Guisheng Zhao

To examine the local actions of IGF signaling in skeletal tissue in a physiological context, we have used Cre-mediated recombination to disrupt selectively in mouse osteoblasts the gene encoding the type 1 IGF receptor (Igf1r). Mice carrying this bone-specific mutation were of normal size and weight but, in comparison with normal siblings, demonstrated a striking decrease in cancellous bone volume, connectivity, and trabecular number, and an increase in trabecular spacing. These abnormalities correlated with a striking decrease in the rate of mineralization of osteoid that occurred despite an unexpected osteoblast and osteoclast hyperactivity, detected from the significant increments in both osteoblast and erosion surfaces. Our findings indicate that IGF1 is essential for coupling matrix biosynthesis to sustained mineralization. This action is likely to be particularly important during the pubertal growth spurt when rapid bone formation and consolidation are required. To examine the local actions of IGF signaling in skeletal tissue in a physiological context, we have used Cre-mediated recombination to disrupt selectively in mouse osteoblasts the gene encoding the type 1 IGF receptor (Igf1r). Mice carrying this bone-specific mutation were of normal size and weight but, in comparison with normal siblings, demonstrated a striking decrease in cancellous bone volume, connectivity, and trabecular number, and an increase in trabecular spacing. These abnormalities correlated with a striking decrease in the rate of mineralization of osteoid that occurred despite an unexpected osteoblast and osteoclast hyperactivity, detected from the significant increments in both osteoblast and erosion surfaces. Our findings indicate that IGF1 is essential for coupling matrix biosynthesis to sustained mineralization. This action is likely to be particularly important during the pubertal growth spurt when rapid bone formation and consolidation are required. Body size and linear bone growth in mammals is affected by cellular signaling pathways controlled by growth factors and hormones (1Efstratiadis A. Int. J. Dev. Biol. 1998; 42: 955-976PubMed Google Scholar). In this regard, a major growth-promoting signaling system consisting of the insulin-like growth factors (IGF, 1The abbreviations used are: IGF, insulin-like growth factor; PBS, phosphate-buffered saline; X-gal, 5-bromo-4-chloro-3-indolyl-β-d-galactopyranoside; AP, alkaline phosphatase IGF1 and IGF2) and the type 1 IGF receptor (IGF1R) regulates embryonic growth, as shown by gene knockout experiments in mice (1Efstratiadis A. Int. J. Dev. Biol. 1998; 42: 955-976PubMed Google Scholar). IGF1 acting through IGF1R also plays central roles in postnatal growth either independently or by mediating growth hormone functions (2Lupu F. Terwilliger J.D. Lee K. Segre G.V. Efstratiadis A. Dev. Biol. 2001; 229: 141-162Crossref PubMed Scopus (643) Google Scholar). Signaling through the IGF1R tyrosine kinase receptor not only promotes cell proliferation, but also mediates anti-apoptotic actions (3Baserga R. Resnicoff M. D'Ambrosio C. Valentinis B. Vitam. Horm. 1997; 53: 65-98Crossref PubMed Scopus (61) Google Scholar, 4Nakae J. Kido Y. Accili D. Endocr. Rev. 2001; 22: 818-835Crossref PubMed Scopus (357) Google Scholar). The IGF system includes a second receptor (IGF2R) devoid of signaling properties, but serving IGF2 turnover, and at least six IGF-binding proteins (IGFBPs) of obscure functional significance (single and also some double mouse mutations ablating IGFBPs have not revealed as yet significant consequences in growth impairment). 2J. Pintar, personal communication. The IGFs are produced locally in various tissues, including bones, and exert autocrine/paracrine functions, but they are also present in serum, mostly associated with IGFBPs. Whether the circulating IGFs act systemically as hormones is currently controversial (5Yakar S. Liu J.L. Stannard B. Butler A. Accili D. Sauer B. LeRoith D. