Microphthalmia-associated transcription factor interactions with 14-3-3 modulate differentiation of committed myeloid precursors

Abstract

The microphthalmia-associated transcription factor (MITF) is required for terminal osteoclast differentiation and is a target for signaling pathways engaged by colony stimulating factor (CSF)-1 and receptor-activator of nuclear factor-kappaB ligand (RANKL). Work presented here demonstrates that MITF can shuttle from cytoplasm to nucleus dependent upon RANKL/CSF-1 action. 14-3-3 was identified as a binding partner of MITF in osteoclast precursors, and overexpression of 14-3-3 in a transgenic model resulted in increased cytosolic localization of MITF and decreased expression of MITF target genes. MITF/14-3-3 interaction was phosphorylation dependent, and Ser173 residue, within the minimal interaction region of amino acid residues 141-191, was required. The Cdc25C-associated kinase (C-TAK)1 interacted with an overlapping region of MITF. C-TAK1 increased MITF/14-3-3 complex formation and thus promoted cytoplasmic localization of MITF. C-TAK1 interaction was disrupted by RANKL/CSF-1 treatment. The results indicate that 14-3-3 regulates MITF activity by promoting the cytosolic localization of MITF in the absence of signals required for osteoclast differentiation. This work identifies a mechanism that regulates MITF activity in monocytic precursors that are capable of undergoing different terminal differentiation programs, and it provides a mechanism that allows committed precursors to rapidly respond to signals in the bone microenvironment to promote specifically osteoclast differentiation.

Figure 1.

MITF binds to 14-3-3 in phosphodependent manner. (A) anti-MITF immunoprecipitates prepared from RAW264 C4 cells or COS-7 cells were analyzed by immunoblot. The MITF doublet is indicated by arrows. (B) GST-fusion full length MITF proteins expressed in bacteria were incubated with whole cell extract from COS-7 cells transfected with Myc-14-3-3ζ. MITF was pulled down (GST PD), and complexes were analyzed by immunoblot with anti-Myc antibody. (C) FLAG-MITF and GST-14-3-3ζ, -β, -ε, -η, and -γ were coexpressed in COS-7; 14-3-3s were pulled down; and complexes were analyzed by immunoblot with anti-FLAG antibody. (D) FLAG-MITF expression vector and either GST-14-3-3 (wt) or dimerization-deficient GST-14-3-3 (dm) were coexpressed in COS-7 cells, 14-3-3 was pulled down, and complexes were analyzed by immunoblot. In each panel, whole cell lysates were analyzed by immunoblot with the appropriate antibodies, as indicated (input controls).

Figure 2.

Mapping of the 14-3-3 binding domain in MITF. (A and B) COS-7 cells were transiently transfected with FLAG-MITF or various FLAG-MITF truncation mutations (A) or short targeted deletion mutations (B) and GST-14-3-3 (or GST), as indicated. 14-3-3 was pulled down (GST PD), and complexes were analyzed by Western with anti-FLAG antibody. Whole cell lysates were also immunoblotted with anti-FLAG or anti-GST antibody (input controls).

Figure 3.

Identification of S173 as a 14-3-3 binding site. In each panel, FLAG- and GST-tagged proteins, as indicated, were expressed in COS-7 cells, and complexes from whole cell lysates were pulled down (GST PD), and analyzed by Western analysis with the antibody indicated. Whole cell lysate was also immunoblotted with anti-FLAG or anti-GST antibody (input controls). The experiments in all panels were all preformed a minimum of three times, and representative results are presented. (A) FLAG-MITF and GST-14-3-3 were coexpressed. Thirty-six hours after transfection, cells were left untreated or treated with 100 nM CalA for 1 h, or the whole cell lysate was treated with 40 U/μl CIAP. (B) FLAG-MITF wild type or versions with serine or threonine point mutations (S100A, S173A, T186A, and S409A) were coexpressed with GST-14-3-3ζ, or GST alone, as indicated. (C) MITF expression vector or FLAG-MITF S173A was coexpressed with GST-14-3-3 (or GST). After 36 h, cells were either not treated or treated with 100 nM CalA for 1 h. (D) Lysates from COS-7 cells transfected with GST-14-3-3 treated or not with CalA were immobilized onto glutathione-Sepharose beads and then incubated with lysate from COS-7 cells independently transfected with FLAG-MITF, also treated or not with CalA, as indicated. (E) GST-MITF and either Myc-14-3-3 or Myc 14-3-3 K49E were coexpressed. Endogenous 14-3-3 band is marked by asterisk.

Figure 4.

