CaMKIV-Mediated Phosphorylation Inactivates Freud-1/CC2D1A Repression for Calcium-Dependent 5-HT1A Receptor Gene Induction

Abstract

Calcium calmodulin-dependent protein kinase (CaMK) mediates calcium-induced neural gene activation. CaMK also inhibits the non-syndromic intellectual disability gene, Freud-1/CC2D1A, a transcriptional repressor of human serotonin-1A (5-HT1A) and dopamine-D2 receptor genes. The altered expression of these Freud-1-regulated genes is implicated in mental illnesses such as major depression and schizophrenia. We hypothesized that Freud-1 is blocked by CaMK-induced phosphorylation. The incubation of purified Freud-1 with either CaMKIIα or CaMKIV increased Freud-1 phosphorylation that was partly prevented in Freud-1-Ser644Ala and Freud-1-Thr780Ala CaMK site mutants. In human SK-N-SH neuroblastoma cells, active CaMKIV induced the serine and threonine phosphorylation of Freud-1, and specifically increased Freud-1-Thr780 phosphorylation in transfected HEK-293 cells. The activation of purified CaMKIIα or CaMKIV reduced Freud-1 binding to its DNA element on the 5-HT1A and dopamine-D2 receptor genes. In SK-N-SH cells, active CaMKIV but not CaMKIIα blocked the Freud-1 repressor activity, while Freud-1 Ser644Ala, Thr780Ala or dual mutants were resistant to inhibition by activated CaMKIV or calcium mobilization. These results indicate that the Freud-1 repressor activity is blocked by CaMKIV-induced phosphorylation at Thr780, resulting in the up-regulation of the target genes, such as the 5-HT1A receptor gene. The CaMKIV-mediated inhibition of Freud-1 provides a novel de-repression mechanism to induce 5-HT1A receptor expression for the regulation of cognitive development, behavior and antidepressant response.

Keywords: 5-HT1A receptor; gene repression; major depressive disorder; phosphorylation; raphe; transcription factor.

Figure 1

Freud-1 phosphorylation by CaMKIIα and CaMKIV in vitro. (A,B) In vitro kinase assay. Recombinant purified human Freud-1 (hF1) was incubated in the absence or presence of CaMKIIα (A) or CaMKIV (B) and [³²P] incorporation was detected via autoradiography (upper panels); Freud-1 protein was detected in the same gels via Coomassie Blue staining (lower panels). Data are representative of at least three independent experiments. (C,D) Phosphorylation of Freud-1 mutants. Above: Purified Freud-1 wild-type or mutants Ser644Ala and Thr780Ala or the double mutant were incubated in the presence of CaMKIIα (C) or CaMKIV (D) and subjected to autoradiography. Purified Freud-1 protein was detected via Western blot using S-tag antibody. Below: Quantification of Freud-1 phosphorylation relative to wild-type (WT) Freud-1. Data represent mean ± SE of three separate experiments; * p < 0.05; ** p < 0.01 compared to WT, by using one-way ANOVA with Tukey post-test.

Figure 2

CaMKIV-mediated phosphorylation of Freud-1 in SK-N-SH cells. (A) Phospho-serine (pSer)/phospho-threonine (pThr): Lysates from transfected SK-N-SH cells were used in pulldown of His/S-tagged Freud-1. Cells were transfected with plasmids for constitutively active CaMKIV Δ1-317 (+), His/S-tagged Freud-1 (+) or control vector (−). Input and elution fractions were analyzed via Western blot using S-tag antibody for Freud-1 (arrow), and pSer or pThr antibody (1:5000) for phosphorylation. (B) Phospho-Thr780: HEK-293 cells were co-transfected with constitutively active CaMKIV Δ1-317 (+) and His/S-tagged Freud-1 (lane 1), vector and Freud-1 (lane 2), CaMKIV (lane 3) or vectors only (lane 4). Western blots were probed with anti-phospho-T780 Freud-1 as shown on a representative blot. Right, band intensities were quantified using Image J win64 Software for densitometry analysis of peak areas. Data are expressed as % above vector control (0%) as mean ± SEM, n = 3; **** p < 0.0001 vs. vector; ** p < 0.01 between groups, one-way ANOVA, Tukey post-test.