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 7324-7329Crossref PubMed Scopus (1191) Google Scholar, 6D'Ercole A.J. Calikoglu A.S. Growth Horm. IGF Res. 2001; 11: 261-265Crossref PubMed Scopus (21) Google Scholar). A number of in vitro and in vivo studies are progressively unraveling the significance of the IGF system for skeletal development and metabolic control (for a review see Ref. 7Conover C.A. Growth Horm. IGF Res. 2000; 10 Suppl. B: S107-S110Crossref PubMed Scopus (46) Google Scholar). IGF1, by stimulating the proliferation of chondrocytes in the growth plate, plays an essential role in longitudinal bone growth (2Lupu F. Terwilliger J.D. Lee K. Segre G.V. Efstratiadis A. Dev. Biol. 2001; 229: 141-162Crossref PubMed Scopus (643) Google Scholar) and is also involved in the formation of trabecular bone. In fact, chondrocytes and bone cells produce IGFs and express IGF1R (see for example Refs. 8Shinar D.M. Endo N. Halperin D. Rodan G.A. Weinreb M. Endocrinology. 1993; 132: 1158-1167Crossref PubMed Scopus (88) Google Scholar and 9Wang E. Wang J. Chin E. Zhou J. Bondy C.A. Endocrinology. 1995; 136: 2741-2751Crossref PubMed Google Scholar). Studies using osteoblast culture systems have shown that IGF1 stimulates osteoblast proliferation, accelerates their differentiation, and enhances bone matrix production (10Canalis E. Bone. 1993; 14: 273-276Crossref PubMed Scopus (95) Google Scholar, 11Birnbaum R.S. Bowsher R.R. Wiren K.M. J. Endocrinol. 1995; 144: 251-259Crossref PubMed Scopus (55) Google Scholar). In addition, IGF1 is being recognized as a critical factor for bone cell survival (12Parfitt A.M. Mundy G.R. Roodman G.D. Hughes D.E. Boyce B.F. J. Bone Miner. Res. 1996; 11: 150-159Crossref PubMed Scopus (292) Google Scholar, 13Hughes D.E. Boyce B.F. Mol. Pathol. 1997; 50: 132-137Crossref PubMed Scopus (90) Google Scholar, 14Hill P.A. Tumber A. Meikle M.C. Endocrinology. 1997; 138: 3849-3858Crossref PubMed Scopus (186) Google Scholar). Finally, IGF1 also appears to regulate bone resorption, either directly or through its action on osteoblasts that stimulate in turn the formation and function of osteoclasts (15Hill P.A. Reynolds J.J. Meikle M.C. Endocrinology. 1995; 136: 124-131Crossref PubMed Google Scholar). Because of complex relationships in the signaling processes of the IGF regulatory system (1Efstratiadis A. Int. J. Dev. Biol. 1998; 42: 955-976PubMed Google Scholar) on one hand, and intricacies in bone development (16Olsen B.R. Reginato A.M. Wang W. Annu. Rev. Cell Dev. Biol. 2000; 16: 191-220Crossref PubMed Scopus (780) Google Scholar, 17Karsenty G. Wagner E.F. Dev. Cell. 2002; 2: 389-406Abstract Full Text Full Text PDF PubMed Scopus (1204) Google Scholar) on the other, it has been quite difficult to define individualin vivo aspects of the skeletal actions of IGF1, especially those brought about by local (autocrine/paracrine) mechanisms. Nevertheless, progress is being made with the use of genetically modified mice. For example, mice with targeted overexpression of IGF1 in osteoblasts exhibited an increased bone formation rate and increased trabecular and cortical bone volume (18Zhao G. Monier-Faugere M.C. Langub M.C. Geng Z. Nakayama T. Pike J.W. Chernausek S.D. Rosen C.J. Donahue L.R. Malluche H.H. Fagin J.A. Clemens T.L. Endocrinology. 2000; 141: 2674-2682Crossref PubMed Scopus (271) Google Scholar). Remarkably, these changes occurred without an increase in the total number of osteoblasts, suggesting that locally produced IGF1 could exert its anabolic effects primarily by increasing the performance of resident osteoblasts. Nevertheless, firm establishment of causal relationships necessitates complete ablation of IGF signaling by elimination of IGF1R function. In this regard, however, the invariable neonatal lethality ofIgf1r nullizygous mice (19Liu J.P. Baker J. Perkins A.S. Robertson E.J. Efstratiadis A. Cell. 1993; 75: 59-72Abstract Full Text PDF PubMed Scopus (2597) Google Scholar) precluded an examination of the skeletal role of IGF1 during postnatal growth and development. To circumvent this problem, we have now disrupted selectively theIgf1r gene in mouse osteoblasts using thecre/loxP recombination system, and evaluated the bone phenotype. As we report here, this site-specific ablation of IGF signaling impairs the rate of bone formation and severely retards mineralization of osteoid resulting in decreased cancellous bone volume and altered trabecular structure. A DNA fragment representing the human osteocalcin (OC) promoter (20Clemens T.L. Tang H. Maeda S. Kesterson R.A. DeMayo F. Pike J.W. Gundberg C.M. J. Bone Miner. Res. 1997; 12: 1570-1576Crossref PubMed Scopus (83) Google Scholar) was cloned into the pBluescript SK(−) vector to create pOC. A plasmid containing a cDNA encoding cre, which was modified to include a nuclear localization sequence and the human β-actin 3′-untranslated region (provided by Dr. Thomas Doetschman, University of Cincinnati), was cloned into the pKBpA plasmid (18Zhao G. Monier-Faugere M.C. Langub M.C. Geng Z. Nakayama T. Pike J.W. Chernausek S.D. Rosen C.J. Donahue L.R. Malluche H.H. Fagin J.A. Clemens T.L. Endocrinology. 