuated at all doses in ColCre;Cx43–/flmice (C). *P<0.05, **P<0.01 versus vehicle (one-way ANOVA); n=6-8 per each data point. Data are mean ±s.e.m.; n=12 (vehicle) and 6-8 (PTH-treated groups). e c threefold in Cx43+/fl mice compared with mice of the same was also highest in Cx43+/fl mice, but the changes were not n e genotype treated with vehicle. A significant increase of lesser statistically significant (Fig. 7H). By contrast, osteoclast ci magnitude was also observed in the heterozygous equivalent perimeter was higher in the Cx43–/fl and ColCre;Cx43–/fl S group, Cx43–/fl, whereas BV/TV was not different in the groups, but even in this case the differences were not ell ColCre;Cx43–/fl group relative to the vehicle-treated group statistically significant (Fig. 7I). C (Fig. 7A-D). Osteoblast number was increased in all genotypes Dynamic histomorphometric parameters of bone formation f o with no statistical differences among groups, even though this were assessed at two skeletal sites, to further investigate al parameter was ~30% lower in conditionally Cx43-deleted mice differences in PTH responses at the spine and femur. Abundant n r relative to Cx43+/fl littermates (Fig. 7E). Conversely, other double-calcein labels were observed in wild-type mice after u o static histomorphometric parameters of bone formation, PTH treatment at both the spine and at the endosteal surface J trabecular number and thickness were significantly increased of the tibia. Double labelling was also present in the in Cx43+/fl (~20 and ~40%, respectively) but not in heterozygous equivalent mice, whereas in the majority of ColCre;Cx43–/flor Cx43–/flmice (Fig. 7F,G). Cortical thickness conditional knockout mice only single labels were detected Fig. 6.Region-specific changes in BMC in osteoblast Cx43-deficient mice. (A) Percent change from baseline of whole body BMC after a 4- week treatment with 40 (cid:4)g/kg PTH in a group of 7.4-9.6-month-old mice, showing an attenuated response in conditionally deleted ColCre;Cx43–/flmice (n=10) relative to wild-type animals (n=15), whereas response in Cx43–/flmice (n=11) was intermediate. (B) Response was absent at the lumbar spine in both ColCre;Cx43–/fland heterozygous equivalent Cx43–/fl. (C) Significant increases in bone mass were instead detected in all genotypes on the total femoral area. **P<0.01, *P<0.05 versus vehicle (n=9); #P<0.05 versus ColCre;Cx43–/fl; one-way ANOVA. Connexin43 and PTH in vivo 4193 e c n e ci Fig. 7.Static bone histomorphometric analysis after a 4-week treatment with 40 (cid:4)g/kg PTH. (A-C) Trichrome Masson stain of the proximal S tibia showing less abundant trabecular bone mass in the conditional knockout (cKO) ColCre;Cx43–/flrelative to heterozygous equivalent (Het) ell Cx43–/fland wild-type (WT)Cx43+/flmice. (D) Robust anabolic response occurred in the WT group, and an attenuated response was observed C in Het mice. By contrast, no significant increases in bone volume/total volume were detected in the conditionally deleted ColCre;Cx43–/flmice f relative to the other genotype groups. (E) Osteoblast number was significantly increased by PTH treatment in all groups. (F) Trabecular number o was significantly increased in the wild-type groups only. (G) The same result was observed for trabecular thickness. (H) No significant changes al were detected for cortical thickness. (I) Osteoclast number was not different among the different groups. *P<0.05 versus vehicle (ANOVA); n r n=9-15. u o J (Fig. 8A,C). Consequently, mineral apposition rate (calculated An important role for Cx43 in bone homeostasis and for the in the trabecular and endosteal surfaces) was significantly function of bone-forming cells was postulated by several in lower in ColCre;Cx43–/flmice than in the Cx43+/flgroup in both vitro studies (Civitelli et al., 1993; Donahue et al., 2000; sites (Fig. 8B,D). However, periosteal mineral apposition rate Schiller et al., 2001a), and it was established