Dynamics of MITF localization and MITF-14-3-3 interaction during differentiation of primary osteoclast-like cells. (A and B) BMMs were grown without CSF-1 for 6 h (time 0) and then cultured with RANKL/CSF-1 for 12 or 24 h. (A) Anti-MITF was used for immunolocalization. Representative images are shown. (B) Quantitative analysis of MITF localization. Each experiment was repeated three times, and a minimum of 100 cells were counted for each condition and time point. Error bars indicate SD of the measurements. (C) BMMs were treated with RANKL/CSF-1 for times indicated (in hours). At each time point, anti-MITF IP complex was prepared and analyzed by western. Whole cell lysates were also immunoblotted with anti-MITF or anti-14-3-3 antibody.

Figure 5.

Overexpression of 14-3-3 regulates MITF localization and activity. (A and B) BMMs were grown without CSF-1 for 6 h (time 0) and then cultured with RANKL/CSF-1 for times indicated (in hours). (A) Cytosolic (C) and nuclear (N) fractions of BMM cells were by analyzed by Western with antibodies as indicated. These experiments were repeated twice on cells derived from two different sets of mice. Endogenous 14-3-3 band is marked by asterisk. (B) Results of qRT-PCR for Mitf, TRAP or Cathepsin K (Cat A), as indicated. Results of two experiments, each performed in triplicate are presented. Results were normalized to day 0. Error bars indicate SD of the measurements. (C and D) RAW264C4 cells were transiently cotransfected with GFP-MITF and GST-14-3-3 or dimerization-deficient GST-14-3-3 (dm) or pEBG-GST vector, as indicated. Cells were left untreated or treated with RANKL/CSF-1 for 3 h. (C) Representative photomicrograph images. (D) Quantitative analysis of GFP-MITF localization. Each experiment was repeated four times, and a minimum of 100 cells were counted for each condition and time point. Error bars indicate SD of the measurements. (E) RAW C4 cells were cotransfected with 2 μg of TRAP luciferase reporter construct and 1 μg of wild-type MITF or S173A MITF, and GST-14-3-3 or dimerization-deficient GST-14-3-3 (dm) (1 or 2 μg* as indicated) or pEBG-GST empty vector (up to 5 μg total of DNA). TRAP reporter activity is presented as relative luciferase units (RLU), and the results of four experiments each performed in duplicate are presented. Error bars indicate SD of the measurements.

Figure 6.

MITF interaction with C-TAK1 enhances formation of MITF/14-3-3 complex. (A) GST-MITF and C-TAK1 wt (+) or C-TAK1 kinase dead (KD) were coexpressed in COS-7 cells. GST-MITF was pulled down (GST PD), and complexes were analyzed by immunoblot. (B) FLAG-MITF or FLAG-MITF S173A, and C-TAK1 wt (+) or C-TAK1 kinase dead (KD) were coexpressed in COS-7. MITF was immunoprecipitated with anti FLAG antibody, and the IP complex was analyzed by immunoblot. (C) FLAG-MITF expression vector or various FLAG-MITF truncation and point mutations and C-TAK1 were coexpressed in COS-7 cells, as indicated. MITF was immunoprecipitated with anti-FLAG antibody, and the IP complex was analyzed by Western. Whole cell lysates were also immunoblotted with anti-FLAG and anti C-TAK-1 antibody (input control). (D) FLAG-MITF and GST-C-TAK were coexpressed in RAW264.7 C4 cells, either treated or not treated with CSF1/RANKL for 3 h, as indicated. GST-C-TAK1 was pulled down, and complexes were analyzed by immunoblot with anti-FLAG antibody. In the panels, whole cell lysates were also immunoblotted with anti-FLAG, anti C-TAK-1, or anti-14-3-3 antibody, as indicated (input controls).

Figure 7.

14-3-3 binds to MITF in a localization- and differentiation-dependent manner. RAW C4 cells stably expressing FLAG-MITF (C4-MITF+) were left untreated or treated with RANKL/CSF-1 for the indicated times. (A) Immunolocalization of FLAG-MITF was monitored by anti-FLAG antibody. Representative photomicrograph images are presented. (B) Quantitative analysis of MITF localization. (C) Extent of differentiation, as measured by appearance of multinuclear cells. For B and C, experiments were performed three times each on three independently derived C4MITF+ cell clones, and a minimum of 100 cells were scored for each condition. Error bars indicate SD of the measurements. (D) Cytosolic and nuclear fractions of C4-MITF+ cells were prepared and anti-MITF complexes isolated and analyzed by immunoblot using anti-14-3-3 or anti C-TAK1 antibody. Nuclear and cytosolic fractions were also immunoblotted with anti-FLAG or anti-14-3-3 and anti-Histone H3 antibody. This experiment was repeated three times each on three independently derived C4MITF+ cell clones, and representative results for one cell clone are shown.