2000; 141: 2674-2682Crossref PubMed Scopus (271) Google Scholar) downstream from a sequence representing the rabbit β-globin second intron flanked by remnants of truncated exons. The derived fused (globin-cre) sequences were then subcloned into pOC downstream from the osteocalcin promoter, to create pOC-cre (Fig. 1 A). The insert of this plasmid (OC-cre) was excised and microinjected into fertilized eggs (FVB-N mouse strain). Transgenic lines were established from 11 founders that were identified by Southern analysis of genomic DNA. Two of the transgenic lines (4 and 6) were analyzed in detail and used in our experiments (see “Results”). Differences in the level and specificity of transgenic expression between the two lines were not detected. Expression of cre mRNA was determined by Northern analysis of total RNA (10 μg per lane) using acre cDNA probe. All animals received humane care in compliance with the local Institutional Animal Care and Use Committee. OC-cre mice were mated with homozygous conditional mutants carrying modified Igf1ralleles (with loxP sites flanking exon 3; (21Dietrich P. Dragatsis I. Xuan S. Zeitlin S. Efstratiadis A. Mamm. Genome. 2000; 11: 196-205Crossref PubMed Scopus (104) Google Scholar), to generate OC-cre/Igf1rflox/+ progeny, which were used in subsequent crosses (see “Results”). The OC-cre mice were also crossed with Z/AP reporter mice (22Lobe C.G. Koop K.E. Kreppner W. Lomeli H. Gertsenstein M. Nagy A. Dev. Biol. 1999; 208: 281-292Crossref PubMed Scopus (452) Google Scholar), to estimate the efficiency of Cre-mediated recombination (see “Results”). For routine genotyping of progeny, thecre transgene was detected by PCR (1 min at 94 °C, 1 min at 53 °C, and 1 min at 72 °C, for 30 cycles) using the primers: 5′-CAAATAGCCCTGGCAGATTC-3′ (forward) and 5′-TGATACAAGGGACATCTTCC-3′ (reverse) to generate a 260-bp to a of the promoter and the rabbit β-globin The was detected by PCR (1 min at 94 °C, 1 min at °C, and 1 min at 72 °C, for 30 cycles) with the and (see (forward) and which generate a from the or a from the To the was used in with to generate a was used to the Z/AP reporter as To the specificity and efficiency of Cre-mediated OC-cre mice from two lines and 6) were crossed with the Z/AP mice G. Wagner E.F. Dev. Cell. 2002; 2: 389-406Abstract Full Text Full Text PDF PubMed Scopus (1204) Google Scholar) carrying a transgene (see A and “Results”). recombination the is and the downstream human gene is Mice were and were in PBS, and with in on for For was in one was used for and the for as (22Lobe C.G. Koop K.E. Kreppner W. Lomeli H. Gertsenstein M. Nagy A. Dev. Biol. 1999; 208: 281-292Crossref PubMed Scopus (452) Google Scholar). For the were for min in in was in (1 X-gal, and in at for to with and from were with PBS, in and in for 10 with PBS, and then with to in at For alkaline phosphatase the were with for and then in at for 30 min to were in PBS, in for 10 and with in for at least min or was were in and and in were and in were at 10 and were to were in for min and in in for 1 at and then in in at were then in and to were in containing for 10 For were for min in and then in for at °C, from were in PBS, (see through a and For were in for min and was by in at for 30 were then with PBS, in for 10 and with for min at were in PBS, through an and then The of the was evaluated using a system P. B. R. Int. 1996; PubMed Scopus Google Scholar). were the cortical bone in the and in the For of the and cortical the bone was using a per For of the in the the bone was using a at the growth and for A region of including only cancellous bone was to the growth and for were and as I. S. R. T. G. M. M. D. M. J. Bone Miner. Res. 2001; 16: PubMed Scopus Google Scholar). The were in with an size of either (for the or (for the were using a that not on about the is either or J. F. B. T. R. J. 1995; 96: PubMed Scopus Google Scholar). For the we the total bone volume and the volume as the of tissue volume by the total volume by the bone I. S. R. T. G. M. M. D. M. J. Bone Miner. Res. 2001; 16: PubMed Scopus Google Scholar). For the cortical the bone volume bone volume and cortical were in a region at the For the cancellous bone region in the bone volume trabecular trabecular trabecular number and were and were in bone were in H.H. M.C. of Bone Scholar), and were with a C. This analysis of the in for and bone were using the modified J. J. Pathol. 