by analysis of in the tibia was not significantly different among groups, even mice with a germline null mutation of the Cx43 gene, which though the average was lower in ColCre;Cx43–/fl mice exhibit delayed ossification of both endochondral and (0.423±0.246 (cid:4)m/day) relative to Cx43+/fl (0.614±0.629 intramembranous skeleton and defective osteoblast (cid:4)m/day) and Cx43–/fl mice (0.710±0.497 (cid:4)m/day), a result differentiation (Lecanda et al., 2000). Craniofacial very consistent with the cortical thickness data. Finally, 5- malformations are not present in ColCre;Cx43–/fl mice, most bromo-2(cid:5)-deoxy-uridine (BrdU)-positive cells were observed likely because in these animals Cx43is deleted at around birth on the bone surface in all genotypes after PTH treatment (Fig. (Dacquin et al., 2002), and thus embryonic development would 8E), and the osteoblast mitotic index was not different among be expected to be normal. However, the osteoblast defect is groups (Fig. 8F). reproduced in ColCre;Cx43–/fl mice, a defect that leads to significant osteopenia throughout life. Interestingly, Discussion generalized osteopenia is also present in Gja1Jrt/+mice, which The present study demonstrates that selective deletion of Cx43 carry a point mutation of the Cx43 gene (Flenniken et al., in osteoblasts leads to a marked decrease in peak bone mass 2005), and whose phenotype resembles that of human ODDD, and osteopenia; it also severely attenuates the bone anabolic a rare autosomal dominant condition characterized by response to intermittent administration of PTH. These craniofacial (ocular, nasal and dental) malformations, limb abnormalities are caused by a functional defect in bone- dysmorphisms, spastic paraplegia and neurodegeneration forming cells, which fail to increase their activity in response (Loddenkemper et al., 2002; Schrander-Stumpel et al., 1993). to the hormonal stimulus. Thus, functional Cx43 is required for Human ODDD has been linked to mutations of the Cx43gene normal bone mass acquisition and maintenance and it is (Kjaer et al., 2004; Paznekas et al., 2003; Richardson et al., involved in the mechanism of action of PTH-induced 2004), however both Cx43 null and Gja1Jrt/+ mice exhibit anabolism. impaired skull ossification (Flenniken et al., 2005; Lecanda et 4194 Journal of Cell Science 119 (20) e c n e ci S ell C f o al n r u o J Fig. 8.Dynamic bone histomorphometric analysis and osteoblast proliferation after a 4-week treatment with 40 (cid:4)g/kg PTH. (A) Fluorescent micrographs (200(cid:6)) of undecalcified sections of lumbar spine trabecular bone showing double calcein labels in both wild-type Cx43+/fl(WT) and heterozygous equivalent Cx43–/fl(Het) mice, but only single labels in conditional knockout ColCre;Cx43–/fl(cKO) mice. (B) Mineral apposition rate (MAR) in the lumbar spine was significantly lower in cKO ColCre;Cx43–/flmice than in either HT or WT Cx43+/flgroup (n=5- 6). (C) Calcein labeling and (D) mineral apposition rate in the endosteal surface of the tibia showing attenuated response to PTH in cKO mice (n=5-6). (E) BrdU stain of endosteal tibial surface, showing positively stained cells equally in mice of all genotypes (n=3-4). (F) Osteoblast mitotic index, expressed as percentage of BrdU-labeled cells per total number of cells on the bone surface. *P<0.05 versus Cx43+/fl, one-way ANOVA. al., 2000), whereas osteosclerotic changes are described in confirm that Cx43is required for full osteoblast differentiation patients with ODDD (Paznekas et al., 2003; Schrander- and functional activity, although in vivo interference with gap Stumpel et al., 1993). Such a discrepancy may be related to junctional communication between osteoblasts and other cells species differences, or to mechanisms by which different on the bone microenvironment may also contribute to the ODDD mutations affect connexin function. Nevertheless, there phenotype. Because the 2.3-kb fragment of the (cid:2)(I) collagen 1 is now evidence from different mouse genetic models promoter we used to delete Cx43 is expressed in committed consistently demonstrating that interference with Cx43 in the osteoblasts (Dacquin et al., 2002), it is likely that the decreased postnatal skeleton leads to a low bone mass phenotype. osteoblast number in bone of conditionally deleted mice As noted, the cellular bases of the phenotype observed in reflects a delayed differentiation rather than a decreased conditionally Cx43-deleted mice suggest a defect in osteoblast recruitment of new osteoblasts, a conclusion also supported by differentiation and function, previously observed in the similar proliferation rates of bone cells in wild-type and deleted germline Cx43 null mutants (Lecanda et al., 2000), and very mice. Of course, this conclusion does not exclude other likely present also in the Gja1Jrt/+ mouse (Flenniken et al., functions of Cx43 at earlier stages of osteoblast differentiation 2005). Accordingly, ColCre;Cx43–/fl mice have a low as postulated by studies in the Gja1Jrt/+ mouse (Flenniken et osteoblast number, modestly decreased mineral apposition al., 2005). rate, delayed in vitro osteoblast differentiation, and profound Although low bone mass is present in both ColCre;Cx43–/fl deficit in osteoblast-specific gene expression. These results and Gja1Jrt/+ (Flenniken et al., 2005) mutants, the molecular Connexin43 and PTH in vivo 4195 mechanisms leading to osteopenia may be different. In the to produce new bone under the hormonal stimulus, as Gja1Jrt/+ mice the mutation is germline and acts as dominant demonstrated by ~70% lower mineral apposition rate in negative (Roscoe et al., 2005; Shibayama et al., 2005), whereas ColCre;Cx43–/fl than in wild-type mice after a 4-week in our model the mutation is recessive and it occurs only in treatment with PTH, despite a significant increase in osteoblast committed osteoblasts. Furthermore, osteoblasts also express number. Considering that osteoblast number is decreased in Cx45(Civitelli et al., 1993) and although this connexin forms untreated ColCre;Cx43–/fl mice, the results seem to indicate gap junction channels of different biophysical properties than that the hormone is still able to stimulate osteoblast recruitment those formed by Cx43 (Steinberg et al., 1994; Veenstra et al., to the bone surface in conditional Cx43-deficient mice, 1994), Cx45 might be sufficient to support some degree of gap although these cells are obviously impaired in their ability to junctional communication in the absence of Cx43. This may synthesize new bone in response to PTH. Because Cx43 provide a partial compensatory mechanism for the lack of deletion occurs in cells that are already fully committed to the Cx43, even though Cx45 expression is not upregulated in osteogenic lineage, it is likely that some of PTH effects, for conditionally deleted cells. By contrast, the Gja1Jrt/+ variant example recruitment or proliferation of osteoprogenitors, occur may interfere with both connexins, or other interacting at a stage when Cx43 deletion has not yet taken place, or are proteins, thus inhibiting the function of both Cx43 and Cx45 Cx43 independent. Although earlier studies indicated that (Giepmans, 2004; Saez et al., 2003). These concepts are not at intermittent PTH administration activates existing bone lining odds with the established notion that Cx45 overexpression cells without affecting cell proliferation (Dobnig and Turner, reduces Cx43 function (Koval et al., 1995; Lecanda et al., 1995), an increase in bone marrow osteoprogenitor cells has 1998), because while in a mixed Cx43/Cx45 environment the been reported in response to PTH in rats (Kostenuik et al., biophysical properties of Cx45 prevail, in a Cx43 null 1999) and mice (Tanaka et al., 2004), and osteoblast number background, as it occurs in our mouse model, the presence of is consistently increased in mice (Iida-klein et al., 