Fig 2: Expression levels of CIITA in osteocytes are associated with the status of bone lesions in myeloma patients.Bone marrow biopsies were randomly collected from myeloma patients (n = 15) and then stained with the anti-CIITA antibody. The immunohistochemically stained slides were semi-quantified by H-score, which reflects the totality of percentage of stained cells in each intensity. a Correlation between the H-scores of CIITA+ osteocytes and the numbers of bone lesions in myeloma patients. b Representative images of immunohistochemical staining show the expression of CIITA in osteocytes in the biopsy segments of the two patients highlighted with red or blue in (a). Scale bar, 50 µm. c–e Bone marrow aspirates of myeloma patients were subjected to ELISA. Shown are the correlation coefficients of the H-scores of CIITA+ osteocytes with bone marrow serum levels of RANKL (c, left) or sclerostin (c, right), the correlation coefficients of the levels of TP mRNA in myeloma cells with bone marrow serum levels of RANKL (d, left) or sclerostin (d, right), and the correlation coefficients of the H-scores of CIITA+ osteocytes with the levels of TP mRNA in myeloma cells (e). The correlations were evaluated using Pearson coefficient with two-tailed P value. r, correlation coefficient. f Depiction of osteocyte CIITA–mediated bone destruction in multiple myeloma. Myeloma TP/2DDR upregulates CIITA expression in osteocytes through the STAT1/IRF1 signaling pathway. The increased CIITA via AP2a as a link binds to the promoter of Tnfsf11 or Sost and activates histone acetylation at H3K14, leading to the expression of RANKL or sclerostin in osteocytes.

Fig 3: Deficiency of CIITA expression in osteocytes impairs myeloma-induced bone lesions.a Western blots show the expression of CIITA in osteocytes, MSCs, and osteoblasts isolated from wild-type (WT) or osteocyte-specific Ciita-knockout (Ciita KO) mice. The expression of GAPDH served as protein loading controls. Images shown are representative of two independent experiments. b–k WT and osteocyte-specific Ciita KO mice were intrafemorally injected with the murine myeloma cell line Vk12598 (1 × 106 cells/mouse). The mice not receiving myeloma cells (No MM) served as controls. Shown are the concentrations of M-proteins in mouse sera (b), representative µ-CT images of mouse femurs (c), and percentages of BV/TV (d), Tb.Th (e), Tb.N (f), Tb.Sp (g), OS/BS (h), Ob.S/BS (i), ES/BS (j), and Oc.S/BS (k). l BFR/BS was measured by calcein injection, and the bone sections were imaged and analyzed. Shown are representative images and summarized data of bone formation in mouse femurs. Scale bar: 20 µm. m-p Osteocytes isolated from WT or Ciita-KO mice injected with or without myeloma cells, were cultured for generating the osteocyte CM. The osteoblast precursor MC3T3-E1 and the osteoclast precursor Raw264.7 were cultured with the osteocyte CM. After culturing, the cells were stained with Alizarin red-S for osteoblast differentiation or TRAP staining for osteoclast differentiation. Shown are the summarized data from Alizarin red-S staining (m) and the numbers of multinuclear (=3) TRAP+ cells (n, left). Real-time PCR and ELISA assays show the relative expression of Tnfsf11 and Sost mRNA in osteocytes (o) and the serum levels of RANKL or sclerostin in mice (p). (n, right) Primary osteocytes isolated from mouse femurs were knocked down by Ciita siRNA (siCiita). The CM of WT or siCiita was cultured with myeloma cells for 48 hours. Data are mean ± SD (n = 5 mice/group). P values were determined using one-way ANOVA with Tukey’s multiple comparisons test.

