Fig. 2. PHGDH in osteoclasts controls bone resorption and postnatal bone mass.

a–c, Three-dimensional (3D) micro-CT models (a) of the tibial metaphysis (top) and diaphysis (bottom) with quantification of trabecular bone volume (BV/TV; b) and cortical thickness (Ct.Th; c) in wild-type (Phgdh^oc+) and osteoclast-specific PHGDH knockout (Phgdh^oc−) mice (n = 9 Phgdh^oc+ and 7 Phgdh^oc− mice; unpaired, two-tailed Student’s t-test). d,e, TRAP staining of the tibial metaphysis (d) with quantification (e) of osteoclast surface per bone surface (Oc.S/B.S) in Phgdh^oc+ and Phgdh^oc− mice (n = 9 Phgdh^oc+ and 7 Phgdh^oc− mice; unpaired, two-tailed Student’s t-test). Scale bar, 100 µm. f, Serum CTx-I levels in Phgdh^oc+ and Phgdh^oc− mice (n = 9 Phgdh^oc+ and 7 Phgdh^oc− mice; unpaired, two-tailed Student’s t-test). g, Schematic overview of the experimental setup used to study osteoclast differentiation. h, TRAP staining of osteoclast progenitors (upper) and osteoclasts (lower) from Phgdh^oc+ and Phgdh^oc− mice treated with M-CSF alone or M-CSF and RANKL, respectively (n = 4). Scale bar, 50 µm. i–l, Number of BMMCs (i), total number of TRAP-positive (TRAP⁺) MNCs (j), number of TRAP⁺ MNCs with indication of number of nuclei per cell (k) and average size of TRAP⁺ MNCs (l) from Phgdh^oc+ and Phgdh^oc− mice after M-CSF and RANKL treatment (n = 4; all unpaired, two-tailed Student’s t-test). m, Phgdh (P < 0.0001), Csf1r, Tnfrsf11a (encoding RANK), Nfatc1 (P = 0.002), Myc, Calcr (P = 0.0002), Acp5 (P = 0.008), Ctsk (P = 0.0002), Dcstamp (P = 0.0016), Atp6v0d2 (P = 0.007), Mmp9 (P = 0.010) and Irf8 mRNA levels in cultured osteoclasts from Phgdh^oc+ and Phgdh^oc− mice after M-CSF and RANKL treatment (n = 4; unpaired, two-tailed Student’s t-test with *P < 0.05 versus Phgdh^oc+). Data are means ± s.d.

Source data