The phospho-occupancy of an atypical SLIMB-binding site on PERIOD that is phosphorylated by DOUBLETIME controls the pace of the clock

Abstract

A common feature of animal circadian clocks is the progressive phosphorylation of PERIOD (PER) proteins, which is highly dependent on casein kinase Idelta/epsilon (CKIdelta/epsilon; termed DOUBLETIME [DBT] in Drosophila) and ultimately leads to the rapid degradation of hyperphosphorylated isoforms via a mechanism involving the F-box protein, beta-TrCP (SLIMB in Drosophila). Here we use the Drosophila melanogaster model system, and show that a key step in controlling the speed of the clock is phosphorylation of an N-terminal Ser (S47) by DBT, which collaborates with other nearby phosphorylated residues to generate a high-affinity atypical SLIMB-binding site on PER. DBT-dependent increases in the phospho-occupancy of S47 are temporally gated, dependent on the centrally located DBT docking site on PER and partially counterbalanced by protein phosphatase activity. We propose that the gradual DBT-mediated phosphorylation of a nonconsensus SLIMB-binding site establishes a temporal threshold for when in a daily cycle the majority of PER proteins are tagged for rapid degradation. Surprisingly, most of the hyperphosphorylation is unrelated to direct effects on PER stability. We also use mass spectrometry to map phosphorylation sites on PER, leading to the identification of a number of "phospho-clusters" that explain several of the classic per mutants.

Figure 1.

Identification of N-terminal sequences required for DBT-mediated degradation of dPER in cultured S2 cells. S2 cells were cotransfected with wild-type or mutant variants of dper containing plasmids (pAc-dper-V5-His [A–C] or pAc-3XFlag-His-dper/Tev100-6Xc-myc [D]) and pMT-dbt-V5-His, and collected at the indicated times (hours) post-dbt induction. dPER variants were detected by immunoblotting in the presence of α-V5 (A–C) or α-Flag antibodies (D).

Figure 2.

Using the TEV/TAG strategy to evaluate phosphorylation in the first 100 amino acids of dPER. (A) Schematic model showing the pAc-3XFlag-His-dper/Tev100-6Xc-myc construct and resulting cleavage products. (B) Extracts were prepared from S2 cells expressing pAc-3XFlag-His-dper/Tev100-6Xc-myc with (+) or without (−) coexpression of pMT-dbt-V5-His. (Right) Extracts were subjected to TEV cleavage, incubated with α-Flag beads, and immune complexes treated with λ-phosphatase (+) or mock treated (−). dPER(1–100) was detected using α-Flag antibodies. Hypo- and hyperphosphorylated isoforms are indicated. (*) Hyperphosphorylated isoforms only detected in the presence of induced dbt. (C) S2 cells coexpressing pAc-3XFlag-His-dper/Tev100-6Xc-myc and pMT-dbt-V5-His were harvested at the indicated times post-dbt induction and extracts were either TEV-treated or mock-treated. Mock-treated extracts were subjected to immunoblotting in the presence of α-c-myc antibodies to detect full-length dPER (top), and TEV-treated extracts were used to detect dPER(1–100) using α-Flag antibodies (bottom).

Figure 3.

S47 is a key phospho-determinant mediating dPER–SLIMB interactions. (A) S2 cells coexpressing pAc-3XFlag-His-dper/Tev100-6Xc-myc and pMT-dbt-V5-His were collected at the indicated times post-dbt induction. Extracts were either treated with TEV to yield amino acids 1–100 or amino acids 101–1224, or mock-treated retaining the full-length amino acids 1–1224. The top panel shows the input used in GST-SLIMB pull-down assays and the bottom panel shows the bound dPER proteins. The amino acid 1–100 fragment was detected using α-Flag antibodies, whereas full-length dPER and the amino acid 101–1224 fragment were detected using α-c-myc antibodies. Hyperphosphorylated and hypophosphorylated dPER isoforms are indicated as Hyper-P and Hypo-P, respectively. (B–D) Extracts collected from S2 cells expressing wild-type (WT) or mutant derivatives of pAc-3XFlag-His-dper/Tev100-6Xc-myc with (+) or without (−) induced dbt were subjected to TEV cleavage, subjected to GST-SLIMB pull-down assays and the amino acid 1–100 dPER fragment detected with α-Flag antibodies. (B) Asterisk (*) denotes S44–48A mutant hyperphosphorylated isoforms with faster mobility as compared with wild-type hyperphosphorylated isoforms.

