Identification of the homeobox protein Prx1 (MHox, Prrx-1) as a regulator of osterix expression and mediator of tumor necrosis factor α action in osteoblast differentiation - PubMed (original) (raw)

Identification of the homeobox protein Prx1 (MHox, Prrx-1) as a regulator of osterix expression and mediator of tumor necrosis factor α action in osteoblast differentiation

Xianghuai Lu et al. J Bone Miner Res. 2011 Jan.

Abstract

Tumor necrosis factor α (TNF-α) promotes bone loss and inhibits bone formation. Osterix (Osx, SP7) is a transcription factor required for osteoblast (OB) differentiation because deletion results in a cartilaginous skeleton. We previously described a TNF suppressor element in the Osx promoter that was used to isolate nuclear proteins mediating TNF inhibition of OB differentiation. Nuclear extracts from TNF-treated pre-OBs were incubated with the TNF suppressor element for protein pull-down, and tryptic fragments were analyzed by mass spectrometry. Chromatin immunoprecipitation (ChIP) assay confirmed eight bound transcription factors. One protein, the paired related homeobox protein (Prx1), had been shown previously to have a critical role in limb bud formation and skeletal patterning. PCR revealed Prx1 expression in primary stromal cells (MSCs), C3H10T1/2 cells, and MC3T3 preosteoblasts. TNF stimulated a 14-fold increase in mRNA for Prx1, rapid cell accumulation in MC3T3 cells, and expression in periosteal and trabecular lining cells in vivo. Transient expression of Prx inhibited transcription of Osx and RUNX2. Expression of the Prx1b isoform or Prx2 decreased Osx and RUNX2 mRNA and OB differentiation in preosteoblasts. Silencing of Prx1 with siRNA abrogated TNF suppression of Osx mRNA and increased basal Osx expression. Electrophoretic mobility shift revealed Prx1b as the preferred isoform binding the Osx promoter. These results identify the homeobox protein Prx1 as an obligate mediator of TNF inhibition of Osx and differentiation of OB progenitors. Activation of Prx1 by TNF may contribute to reduced bone formation in inflammatory arthritis, menopause, and aging.

© 2011 American Society for Bone and Mineral Research.

PubMed Disclaimer

Figures

Fig. 1

Fig. 1

Protein binding to the TNF-α repressor element of the Osx promoter is confirmed by ChIP assays. C3H10T1/2 cells treated with TNF-α (10 ng/mL) or control medium for 24 hours were used for ChIP assays. At the top is a map of the Osx promoter, including the TNF-α response element region. Primer sequences are listed in Table 1. (A) ChIP assays used antibodies against normal mouse IgG (as control), HCC1, HSP70, radixin, NF-κB p65, VDR, moesin, or TIF1. PCR reactions used the primer pair flanking the TNF-α response element region indicated on the map. Molecular-weight markers are on each end. (B) ChIP assays used antibodies against normal mouse IgG (as control), moesin, TIF1, or Prx1. Other controls were done to test the wash step and the effect of no antibody, as well as to verify the presence of the relevant sequence in the starting DNA sample (input). The primers used were the same as in panel A. (C) PCR reactions used the same DNA as in panels A and B and primers that were upstream or downstream of the TNF-α response element.

Fig. 2

Fig. 2

Prx expression is regulated by TNF-α. (A) Gene structures of the Prx1 isoforms and Prx2. (B–D) RNA was prepared from C3H10T1/2 cells treated with control medium or TNF-α (10 ng/mL) for 24 hours. Quantitative RT-PCR using primer pairs specific for Prx1a (B), Prx1b (C), or Prx2 (D) (Table 1) was performed.

Fig. 3

Fig. 3

TNF-α stimulates expression of Prx1 in vitro and in vivo. (A) MC3T3 cells treated with control medium or TNF-α (10 ng/mL) for the indicated times and labeled with fluorescein-conjugated anti-Prx1 antibody (FITC-Prx-1). Increased signal intensity is seen within 15 minutes and is sustained. Nuclei were counterstained with DAPI, and cells were examined by confocal microscopy with a ×40 oil-immersion objective. (B) As in panel A, showing a closer view of increased cytoplasmic and punctuate nuclear accumulation after TNF-α treatment from another experiment. (C) Immunostaining shows Prx detection in tibial periosteal lining cells from TgTNF arthritic mice but little or no signal in wild-type littermates. Shown are TgTNF, wild-type, control antibody, and hematoxylin and eosine (H&E)–stained sections from the proximal tibia. Bones were obtained from 8-week-old mice.