2002; Knopp Cx45 would allow a certain degree of cell-cell communication et al., 2005). Furthermore, because osteoblast number was that may partially compensate for lack of Cx43. increased and osteoblast proliferation was not altered in the The consequences of osteoblast-specific ablation of Cx43 conditionally deleted animals, it is unlikely that our gene e c are more severe under the stimulatory action of intermittent manipulation may have affected the anti-apoptotic action of n administration of PTH, reflected by the dramatic attenuation of PTH to a substantial degree (Jilka et al., 1999). e ci the anabolic effect of PTH in ColCre;Cx43–/fl mice. Although the molecular aspects of the interaction between S Interestingly, although responses of lesser magnitude were also Cx43- and PTH-induced bone anabolism remain to be ell observed at intermediate doses of PTH in Cx43–/fl mice, in elucidated, we had previously observed that PTH upregulates C which the abundance of Cx43 in osteoblasts is reduced, the Cx43 expression and function in osteoblasts (Civitelli et al., of highest dose of PTH used (80 (cid:4)g/kg) elicited a response 1998), and more recently, we demonstrated that interference al similar to wild-type mice. By contrast, effects on bone mass with Cx43 alters transcriptional regulation of specific gene n that were maximal in wild-type and heterozygous equivalent promoter elements, via MAP kinase- and protein kinase C- r u animals were never achieved in the conditionally deleted mice, dependent pathways (Stains et al., 2003; Stains and Civitelli, Jo and no further gains were obtained with doses above 20 (cid:4)g/kg. 2005b). Because PTH signal transduction involves both of Thus, the gains in bone mass that can be induced by PTH are these pathways, it is possible that Cx43 is required to minimal, though not totally absent, when osteoblasts are appropriately integrate PTH-activated signals and/or to deprived of Cx43 in vivo, a conclusion consistent with the equalize hormonal responses throughout the osteoblast notion that the anabolic response to PTH requires functional network (Stains and Civitelli, 2005a). Consistent with this Cx43. Instead, reduced Cx43 abundance in Cx43–/fl mice may hypothesis, in preliminary results we find that interference with be sufficient to support some osteoblast functions but not Cx43 function reduces the capacity of osteoblastic cells to others. In particular, PTH upregulation of Cx43 expression increase osteocalcin gene transcription under stimulation by (Civitelli et al., 1998) is likely to be attenuated when Cx43 is PTH (De Marzo et al., 2005). It is worth mentioning that the decreased, and this may contribute to attenuation of PTH distribution of PTH receptors is not uniform throughout the anabolic effect we have seen in Cx43–/fl mice. Similar bone tissue, and even within cell lines, certain signal responses observations have been made in the study of Cx43 function in are not homogeneous (Civitelli et al., 1992). The present results astroglia, where the Cx43flallele shows haploinsufficiency for have interesting ramifications for development of therapeutic some phenotypical parameters but not for others (Theis et al., strategies for bone anabolism. The nature of the defect in 2003). We also observed skeletal site-specific differences in response to intermittent PTH in our animal model makes it Cx43 sensitivity to PTH anabolic effect in the conditionally likely that similar attenuations of bone mass responses may deleted animals by regional DEXA analysis. It is possible that occur for other anabolic agents or stimuli, i.e. mechanical load, lack of Cx43attenuates PTH response on trabecular bone to a because activation of bone-forming cell function is the ultimate greater extent than it does on cortical bone, thus potentially requirement for manufacturing new bone. It is also reasonable explaining the normal response in femur observed by DEXA to believe that