Dickkopf-1 negatively regulates the expression of osteoprotegerin, a key osteoclastogenesis inhibitor, by sequestering Lrp6 in primary and metastatic lytic bone lesions

Abstract

Recently, an inverse role for Wnt signaling in the development of osteoclasts in the bone was demonstrated. In the present study, we examined whether there is a commonality in the mechanism of bone resorption and lysis that occur in a diverse set of bone metastatic lesions, as well as in primary bone lesions. Compared with control bone tissue and bone biopsies from patients with nonmetastatic primary tumors (i.e., breast carcinoma, lung adenocarcinoma, and prostate carcinoma), patients with bone metastatic lesions from the three aforementioned primary tumors, as well as osteolytic lesions obtained from the bone biopsies of patients with multiple myeloma, demonstrated an upregulated expression of the glycoprotein Dickkopf-1 at both the mRNA and protein levels. Additionally, by coimmunoprecipitation, Dickkopf-1 pulled-down low-density lipoprotein receptor-related protein 6 (Lrp6), which is a key downstream effector of the Wnt signaling pathway. The expression of Lrp6 was unaltered in the osteometastatic lesions. This negative regulation was associated with a lowered expression of osteoprotegerin in the osteometastatic lesions, an observation that was previously reported to promote osteoclastogenesis. These findings provide a common mechanism for the inverse relationship between the Wnt signaling pathway and the development of primary or metastatic bone lesions. Pharmacological modulation of the Wnt signaling pathway might benefit the clinical management of primary and metastatic bone lesions.

Figure 1

Histogram depicting alteration of the mRNA transcripts of Dickkopf-1 in the metastatic bone lesions and multiple myeloma. The normalized mRNA expression was estimated from the different transcripts from the standard curves. The primary tumors had no evidence of radiologic lesions but were biopsied because of persistent complaints of bone pain. The differences in the means between the groups were statistically significant when compared by analyses of variance. mRNA, messenger ribonucleic acid, #Metastatic bone lesion versus control (P <0.01); ∗metastatic bone lesion versus its corresponding primary tumor (P <0.01).

Figure 2

Western blot showing elevated levels of Dickkopf-1 in the osteometastatic lesions compared with the controls A representative blot is shown here. Triplicate samples were analyzed to obtain the quantitative intensities of the protein signals for Dickkopf-1 and β-actin (the loading control). These means between the groups were different and were statistically significant when examined by analyses of variance (∗P <0.01).

Figure 3

(A and B) Coimmunoprecipitation of the pulled-down Dickkopf-1–Lrp6 immunoblots with either antibody shows the binding. (C) Unaltered levels of Lrp6 in the osteometastatic lesions compared with the controls. All of the panels are representative Western blots. (A and B) In the far left and middle panel, the immunoprecipitated lysates were blotted for Dickkopfl-1 and Lrp6, respectively. GAPDH was used to demonstrate equal loading of the loading lanes. The histograms below each blot are the quantitative intensities of the protein signals. (C) The far right panel shows the expression of Lrp6 in the control, primary, and metastatic lysates. β-actin was used as a loading control. The histogram below represents the quantified intensities of the protein signals. All of the assays were performed in triplicate from pooled samples. GAPDH = glyceraldehyde-3-phosphate dehydrogenase.

Figure 4

Western blots depicting diminished levels of osteoprotegerin in the osteometastatic lesions are shown. A representative blot is shown here. Triplicate samples were analyzed to obtain the quantitative intensities of the protein signals for osteoprotegerin and β-actin (the loading control). The differences in the means between the groups were statistically significant when compared by analyses of variance. #Metastatic bone lesion versus control (P <0.01); ∗metastatic bone lesion versus its corresponding primary tumor (P <0.01).