Fig. 4

Fig. 4

TNF-α inhibition of Osx mRNA is mediated by Prx1. C3H10T1/2 cells were transfected with control or Prx1 siRNA, followed by treatment with control medium or TNF-α (10 ng/mL) for 24 hours. RNA was prepared and analyzed by quantitative RT-PCR using Prx1-, Osx-, or Runx2-specific primer pairs, as shown in Table 1. (A) Prx1. (B) Osx. (C) Runx2.

Fig. 5

Fig. 5

Prx1 binds the Osx promoter. Nuclear extract was prepared from C3H10T1/2 cells treated with TNF-α (10 ng/mL) for 18 hours. EMSA was carried out using 32P-labeled dsDNA TNF-α response element or homeobox-binding site oligos, along with the indicated combinations of nuclear extract, antibodies, and cold probes. (A) EMSA showing labeled TNF-α response element–binding nuclear protein from control or TNF-α-treated C3H10T1/2 cells. The arrow on the left of the gel identifies the Prx1 band. TNF-α increased binding to the probe (control, +NE versus TNF, +NE). The binding is supershifted by a Prx1 antibody (αPrx#2) but not by a Prx2 antibody (αPrx2) or a commercial Prx1 antibody (αPrx1#1) and competed by 100X unlabeled probe (TRE) but not by 100X unlabeled control probe. (B) EMSA using a probe from a homeobox homologous site upstream of the TNF-α site that includes a contiguous RUNX2-binding site. The arrows on the left of the gel identify Prx1 bands. The binding is reduced or supershifted by two different Prx1 antibodies (αPrx1, #1 or #2) and competed by 100X cold probe. TNF-α treatment did not change the binding to this site.

Fig. 6

Fig. 6

Delineation of Prx isoforms binding the Osx promoter TNF-α response element. Recombinant Prx1a, Prx1b, and Prx2 proteins were synthesized by in vitro transcription/translation. Nuclear extract was prepared from C3H10T1/2 cells treated with TNF-α (10 ng/mL) for 24 hours. EMSA was carried out using 32P-labeled dsDNA TNF-α response element or homeobox-binding site oligos, along with the indicated combinations of recombinant proteins and nuclear extract. Controls were prepared without recombinant protein. (A) Recombinant Prx1b and Prx2, but not Prx1a, bind the TNF-α response element in the presence of nuclear extract. (B) Recombinant Prx1b fails to bind the TNF-α response element in the absence of nuclear extract. (C) Homeobox-binding site (upstream of the TNF-α element). Recombinant Prx1b and Prx2 bind the upstream homeobox element with or without nuclear extract. Arrows indicate the locations of protein-DNA complexes.

Fig. 7

Fig. 7

Osteoblast differentiation and Osx and RUNX2 expression are inhibited by Prx1b and Prx2 in preosteoblasts. MC3T3 cells were transiently transfected with Prx1a, Prx1b, or Prx2 expression vectors using the Amaxa nucleofection system. RNA was prepared 48 hours after transfection. (A) Quantitative RT-PCR using primers specific for Osx. (B) Quantitative RT-PCR using primers specific for RUNX2. (C) MC3T3 cells transiently transfected with Prx1a, Prx1b, or Prx2 expression vectors using the Amaxa nucleofection system. Transfected cells were cultured in mineralization medium for 14 days and then stained with alizarin red.

Fig. 8

Fig. 8

Prx1b and Prx2 inhibit osteoblast differentiation and Osx and RUNX2 expression in primary MSC cultures. Mouse primary MSC suspensions were prepared at 5 × 106 cells/mL in α-MEM and 10% FBS and transiently transfected with Prx1a, Prx1b, or Prx2 expression vectors using the Amaxa nucleofection system. Transfected cells were plated in 12-well plates (1 mL/well) and supplemented with 50 &g/mL of

l

-ascorbate and 5 mM β-GP. RNA was prepared after 48 hours, and quantitative RT-PCR was performed. (A) Osx. (B) RUNX2.

References

    1. Khosla S, Westendorf JJ, Oursler MJ. Building bone to reverse osteoporosis and repair fractures. J Clin Invest. 2008;118:421–428. -PMC -PubMed
    1. Marie PJ. Transcription factors controlling osteoblastogenesis. Arch Biochem Biophys. 2008;473:98–105. -PubMed
    1. Franceschi RT, Ge C, Xiao G, Roca H, Jiang D. Transcriptional regulation of osteoblasts. Ann NY Acad Sci. 2007;1116:196–207. -PubMed
    1. Lian JB, Stein GS, Javed A, et al. Networks and hubs for the transcriptional control of osteoblastogenesis. Rev Endocrinol Metab Disord. 2006;7:1–16. -PubMed
    1. Komori T. Regulation of osteoblast differentiation by transcription factors. J Cell Biochem. 2006;99:1233–1239. -PubMed

Publication types

MeSH terms

Substances

LinkOut - more resources