the osteoanabolic response to PTH could be in conditionally Cx43-deleted mice. Envelope- or site-specific enhanced by increasing gap junctional communication using effects of PTH have been reported, with more pronounced bone pharmacologic agents, thus allowing lesser doses or less mass increments observed in the trabecular than in the cortical frequent parenteral administration of PTH. Furthermore, the component (Calvi et al., 2001; Gunness-Hey and Hock, 1984; requirement of osteoblast/osteocyte Cx43 for the anti-apoptotic Iida-klein et al., 2002). action of bisphosphonates (Plotkin et al., 2005), which are also The attenuated osteoanabolic response to PTH is the widely used in the therapy of osteoporosis, can now be tested consequence of a failure of Cx43-deficient bone-forming cells in vivo. 4196 Journal of Cell Science 119 (20) In summary, we have demonstrated that selective Cx43gene and bone, as described (Lecanda et al., 2000; McLeod, 1980). For whole-mount X- deletion in osteoblasts results in adult osteopenia, delayed gal staining, carcasses of newborn mice were fixed for 2 hours in 2% formaldehyde, osteoblast differentiation, and greatly attenuated osteoanabolic 0.02% paraformaldehyde, 5 mM EGTA, 0.1 mM MgCl2and 0.1 M NaPO4, pH 7.3, then washed in a solution containing NP-40 and stained for 3 hours in X-gal response to PTH, the consequence of a failure of Cx43- substrate (5-bromo-4-chloro-3-indolyl-D-galactopyranoside) 1 mg/ml, as described deficient bone-forming cells to mount a full response to the (Frendo et al., 1998). They were then transferred to a 1% KOH solution in 20% hormone. Cx43-mediated gap junctional communication glycerol until they were cleared and then stored in glycerol. For X-gal staining of bone sections, tibiae or lumbar spine were fixed in 2% paraformaldehyde and 0.02% represents a potential target for modulation of bone anabolic glutaraldehyde for 1 hour, and then decalcified in 4% EDTA for 17 days. Decalcified stimuli. bones were incubated in 0.1% (v/v) X-gal substrate (see above) for 12 hours, post- fixed in 4% paraformaldehyde and embedded for paraffin sectioning (Hens et al., 2005). Sections were counterstained with eosin. Materials and Methods Transgenic mice Bone mineral density (BMD) measurements Development of the mouse model used in these studies has already been reported Total body BMC and BMD were monitored by DEXA using a PIXImus scanner in some detail (Castro et al., 2003). Briefly, a mouse strain harboring a mutant (GE/Lunar, Madison, WI), under anesthesia with 100 mg/kg ketamine and 10 mg/kg ‘floxed’ Cx43 allele (Cx43fl) (Theis et al., 2001) was mated to mice expressing Cre xylazine i.p., as described (Castro et al., 2004). Heads were excluded from the under control of a 2.3 kb (cid:2)1(I) collagenpromoter fragment (abbreviated as ColCre) analysis by masking. Region-specific BMD was also measured at the spine and (Dacquin et al., 2002), so that Cre-mediated recombination replaces the entire Cx43 femur, by identifying regions of interest corresponding to the L1-L6 area or the reading frame with the lacZ reporter cassette. Homozygous Cx43fl/fl mice were entire femur, respectively. In the latter case, animals were positioned with the femur generated first and crossed with ColCre mice also carrying a Cx43 null allele at a 45°angle with the tibia, and values for both femurs averaged. Calibration was (ColCre;Cx43+/–). This strategy avoids potential effects of activation of Crein the performed daily with a standard phantom as suggested by the manufacturer. The parental germ line. These crosses generate, in approximately equal numbers, the precision of whole-body BMD, assessed by the root mean square method is 1.34% Cx43 conditionally deleted mice, ColCre;Cx43–/fl, as well as three additional (coefficient of variation) (Castro et al., 2004). genotypes, Cx43+/fl(wild-type equivalent), Cx43–/fl(heterozygous equivalent), and ColCre;Cx43+/fl(conditional