Figure 3

CaMK reduces Freud-1 binding to 5-HT1A and D2 DREs in vitro. Purified Freud-1 was analyzed via EMSA using end-labeled 5-HT1A (5′DRE, 3′DRE) (A) or D2-DRE probes (B). Two Freud-1/DNA complexes were detected, as observed previously (arrows). Standard conditions or CaMKIIα buffer at 30 °C (30 °C buffer; or all samples in (B)) for kinase activity were used, with or without CaMKIIα or CaMKIV, as indicated. Unlabeled 200x molar excess of unlabeled DRE competitor (200XComp) blocked the interaction, but unrelated E2F competitor did not. (C) Concentration-dependent Freud-1-DRE complex formation. Bacterially expressed, purified wild-type human Freud-1 (WT) or Freud-1 mutants S644A, T780A, S644A/T780A and S644D/T780D were analyzed via EMSA with labeled 5-HT1A DRE using increasing concentrations (1.5—wild-type only, 2, 2.5, 3 ug) of Freud-1 protein, with no protein added to the first lane. Two complexes were seen, except for T780-containing mutants where a third complex was also seen. Data are representative of at least three independent experiments.

Figure 4

Active CaMKIV but not CaMKIIα attenuates Freud-1 repression. Freud-1 repressor activity was determined using the Gal4 hybrid system (Supplementary Figure S3). (A) SK-N-SH cells were co-transfected with reporter constructs G₅P, Gal4-DBD or Gal4-DBD-Freud-1, and CaMKIIα vector (SRα), constitutively active CaMKIIα T286D, CaMKIV vector (PSG5) or constitutively active CaMKIV Δ1-317. Reporter activity determined in triplicate samples was normalized to β-galactosidase activity and is presented as % of control for each vector. Data represent mean ± SEM of three independent experiments. (B) Freud-1 phosphorylation site mutants are resistant to CaMKIV suppression. SK-N-SH cells were transfected with G₅P, CaMKIV vector (PSG5) or CaMKIV Δ1-317 and Gal4 DBD, Gal4 DBD hFreud-1 or Gal4-DBD Freud-1 mutants. Values for reporter activity are expressed as %CaMKIV-induced suppression = 100 × [CaMKIV Δ1-317]/[PSG5] of reporter activity for each Gal4-Freud-1 construct normalized to the ratio for Gal4-DBD vector (100%). Triplicate transfections from three independent experiments are presented as mean ± SEM. * p < 0.05; via unpaired two-tailed t-test.

Figure 5

Repression by Freud-1 phospho-site mutants is resistant to active CaMKIV. (A) Calcium-CaMKIV-mediated de-repression of 5-HT1A-DRE. SK-N-SH cells were co-transfected with 5-HT1A DRE-SV40 promoter-luciferase construct, CaMKIV or vector (PSG5) and human Freud-1 (hF1) or S644A/T780A-Freud-1 (hF1 AA) or vector (pcDNA3). Cells were treated with vehicle (DMSO) or 40 mM KCl and 1 uM ionomycin for 4 h prior to cell collection. Relative luciferase activity from triplicate samples was normalized to activity in pcDNA3 vector samples (dashed line). Mean ± SEM (n = six experiments). ** p < 0.01 vs. DMSO control, ANOVA with Bonferroni post-test. (B) Freud-1 repression of the 5-HT1A promoter. SK-N-SH cells were co-transfected with human 5-HT1A promoter-luciferase constructs, -1517 (lacking DREs) or -1790ΔRE-1 containing the 5′ and 3′-DRE with the RE-1 deleted [31] and Freud-1 expressing plasmid or vector pcDNA3. Samples were assayed for reporter activity and are presented as % of activity of the DRE-lacking -1517 construct (see Supplementary Figure S3). * p < 0.05; ** p < 0.01 by unpaired two-tailed t-test.

Figure 6

Structural model of Freud-1 domains. Freud-1/CC2D1A 3D structure predicted by AlphaFold [44] was obtained at: https://www.genecards.org/cgi-bin/carddisp.pl?gene=CC2D1A (accessed on 11 March 2024). Shown are the Freud-1 DM14 and C2 domains as predicted by the model and the site of the T780 residue (red asterisk). Below are blow-ups of the boxed region in two different rotations which show hydrogen bonding of the T840 (in red): T780-OH forms coordinate hydrogen bonds with V773=O…HO–T780 and G782–N…T780–OH.