Figure 4.

Daily cycles in dper protein and RNA are altered in dper mutant flies. Head extracts were prepared and used to detect either dPER protein (A–F) or dper RNA (G–J). For the analysis of dPER protein, the different transgenes were evaluated in the wper⁰ genetic background, whereas the levels of dper transcripts were determined in both wper⁰ and wper⁺ genetic backgrounds. The dPER protein profiles of wild type (A,D, lanes 1,2) and four mutants (B–F) were monitored throughout a daily cycle by immunoblotting in the presence of α-HA. Open and closed triangles (A,B) denote hyper- and hypophosphorylated dPER, respectively. (G–J) dper mRNA profiles obtained by semiquantitative RT–PCR using cbp20 for normalization. The control dper mRNA profile (n = 4) derived from wper⁰;per⁺-HAHis flies is shown in all four plots to facilitate comparisons with each mutant profile (n = 2). The error bars shown are SEM.

Figure 5.

GST-SLIMB pull-down assays using fly head extracts further support S47 as a key phospho-determinant underlying SLIMB binding to dPER. (A,B) Head extracts were prepared and aliquots containing equal amounts of total protein were incubated with glutathione resins bound with GST-SLIMB or GST. The relative amount of dPER-HA in the starting material (input) and that bound to the resins (GST/GST-Sb) was visualized by immunoblotting in the presence of α-HA antibodies.

Figure 6.

Phosphospecific antibodies detect pS47-containing isoforms of dPER in S2 cells and flies. (A) Extracts were prepared from S2 cells expressing either a wild-type (WT) or S47A version of pAc-3XFlag-His-dper/Tev100-6Xc-myc. In some cases cells were cotransfected with pMT-dbt-V5-His (+), whereas this was omitted in other cases (−). dPER proteins were immunoprecipitated (IP) with α-Flag beads, treated in the absence (−λPP) or presence (+λPP) of λ-phosphatase, and analyzed by Western blotting (WB) in the presence of the indicated antibody (left of panel). (B, top) Cells were cotransfected with pAc-3XFlag-His-dper-6Xc-myc and pMT-dbt-V5-His and collected at the indicated times post-dbt induction. Extracts were prepared and divided into three separate aliquots: (1) One aliquot was immunoprecipitated (IP) with α-Flag beads and analyzed by immunoblotting in the presence of α-pS47 antibodies (top); (2) another aliquot was directly analyzed by immunoblotting in the presence of α-c-myc antibodies (middle); and (3) a final aliquot was subjected to GST-SLIMB pull-down assays and bound material visualized by immunoblotting in the presence of α-c-myc antibodies (bottom). (C,D) Head extracts were prepared from wper⁰;per⁺-HAHis flies collected at the indicated times (ZT). (C) dPER-HAHis-containing immune complexes recovered using α-HA beads were subjected to Western blotting (WB) in the presence of the indicated antibody (left of panel). Open and closed triangles indicate hyper- and hypophosphorylated pS47-containing isoforms of dPER, respectively. (D) An aliquot was directly analyzed (top), whereas another fraction was subjected to GST-SLIMB pull-down assays and bound material analyzed (bottom).

Figure 7.

Ser47 of Drosophila PER is phosphorylated by DBT and CKIδ in vitro. (A,B) Extracts were prepared from S2 cells express-ing wild-type or ΔdPDBD versions of pAc-3XFlag-His-dper/Tev100-6Xc-myc. (Left) dPER proteins were immunoprecipitated using α-Flag beads and subjected to in vitro kinase assays, followed by Western blotting (WB) in the presence of the indicated antibody. dPER containing immune complexes were either mock-treated (−) or pretreated with λ-phosphatase (+) prior to in vitro incubation in the absence (−) or presence of the indicated kinase.