Molecular mechanism of hyperactive tooth root formation in oculo-facio-cardio-dental syndrome

Abstract

Mutations in the B-cell lymphoma 6 (BCL6) interacting corepressor (BCOR) cause oculo-facio-cardio-dental (OFCD) syndrome, a rare X-linked dominant condition that includes dental radiculomegaly among other characteristics. BCOR regulates downstream genes via BCL6 as a transcriptional corepressor. However, the molecular mechanism underlying the occurrence of radiculomegaly is still unknown. Thus, this study was aimed at identifying BCOR-regulated genetic pathways in radiculomegaly. The microarray profile of affected tissues revealed that the gene-specific transcriptional factors group, wherein nucleus factor 1B, distal-less homeobox 5, and zinc finger protein multitype 2 (ZFPM2) were the most upregulated, was significantly expressed in periodontal ligament (PDL) cells of the diseased patient with a frameshift mutation (c.3668delC) in BCOR. Wild-type BCOR overexpression in human periodontal ligament fibroblasts cells significantly hampered cellular proliferation and ZFPM2 mRNA downregulation. Promoter binding assays showed that wild-type BCOR was recruited in the BCL6 binding of the ZFPM2 promoter region after immunoprecipitation, while mutant BCOR, which was the same genotype as of our patient, failed to recruit these promoter regions. Knockdown of ZFPM2 expression in mutant PDL cells significantly reduced cellular proliferation as well as mRNA expression of alkaline phosphatase, an important marker of odontoblasts and cementoblasts. Collectively, our findings suggest that BCOR mutation-induced ZFPM2 regulation via BCL6 possibly contributes to hyperactive root formation in OFCD syndrome. Clinical data from patients with rare genetic diseases may aid in furthering the understanding of the mechanism controlling the final root length.

Keywords: BCL6; BCOR; OFCD syndrome; ZFPM2; radiculomegaly.

FIGURE 1

Microarray gene profiling and in vitro assay of HPdLF cells. (A) qRT-PCR validation of microarray analysis for upregulating genes with a fold change greater than 15. Expression levels of ZFPM2, NFIB, and DLX5 relative to GAPDH expression were determined. Data are expressed as the means ± standard deviations; t-test, ****p < 0.0001, ***p < 0.001. (B) Overexpression of WT BCOR in HPdLF cells in vitro. HPdLF cells stably expressing WT BCOR were confirmed using an immunofluorescence assay. HPdLF cells were immunostained with anti-FLAG antibody and FITC-labeled secondary antibodies. Cell nuclei were counterstained with DAPI. An empty plasmid was used as a negative control. Samples were analyzed and photographed using an epifluorescence microscope. Scale bars represent 100 μm. (C) The MTT assay revealed that WT BCOR transfection suppressed HPdLF cell proliferation compared to transfection with empty plasmids. Points and bars indicate means and standard deviations, respectively, of the triplicate values. ANOVA, *p < 0.05. (D) qRT-PCR analysis of NFIB, DLX5, and ZFPM2 expression after WT BCOR overexpression. Data are expressed as means ± standard deviations; t-test, ***p < 0.001; ns, not significant.

FIGURE 2

Identifying BCOR downstream genes. (A) BCL6 consensus logo. (B) Prediction of BCL6 binding sites in the ZFPM2 promoter region. Three BCL6 binding sites within 2 kb of exon 1 of ZFPM2 were identified using LASAGNA-Search 2.0. TSS represents transcription start site. (C,D) ChIP assay for the occupation of BCL6 on the promoter of ZFPM2 in COS-7 and HPdLF cells. IgG, n = 3; empty plasmid/WT BCOR, n = 3. Data are expressed as the means ± standard deviations from triplicate experiments. ANOVA was used to assess differences in fold change within each group. *p < 0.05, ****p < 0.0001; ns, not significant. (E) Validation of BCL6 suppression in COS-7 cells using qRT-PCR. t-test, ****p < 0.0001. (F) BCL6 knockdown co-transfected with WT BCOR failed to recruit promoter binding sites in COS-7 cells. Data are expressed as the means ± standard deviations from triplicate experiments. ANOVA, ****p < 0.0001; ns, not significant.

FIGURE 3

Mut BCOR failed to bind the BCL6 promoter sites on ZFPM2. (A) The Mut BCOR plasmid (c.3668delC) have a nuclear localization signal (NLS1 and NLS2), which is the same genotype as that of our patient from our previous research. (B) Expression of ZFPM2 mRNA after WT or Mut BCOR overexpression in HPdLF cells. WT BCOR repressed ZFPM2 expression, but Mut BCOR did not. t-test, ***p < 0.001; ns, not significant. (C) ChIP assay for the occupation of BCL6 on the promoter of ZFPM2 in Mut BCOR-overexpressed COS-7 cells and HPdLF cells. Mut BCOR failed to bind BCL6 binding sites 1, 2, and 3 in the ZFPM2 promoter region in COS-7 and HPdLF cells. ANOVA, ns = not significant.

FIGURE 4

ZFPM2 regulates ALP expression in Osx and miRNA in the tooth development pathway. (A) Osx and miRNA in the tooth development pathway. Compared with that in the healthy subject, RUNX2, KLF4, NOTCH3, NOTCH4, and ALP were considerably elevated in the patient with the OFCD syndrome. A) Osx control of Dspp demonstrates its significance in dentinogenesis. B) Osx controls cementoblast differentiation via the Wnt-β-catenin pathway. C) Role of miRNAs in fine-tuning tooth development. The yellow boxes indicate the elevated signals above a fold change of 2 in the microarray analysis. (B) The expression of differentially expressed genes was determined using qRT-PCR. ANOVA, ****p < 0.0001. (C) qRT-PCR validation of ZFPM2 suppression in Mut PDL cells. t-test, *p < 0.05. (D) MTT assays showed that Mut PDL cell proliferation was inhibited after ZFPM2 knockdown. Points and bars indicate the means and standard deviations of the triplicate values. ANOVA, *p < 0.05. (E) After ZFPM2 inhibition, mRNA ALP expression was significantly reduced in Mut PDL cells whereas the mRNA expression of RUNX2, KLF4, NOTCH3, and NOTCH4 did not change. ANOVA, ****p < 0.0001.