1997; 14: Scholar), and to the were for and of bone and were using a (18Zhao G. Monier-Faugere M.C. Langub M.C. Geng Z. Nakayama T. Pike J.W. Chernausek S.D. Rosen C.J. Donahue L.R. Malluche H.H. Fagin J.A. Clemens T.L. Endocrinology. 2000; 141: 2674-2682Crossref PubMed Scopus (271) Google Scholar, H.H. D. W. Int. PubMed Scopus Google Scholar). In were to the of the to that only sites were In were made in the to the of trabecular bone in the For and animals were with on 1 and and or All with the of the of the of Bone and A.M. J.A. Malluche H. R.R. J. Bone Miner. Res. 2: PubMed Scopus Google Scholar). are as All were significance level of was of the two at was using the of from a at was using the All were using the for conditional crosses between and of mice. used in this in which exon of theIgf1r gene is by loxP sites in have been (21Dietrich P. Dragatsis I. Xuan S. Zeitlin S. Efstratiadis A. Mamm. Genome. 2000; 11: 196-205Crossref PubMed Scopus (104) Google Scholar). To expression in the osteoblasts of we used the osteocalcin (OC) gene promoter and mice (see and Northern analysis demonstrated that the transgene encoding the was in and in of the For the of our it was important to the specificity and efficiency of Cre-mediated recombination in osteoblasts. the could not as a reporter for this the of skeletal tissue precluded an of the level of DNA by Southern analysis only a bone were but difficult for bone as in and could not be For these we used an and crossed OC-cre with Z/AP mice G. Wagner E.F. Dev. Cell. 2002; 2: 389-406Abstract Full Text Full Text PDF PubMed Scopus (1204) Google Scholar) carrying a which of a sequence encoding and a flanked by sites that is by a human alkaline phosphatase gene A). The gene is Cre-mediated of the efficiency of action be by in various the in and the of for alkaline To the of of the expression during from were for As shown in cells were detected in at the of from these revealed that the were osteoblasts and were also for X-gal, to estimate the efficiency of Cre-mediated cells and cells the of for and were (Fig. In derived from double exhibited a in and the of in the cells of the osteoblast the was that for X-gal, to of the into the In the double transgenic cells the representing of to for To also an estimate for the of Cre-mediated DNA brought about by the we the of osteoblasts and in from mice (4 that of the cells being of were for expression in the control the in the double transgenic was only of the in osteoblasts and was or including and skeletal from the mice for and for (Fig. PCR analysis using DNA from of that Cre-mediated recombination occurred in bone the that a of the for in osteoblasts and is not we that a level of by from the with the Z/AP reporter the performance of the OC-cre is on the of the the was used in with various cre to including a S. Liu J.L. Stannard B. Butler A. Accili D. Sauer B. LeRoith D. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 7324-7329Crossref PubMed Scopus (1191) Google Scholar), a I. Zeitlin S. 2000; PubMed Scopus Google Scholar), a 2000; PubMed Google Scholar), and a T. P. Efstratiadis A. 2001; PubMed Scopus Google Scholar), complete Cre-mediated recombination has been Xuan and A. the of crosses between animals with an were for a of effects using the as a bone as The abnormalities were but in animals not Because of the expression of the cre transgene from osteoblasts for the in the of an For these animals be to as in cells osteoblasts the of the gene by the genetically but normal mice be as from To bone and turnover, and were on and control at and of when the skeletal in the mouse is C. S. N. Donahue L.R. Rosen C.J. S. J. Bone Miner. Res. 2001; 16: PubMed Scopus Google Scholar). these the and normal animals were in size and in were detected at in with a and for and control with for of and of the gene in osteoblasts not to the postnatal growth in the the were on the of the the analysis at was on the at this is a of bone