Functions of vasopressin and oxytocin in bone mass regulation

Abstract

Prior studies show that oxytocin (Oxt) and vasopressin (Avp) have opposing actions on the skeleton exerted through high-affinity G protein-coupled receptors. We explored whether Avp and Oxtr can share their receptors in the regulation of bone formation by osteoblasts. We show that the Avp receptor 1α (Avpr1α) and the Oxt receptor (Oxtr) have opposing effects on bone mass: Oxtr(-/-) mice have osteopenia, and Avpr1α(-/-) mice display a high bone mass phenotype. More notably, this high bone mass phenotype is reversed by the deletion of Oxtr in Oxtr(-/-):Avpr1α(-/-) double-mutant mice. However, although Oxtr is not indispensable for Avp action in inhibiting osteoblastogenesis and gene expression, Avp-stimulated gene expression is inhibited when the Oxtr is deleted in Avpr1α(-/-) cells. In contrast, Oxt does not interact with Avprs in vivo in a model of lactation-induced bone loss in which Oxt levels are high. Immunofluorescence microscopy of isolated nucleoplasts and Western blotting and MALDI-TOF of nuclear extracts show that Avp triggers Avpr1α localization to the nucleus. Finally, a specific Avpr2 inhibitor, tolvaptan, does not affect bone formation or bone mass, suggesting that Avpr2, which primarily functions in the kidney, does not have a significant role in bone remodeling.

Keywords: osteoblast; osteoporosis; skeleton.

Fig. 1.

Avp interacts with both osteoblast Avpr1α and Oxtr in bone mass regulation. (A) Representative wells showing the effect of Avp (10⁻⁸ M) on alkaline phosphatase-positive colonies 1 or 2 wk after the induction of differentiation in bone marrow stromal cell cultures isolated from Oxtr^−/− or Oxtr^+/+ mice. (B) mRNA expression (by qPCR) of osteoblast genes, namely Alp, Runx2, Atf4, and osterix (Sp7), in differentiating bone marrow stromal cell cultures from Oxtr^−/− mice following a 6-h exposure to Avp. *P < 0.05, **P < 0.01, triplicate. (C and D) Effect of Avp on mRNA expression (qPCR) of the osteoblast genes osteocalcin (Bglap), bone sialoprotein (Ibsp), Runx2, and Sp7 (comparison with wild type at each time point, *P < 0.05, **P < 0.01, triplicate) (C) or Runx2 protein expression (Western blot) (D) in differentiating (10 d) bone marrow stromal cell cultures isolated from Avpr1α^−/−, Oxtr^−/−:Avpr1α ^−/−, or wild-type mice. (E) Histomorphometry of spinal trabecular bone from Avpr1α^−/−, Oxtr^−/−:Avpr1α ^−/−, or wild-type mice, expressed as fractional bone volume (BV/TV), together with representative images. *P < 0.05, **P < 0.01 compared with respective wild-type littermates, as shown. Results are shown as mean ± SEM. Statistics by unpaired Student’s t-test, comparison with 0-dose control. n = 5 mice per group.

Fig. 2.

Avp triggers nuclear localization of the Avpr1α. (A) Primary bone marrow stromal cells from Avpr1α^+/+ or Avpr1α^−/− mice were stimulated with Avp (10⁻⁸ M, 30 min) and stained with anti-Avpr1α antibody (red). Green indicates phalloidin staining. (Top) Midsectional confocal microscopy shows that at 30 min there is complete nuclear localization of the Avpr1α in Avpr1α^+/+ cells, but this localization is absent in Avpr1α^−/− cells. (Lower) The outer nuclear membrane was removed from isolated nuclei to produce nucleoplasts (19). Nuclei were stained with anti-Avpr1α antibody (red) and counterstained with anti-Kpnb1 antibody (green). Avpr1αs were visualized intranuclearly (red) at 30 min after Avp stimulation. Avpr1α^−/− cells expectedly showed no staining with the Avpr1α antibody. (B) Primary osteoblasts were treated with Avp (10⁻⁸ M) for 15 or 30 min. Western blots show Avpr1α in the cytosolic and nuclear fractions. Markers included β-actin (cytosol) and lamin B1 (nuclear). Control (Ctl): whole-cell lysates. (C) Proteins in the immunoprecipitate from MC3T3.E1 preosteoblasts (with anti-Avpr1α antibody) were identified using MS-Fit software (prospector.ucsf.edu/prospector/cgi-bin/msform.cgi?form=msfitstandard), and all other proteins were recognized as nuclear proteins. Analysis of the spectra (FindPept database) revealed three peptides corresponding to Avpr1α intracellular loops in the in-gel band (45 kDa) at mass-to-charge ratios (m/z) of 1202.55 [(K)/GLLVTPCVSSVK/(S), residues 274–285], 1258.54 [(K)/FAKDDSDSMSR/(R), residues 378–388], and 2184.92 [(R)/RQTSYSNNRSPTNSTGTWK/(D), residues 389–407]. (D) Western immunoblot showing the effect of Avp (10⁻⁸ M) on the phosphorylation of Erk (pErk) in whole-cell lysates from bone marrow stromal cell cultures obtained from Avpr1α^−/− or Oxtr^−/−:Avpr1α ^−/− or wild-type mice. Total Erk (tErk) is shown.

Fig. 3.

Avpr2 inhibition does not affect bone formation or bone mass. (A) Effect of a specific Avpr2 inhibitor, tolvaptan (20 mg⋅kg⁻¹⋅d⁻¹), administered i.p. for 4 wk to sham-operated (Sham) or ovariectomized (OVX) mice, on BMD (PIXImus, Lunar-GE), expressed as total bone, spine (L4–L6), left femur (L-Femur), and left tibia (L-Tibia) BMD. (B) Fractional bone volume (BV/TV) assessed from von Kossa-stained sections of spinal trabecular bone (representative images are shown). (C, Lower) Bone formation parameters, namely mineralizing surface (MS), mineral apposition rate (MAR), and bone formation rate (BFR) following dual calcein (green) and xylenol orange (red) labeling. (Upper) Representative images are shown. Statistics: unpaired Student’s t test, corrected for multiple comparisons by Bonferroni; vehicle versus tolvaptan in Sham and OVXgroups yielded P > 0.1; n = 5 mice per group.

Fig. 4.

Selective Oxtr deletion in osteoblasts attenuates lactation-induced bone loss, excluding an action of Oxt on the Avpr1α. Sequential areal BMD (aBMD) measurements (PIXImus, Lunar-GE), quantitated as changes (Δ) from prepregnancy BMD (0) in total bone, spine (L4–L6 and >L6), left and right femur (L- and R-Femur), and left and right tibia (L- and R-Tibia) at different time points during pregnancy week 2 (P2), lactation weeks 2 and 3 (L2 and L3) and weaning weeks 1 and 3 (SL1 and SL3) in Col2.3Cre⁺/Oxtr^fl/fl (n = 4) or Col2.3Cre⁻/Oxtr^fl/fl mice (n = 6 mice). Statistics: unpaired Student’s t test; *P ≤ 0.05, **P ≤ 0.01.

Fig. S1.

Oxt injection reduces serum Avp levels. Two-month-old wild-type mice were injected twice with Oxt or Avp (4 µg/mL, 12 h apart) and were killed 12 h after the second injection. Mice injected with Oxt had reduced serum levels of Avp, whereas those injected with Avp had reduced serum levels of Oxt compared with vehicle-injected controls.