Fig 4: Depletion of CIITA in osteocytes abrogates effects of 2DDR on cytokine secretion, osteoclastogenesis, and suppressed osteoblastogenesis.MLO-Y4 and MLO-A5 osteocytes carried with non-targeted control shRNAs (shCtrl) or Ciita shRNAs (shCiita) were cultured with or without 2DDR for 2 days, given fresh medium, and cultured for another 2 days, and then the cells and the CM were collected. The osteoblast precursor MC3T3-E1 and the osteoclast precursors Raw264.7 cells or primary monocytes were cultured with or without the osteocyte CM. After culturing, the cells were subjected to Alizarin red-S staining to assess osteoblast differentiation, to TRAP staining to assess osteoclast differentiation, or to real-time PCR to assess the differentiation-associated gene expression. a–f Shown are summarized data for Sost mRNAs (a), Alizarin red-S staining (b), the relative expression of osteoblast differentiation–associated genes Alp, Col1a1, and Bglap (c), Tnfsf11 mRNA (d), the numbers of multinuclear (=3) TRAP+ cells (e), and the relative expression of osteoclast differentiation–associated genes Trap, Calcr, and Ctsk (f). Data are mean ± SD (n = 3 biological replicates). g–j Osteocytes isolated from WT or osteocyte-specific Ciita-KO mice (n = 5 mice/group) were cultured with or without 1 mM 2DDR for 2 days for collecting the cells and the osteocyte CM. MC3T3-E1 or Raw264.7 cells cultured with osteocyte CM were subjected to Alizarin red-S staining, TRAP staining, or real-time PCR. Shown are summarized data for Sost mRNA levels in osteocytes (g), Alizarin red-S staining (h), Tnfsf11 mRNA levels in osteocytes (i), and numbers of multinuclear (=3) TRAP+ cells (j, left). Box plots (g, i) indicate median, 25th and 75th percentiles, and minima and maxima of the distribution. Data are presented as mean ± SD. (j, right) Primary osteocytes wild-type or carried with Ciita siRNAs (siCiita) were cultured in the same fashion as in (a) to collect the CM. The numbers of multinuclear (=3) TRAP+ cells were counted in the cultures of primary monocytes with or without the osteocyte CM. Data are means ± SD. P values were determined using one-way ANOVA with Tukey’s multiple comparisons test.

Fig 5: Myeloma TP/2DDR enhances CIITA expression in osteocytes via the STAT1/IRF1 signaling pathway.a–b The levels of CIITA mRNA (a, n = 3 biological replicates) or protein (b) in MLO-Y4 or MLO-A5 osteocytes co-cultured with the high TP–expressing ARP-1 cells carrying non-targeted control shRNAs (shCtrl) or TP shRNAs (shTP) or co-cultured with low TP–expressing MM.1 S cells carrying empty control vector (Vec) or TP cDNAs (TP). c–e Levels of CIITA mRNA (c, n = 3 biological replicates) and protein (d) as well as phosphorylated (p) or non-phosphorylated ERK1/2, Akt, Syk, STAT1, and JAK1 (e) in MLO-Y4 and MLO-A5 cells cultured with 2DDR. (GAPDH: protein loading controls). f Levels of the p-STAT1 or STAT1 in primary osteocytes cultured with the CM of shCtrl or shTP ARP-1 cells or with the CM of Vec or TP MM.1 S cells. g–h Levels of p-SyK, Syk, p-STAT1, or STAT1 (g), Ciita, Tfsf11, and Sost mRNAs (h, n = 3 biological replicates) in MLO-Y4 or MLO-A5 cells cultured with 2DDR, IgG control, or anti-av antibody (Ab). i Levels of p-STAT1 in whole-cell lysates or cytosolic or nuclear fractions of MLO-Y4 and MLO-A5 cells cultured without or with 2DDR. Loading controls: GAPDH for cytosol fractions (Cyt); H3 for nuclear fractions (Nuc). WCL, whole cell lysate. j The enrichment of p-STAT1 in the promoter of Irf1 gene in MLO-Y4 and MLO-A5 cells cultured with 2DDR. IgG levels served as controls. k The protein levels of IRF1 in MLO-Y4 and MLO-A5 cells cultured with 2DDR. l Co-immunoprecipitation of p-STAT1 with IRF1 in MLO-Y4 and MLO-A5 cells cultured with 2DDR. m The enrichment of p-STAT1 in the promoter of Ciita gene in MLO-Y4 and MLO-A5 cells cultured with 2DDR (n = 3 biological replicates). n–o The relative expression of Ciita (n), Tnfsf11 and Sost (o) mRNA in MLO-Y4 and MLO-A5 cells cultured with 2DDR in the absence or presence of STAT1 inhibitor fludarabine (Flu; 10 µM) (n = 3 biological replicates). b, d, e–g, i, k, and l are representative of two independent experiments. Data are mean ± SD. P values were determined using one-way ANOVA with Tukey’s multiple comparisons test.