heterozygous). All the mouse lines used in this project Bone histomorphometry were developed in a mixed C57BL/6-C129/J background and littermate were used Mice were labeled twice by injection of calcein (15 mg/kg i.p., Sigma-Aldrich) on as controls. Mice were fed regular chow ad libitum and housed in a room maintained days 7 and 2 before euthanasia, and bone samples were prepared according to at constant temperature (25°C) on a 12 hours of light and 12 hours of dark schedule. previously described methods, with some modifications (Castro et al., 2004). Genotyping was performed by PCR on genomic DNA extracted from mouse tails, Briefly, dissected tibiae or lumbar spine were fixed in 70% ethanol and either e after digestion with proteinase K, as described (Lecanda et al., 2000). The Cx43 decalcified in 14% EDTA for 14 days and embedded in paraffin, or left undecalcified enc nCuCllT a GlleAleA w GasC dAe tGecAteTd 3 u(cid:5)s; iCngx 4p3ri-m3(cid:5)e: r5s(cid:5) CAxA43T- 5C(cid:5)G: 5A(cid:5) GTGTGT CGACAA CGGCTT GTGGAA GTGATT GTCCAA aMnads seomn bteridcdherdo mine tmecehthnyiql ume,e athnadc rtyarlatrtaet.e -Prleassitsitca nste acctiiodn sp hwosepreh astatasien eadc tiuvsitiyn gs ttahine ci AGC C 3(cid:5)and Neor-5(cid:5): 5(cid:5)GGA TCG GCC ATT GAA CAA GAT GGA TTG CAC was used for paraffin sections, which were counterstained with methyl green and S 3(cid:5). Primers Cx43-5(cid:5)and Cx43-3(cid:5)amplify a 900-bp product within the Cx43 coding thionin for identification of osteoclasts and osteoblasts (Liu and Kalu, 1990). Eight ell irneg tihoen .n Purlli malelre lNe,e oanr-d5 (cid:5)whhyebnr iudsizeeds wtoit hth ep rnimeoemr yCcxi4n3 r-e3s(cid:5)i,s tiat nacme pclaifisseest tae 1p.r4e-skebn tb oannldy, (cid:4)himsto smecotripohnos mweetrrey lwefat su pnesrtfaoinrmede dfo irn d aynn aarmeaic 1 b7o5n-e8 7h5is (cid:4)tomm odrispthaol mtoe tthrye. gQrouwantthi tpatliavtee C spanning the Neocassette and part of the adjacent Cx43gene (Houghton et al., using the OsteoMeasure software program (Osteometrix, Atlanta, GA) in an of P19C9R9 )b. uPfCfeRr, w0.a0s8 pmerMfo romf eeda cihn daA fiTnPa,l dvCoTluPm, ed GofT 2P,5 d (cid:4)TLT Pr,e a1c (cid:4)tioMn ;p 2r immMers M, 2g.5C lU2, T1a(cid:6)q efoplilflouwoirnegs cpenarcaem meitcerros scoof pbico nsye streemm,o adse dlientga ilwede reel seewstihmeraet e(dC a(sPtarorfi ettt ael.t, 2a0l.0, 41)9. 8T7h)e: al DNA polymerase, 1-5 (cid:4)g genomic DNA. The DNA was denatured at 94°C for 3 trabecular bone volume as a percentage of total tissue volume, trabecular thickness n minutes and amplified for 35 cycles (94°C for 30 seconds, 70°C for 45 seconds and (in (cid:4)m), trabecular number (per (cid:4)m), trabecular separation (in (cid:4)m), osteoblast ur 72°C for 120 seconds) followed by a final extension at 70°C for 20 minutes. perimeter per bone perimeter (in percent) or osteoblast number per trabecular area o Primers UMP(5(cid:5)TCA TGC CCG GCA CAA GTG AGA C 3(cid:5)) and UMPR(5(cid:5) (in number/mm2), osteoclast perimeter per bone perimeter (in percent), and mineral J TCA CCC CAA GCT GAC TCA ACC G 3(cid:5)) were used for the simultaneous apposition rate (in (cid:4)m/day), calculated as the mean distance between two detection of the ‘floxed’ (Cx43fl) and wild-type (Cx43+) alleles, as described (Theis fluorescent labels divided by the number of days between the labels. et al., 2001). These primers generate a 1 kb amplicon corresponding to the Cx43fl allele, and a 900 bp band, corresponding to the wild-type allele. In some Cell culture and phenotypic characterization experiments, the deleted Cx43allele was directly identified in whole bone extracts, Osteoblast-enriched calvaria cultures were prepared from newborn mice by after homogenization and phenol/chloroform extraction. This was accomplished sequential collagenase digestion as described (Castro et al., 2004; Lecanda et al., using primers Cx43delforw(5(cid:5)GGC ATA CAG ACC CTT GGA CTC C 3(cid:5)) and 2000), and