at this These which were with of the trabecular bone in the and revealed that the rate of bone formation for bone was in mice of and significant in osteoblast and osteoclast per bone of were and between mutants and in trabecular bone volume or were not detected at this are shown as per at per at to control as are shown only for significant to control as are shown only for significant volume number volume rate formation are shown as per at per at to control as are shown only for significant in a on mice demonstrated in comparison with the the rate was in the with a increase in mineralization and were made in the not however, the of as by the was increased in the and this was by an increase in osteoclast erosion and In comparison with the normal the mineralization in the mutants were in in trabecular bone volume trabecular and number by an increase in trabecular and is that the cortical was also in the mice of the total number of and the not from the in not To examine bone from and control mice were by using in the of the the cancellous bone volume was normal in the mutants and a decrease in the number of the of which was increased and A). In addition, in trabecular was normal In cortical bone volume and cortical at the of the were not between the two analysis and volume are shown as volume volume number volume are shown as in a in the of to normal between the by or the two of are in the that be from the with a trabecular by was in the was detected by This is to in in that could to in factor to between the two was the use of in the a bone was for a was used for of by have for the that IGF signaling in cells of the is essential for of the mineralization at a normal significant the roles of the IGFs in bone formation and C.A. Growth Horm. IGF Res. 2000; 10 Suppl. B: S107-S110Crossref PubMed Scopus (46) Google Scholar), this be on the about the that we In the mice that we have and of the IGF system have with the of osteoblasts, in which the function of the receptor mediating the signaling of IGF was selectively In fact, of the of Cre-mediated recombination at at least as in of the by using Z/AP reporter we that of cells IGF1R were to to from the during their subsequent In this regard, the of our is of the of action of the IGFs and of the of either one of these growth which roles that are not both embryonic and postnatal IGF2 only during J. Liu J.P. Robertson E.J. Efstratiadis A. Cell. 1993; 75: Full Text PDF PubMed Scopus Google Scholar). These the establishment of a the of in the IGF1R signaling with the of the as in the and functional bone that were evaluated by and the aspects of our are currently and a of be to at a level functional relationships and their to the of R. J.P. J. Mol. Endocrinol. 2001; PubMed Scopus Google Scholar), the mouse that we have as it has brought into downstream from the of which be from the of the of bone in mice at and at a when growth has and the animals are a a be evaluated with a Nevertheless, our in the of bone at IGF signaling is for in osteoblasts anti-apoptotic at normal as by a in osteoblast and a in bone formation In addition, of coupling of and the number of osteoclasts is also this not and at of significant in were not between mutants and fact, the cell were in the a about an unexpected in the of osteoid was detected that was by a in mineralization. In addition, the skeletal representing bone the of bone which were revealed as in the number, and of of with increase in trabecular and normal bone as however, the of as tissue that the bone processes were not in mice. the in with mineralization by signaling the detected changes consequences of the of IGF signaling at the of when the has been is in this regard, that cortical bone volume was not affected in the the effects on trabecular bone could be by the that this skeletal is to the rate of particularly the of the Our indicate despite a of the processes of osteoid production on one and mineralization on the are and be as of two bone osteoblasts not only and and proteins of the matrix but also in the mineralization which to 1995; Google Scholar, Res. 1996; PubMed Scopus Google Scholar, N. R. E. Growth Scholar). and of matrix containing from the of osteoblasts, the are by of the that and are then progressively in the is that the is controlled by the osteoblasts through an of the The for a mineralization is an increase in the between osteoid and This which is increased normal in the is for some of the osteoid to for of mineralization. the normal between the of osteoid and mineralization appears to be in the of IGF as is an in some the between the that this a between the of resulting in and mineralization. to be suggesting that IGF1 could bone mineralization is to our was for example, that IGF1 in of osteoblasts in vitro in with bone formation R.S. Bowsher R.R. Wiren K.M. J. Endocrinol. 1995; 144: 251-259Crossref PubMed Scopus (55) Google Scholar), of an complex into in a of the in bone with Rosen C.J. C.M. J. Bone Miner. Res. 2001; 16: Scopus Google Scholar). is also that is to of the matrix and to be in the of the J. J.P. Int. 1996; Full Text PDF PubMed Scopus (88) Google Scholar). In this regard, it is that IGF1 has been shown to stimulate in human cells by the expression of a A. G. J.P. J. Bone. 2001; PubMed Scopus Google Scholar). is that IGF1 is for by osteoblasts. of the that we have is likely to have important it is that IGF1 is critical for both linear bone growth and skeletal our findings that IGF1 is essential for coupling matrix biosynthesis to sustained mineralization. This action could be particularly important during the pubertal growth spurt when rapid bone formation and consolidation are required. Langub and for in the

Insulin-like growth factor I (IGF-I) is an important growth factor for bone, yet the mechanisms that mediate its anabolic activity in the skeleton are poorly understood. To examine the effects of locally produced IGF-I in bone in vivo, we targeted expression IGF-I to osteoblasts of transgenic mice using a human osteocalcin promoter. The IGF-I transgene was expressed in bone osteoblasts in OC-IGF-I transgenic mice at high levels in the absence of any change in serum IGF-I levels, or of total body growth. Bone formation rate at the distal femur in 3-week-old OC-IGF-I transgenic mice was approximately twice that of controls. By 6 weeks, bone mineral density as measured by dual energy x-ray, and quantitative computed tomography was significantly greater in OC-IGF-I transgenic mice compared with controls. Histomorphometric measurements revealed a marked (30%) increase femoral cancellous bone volume in the OC-IGF-I transgenic mice, but no change in the total number of osteoblasts or osteoclasts. Transgenic mice also demonstrated an increase in the osteocyte lacunea occupancy, suggesting that IGF-I may extend the osteocyte life span. We conclude that IGF-I produced locally in bone osteoblasts exerts its anabolic effect primarily by increasing the activity of resident osteoblasts.

RUNX2, a key transcription factor for osteoblast differentiation, is regulated by ERK1/2 and p38 MAP kinase-mediated phosphorylation. However, the specific contribution of each kinase to RUNX2-dependent transcription is not known. Here we investigate ERK and p38 regulation of RUNX2 using a unique P-RUNX2-specific antibody. Both MAP kinases stimulated RUNX2 Ser319 phosphorylation and transcriptional activity. However, a clear preference for ERK1 versus p38α/β was found when the ability of these MAPKs to phosphorylate and activate RUNX2 was compared. Similarly, ERK1 preferentially bound to a consensus MAPK binding site on RUNX2 that was essential for the activity of either kinase. To assess the relative contribution of ERK1/2 and p38 to osteoblast gene expression, MC3T3-E1 preosteoblast cells were grown in control or ascorbic acid (AA)-containing medium ± BMP2/7. AA-induced gene expression, which requires collagen matrix synthesis, was associated with parallel increases in P-ERK and RUNX2-S319-P in the absence of any changes in P-p38. This response was blocked by ERK, but not p38, inhibition. Significantly, in the presence of AA, BMP2/7 synergistically stimulated RUNX2 S319 phosphorylation and transcriptional activity without affecting total RUNX2 and this response was totally dependent on ERK/MAPK activity. In contrast, although p38 inhibition partially blocked BMP-dependent transcription, it did not affect RUNX2 S319 phosphorylation, suggesting the involvement of other phosphorylation sites and/or transcription factors in this response. Based on this work, we conclude that extracellular matrix and BMP regulation of RUNX2 phosphorylation and transcriptional activity in osteoblasts is predominantly mediated by ERK rather than p38 MAPKs.