grown in (cid:2)-modified essential medium ((cid:2)MEM; Mediatech, Herndon, Cx43delrev(5(cid:5)TGC GGG CCT CTT CGC TAT TAC G 3(cid:5)), which encompass the VA), supplemented with 10% fetal bovine serum (FBS; Atlanta Biologicals, junction between the Cx43gene intron and the (cid:3)-galactosidasecoding region, thus Norcross, GA) and 100 IU/ml penicillin and 100 (cid:4)g/ml streptomycin (Sigma generating a 670 bp amplicon, corresponding to the Cx43-deleted allele (Theis et Chemicals, St Louis, MO). Approximately 3-5 calvariae were pooled to prepare the al., 2001).The ColCretransgene was detected by using the primers Cre 1123-1104: cell cultures used in each experiment. Cx43 gene deletion was assessed in 5(cid:5)-AAG TGC CTT CTC TAC ACC TG-3(cid:5), Cre 982-1002:5(cid:5)-TGC TTA TAA CAC differentiating osteoblasts by (cid:3)-galactosidase activity after fixation in 2% CCT GTT ACG-3(cid:5), MS1: 5(cid:5)-GCT CAG CAA GCT CAC AGC AA-3(cid:5), and LM6: paraformaldehyde, and incubation in a solution containing 1 mg/ml X-gal substrate 5(cid:5)-GAG CTT ACA CAT TTC GTC-3(cid:5). These primers generate 141 bp Cre-specific (see above), as described (Castro et al., 2003). Osteogenic differentiation was amplicon and a 448 bp Cre-negative amplicon. assessed by monitoring alkaline phosphatase activity and in vitro mineralization by von Kossa staining in the presence of 50 (cid:4)g/ml ascorbic acid and 10 mM (cid:3)- Experimental design glycerophosphate, using standard techniques (Castro et al., 2004; Lecanda et al., A total of 157 mice were used for basal phenotypic characterization. For the PTH 2000; Shin et al., 2000). Enzymatic activity was normalized for total protein content studies, 90 (45 males and 45 females) 5- to 6-month-old animals of three genotypes, (Bio-Rad protein assay kit) and expressed as nmol of p-nitrophenol produced from Cx43+/fl, Cx43–/fland ColCre;Cx43–/fl, were subcutaneously injected (5 days a week p-nitrophenyl phosphate per minute per mg of protein. Mineralization was for 4 weeks) with either vehicle (0.9% saline containing 0.1% BSA and 0.001N quantitated by calculating the surface area covered by dark stain per well, using HCl; n=12) or human recombinant PTH (1-34) (Teriparatide®, Eli-Lilly, digital image-processing software (IPLab v.3.5; Scanalytics, Rockville, MD), as Indianapolis, IN) at doses of 10 (cid:4)g/kg (n=19), 20 (cid:4)g/kg (n=18), 40 (cid:4)g/kg (n=18), previously described (Lecanda et al., 2000). and 80 (n=23) (cid:4)g/kg of body weight. Mice were weighed on the second week of treatment, and the amount of PTH injected was adjusted for any change in weight. Cell proliferation A second group of older mice (7.4- to 9.6-month-old; n=45) was also treated with In vivo cell proliferation was assessed by BrdU incorporation, determined by 40 (cid:4)g/kg PTH with the same modalities as just detailed. immunoassay, according to the manufacturer’s instructions (5-Bromo-2(cid:5)-deoxy- uridine labeling and detection Kit III, Roche Molecular Biochemicals). For in vivo Whole-body mounts and X-gal staining labeling, 100 (cid:4)g BrdU (Sigma, St Louis, MO, USA) per gram of body weight in After sacrifice, newborn mice were skinned, eviscerated and maintained for 24 hours PBS was injected i.p. 2 hours before sacrifice. Longitudinal, 5 (cid:4)m sections of in ethanol 100%. After fixation in acetone for 24 hours, the carcasses were stained paraffin-embedded tibiae, prepared as described above, were rehydrated and in a solution containing alizarin red 0.1%, alcian blue 0.3%, acetic acid and 70% incubated for 10 minutes with 30% H2O2in absolute methanol (1:9) and processed ethanol (1:1:1:17). They were then transferred to a 1% KOH solution in 20% in denaturing and blocking solutions following the manufacturer’s protocol. BrdU glycerol until they were cleared and then stored in glycerol for analysis of cartilage incorporated into nuclei was detected by immunostaining (Zymed Laboratories,