In many skeletal diseases, including osteoporosis and disuse osteopenia, defective osteoblast differentiation is associated with increased marrow adipogenesis. The relative activity of two transcription factors, RUNX2 and PPARγ, controls whether a mesenchymal cell will differentiate into an osteoblast or adipocyte. Herein we show that the ERK/MAP kinase pathway, an important mediator of mechanical and hormonal signals in bone, stimulates osteoblastogenesis and inhibits adipogenesis via phosphorylation of RUNX2 and PPARγ. Induction of osteoblastogenesis in ST2 mesenchymal cells was associated with increased MAPK activity and RUNX2 phosphorylation. Under these conditions PPARγ phosphorylation also increased, but adipogenesis was inhibited. In contrast, during adipogenesis MAPK activity and phosphorylation of both transcription factors was reduced. RUNX2 phosphorylation and transcriptional activity were directly stimulated by MAPK, a response requiring phosphorylation at S301 and S319. MAPK also inhibited PPARγ-dependent transcription via S112 phosphorylation. Stimulation of MAPK increased osteoblastogenesis and inhibited adipogenesis, while dominant-negative suppression of activity had the opposite effect. In rescue experiments using Runx2(-/-) mouse embryo fibroblasts (MEFs), wild type or, to a greater extent, phosphomimetic mutant RUNX2 (S301E,S319E) stimulated osteoblastogenesis while suppressing adipogenesis. In contrast, a phosphorylation-deficient RUNX2 mutant (S301A,S319A) had reduced activity. Conversely, wild type or, to a greater extent, phosphorylation-resistant S112A mutant PPARγ strongly stimulated adipogenesis and inhibited osteoblastogenesis in Pparg(-/-) MEFs, while S112E mutant PPARγ was less active. Competition between RUNX2 and PPARγ was also observed at the transcriptional level. Together, these studies highlight the importance of MAP kinase signaling and RUNX2/PPARγ phosphorylation in the control of osteoblast and adipocyte lineages.

Abstract The loss of the tail is among the most notable anatomical changes to have occurred along the evolutionary lineage leading to humans and to the ‘anthropomorphous apes’ 1–3 , with a proposed role in contributing to human bipedalism 4–6 . Yet, the genetic mechanism that facilitated tail-loss evolution in hominoids remains unknown. Here we present evidence that an individual insertion of an Alu element in the genome of the hominoid ancestor may have contributed to tail-loss evolution. We demonstrate that this Alu element—inserted into an intron of the TBXT gene 7–9 —pairs with a neighbouring ancestral Alu element encoded in the reverse genomic orientation and leads to a hominoid-specific alternative splicing event. To study the effect of this splicing event, we generated multiple mouse models that express both full-length and exon-skipped isoforms of Tbxt , mimicking the expression pattern of its hominoid orthologue TBXT . Mice expressing both Tbxt isoforms exhibit a complete absence of the tail or a shortened tail depending on the relative abundance of Tbxt isoforms expressed at the embryonic tail bud. These results support the notion that the exon-skipped transcript is sufficient to induce a tail-loss phenotype. Moreover, mice expressing the exon-skipped Tbxt isoform develop neural tube defects, a condition that affects approximately 1 in 1,000 neonates in humans 10 . Thus, tail-loss evolution may have been associated with an adaptive cost of the potential for neural tube defects, which continue to affect human health today.