Multisite phosphorylation disrupts arginine-glutamate salt bridge networks required for binding of cytoplasmic linker-associated protein 2 (CLASP2) to end-binding protein 1 (EB1)

A group of diverse proteins reversibly binds to growing microtubule plus ends through interactions with end-binding proteins (EBs). These +TIPs control microtubule dynamics and microtubule interactions with other intracellular structures. Here, we use cytoplasmic linker-associated protein 2 (CLASP2) binding to EB1 to determine how multisite phosphorylation regulates interactions with EB1. The central, intrinsically disordered region of vertebrate CLASP proteins contains two SXIP EB1 binding motifs that are required for EB1-mediated plus-end-tracking in vitro. In cells, both EB1 binding motifs can be functional, but most of the binding free energy results from nearby electrostatic interactions. By employing molecular dynamics simulations of the EB1 interaction with a minimal CLASP2 plus-end-tracking module, we find that conserved arginine residues in CLASP2 form extensive hydrogen-bond networks with glutamate residues predominantly in the unstructured, acidic C-terminal tail of EB1. Multisite phosphorylation of glycogen synthase kinase 3 (GSK3) sites near the EB1 binding motifs disrupts this electrostatic "molecular Velcro." Molecular dynamics simulations and (31)P NMR spectroscopy indicate that phosphorylated serines participate in intramolecular interactions with and sequester arginine residues required for EB1 binding. Multisite phosphorylation of these GSK3 motifs requires priming phosphorylation by interphase or mitotic cyclin-dependent kinases (CDKs), and we find that CDK- and GSK3-dependent phosphorylation completely disrupts CLASP2 microtubule plus-end-tracking in mitosis.     

The interaction site of Flap Endonuclease-1 with WRN helicase suggests a coordination of WRN and PCNA

Werner and Bloom syndromes are genetic RecQ helicase disorders characterized by genomic instability. Biochemical and genetic data indicate that an important protein interaction of WRN and Bloom syndrome (BLM) helicases is with the structure-specific nuclease Flap Endonuclease 1 (FEN-1), an enzyme that is implicated in the processing of DNA intermediates that arise during cellular DNA replication, repair and recombination. To acquire a better understanding of the interaction of WRN and BLM with FEN-1, we have mapped the FEN-1 binding site on the two RecQ helicases. Both WRN and BLM bind to the extreme C-terminal 18 amino acid tail of FEN-1 that is adjacent to the PCNA binding site of FEN-1. The importance of the WRN/BLM physical interaction with the FEN-1 C-terminal tail was confirmed by functional interaction studies with catalytically active purified recombinant FEN-1 deletion mutant proteins that lack either the WRN/BLM binding site or the PCNA interaction site. The distinct binding sites of WRN and PCNA and their combined effect on FEN-1 nuclease activity suggest that they may coordinately act with FEN-1. WRN was shown to facilitate FEN-1 binding to its preferred double-flap substrate through its protein interaction with the FEN-1 C-terminal binding site. WRN retained its ability to physically bind and stimulate acetylated FEN-1 cleavage activity to the same extent as unacetylated FEN-1. These studies provide new insights to the interaction of WRN and BLM helicases with FEN-1, and how these interactions might be regulated with the PCNA–FEN-1 interaction during DNA replication and repair.     

Interactions between N- and C-terminal domains of the Saccharomyces cerevisiae high-mobility group protein HMO1 are required for DNA bending

The Saccharomyces cerevisiae high-mobility group protein HMO1 is composed of two DNA-binding domains termed box A and box B, of which only box B is predicted to adopt a HMG fold, and a lysine-rich C-terminal extension. To assess the interaction between individual domains and their contribution to DNA binding, several HMO1 variants were analyzed. Using circular dichroism spectroscopy, thermal stability was measured. While the melting temperatures of HMO1-boxA and HMO1-boxB are 57.2 and 47.2 degrees C, respectively, HMO1-boxBC, containing box B and the entire C-terminal tail, melts at 46.1 degrees C, suggesting little interaction between box B and the tail. In contrast, full-length HMO1 exhibits a single melting transition at 47.9 degrees C, indicating that interaction between box A and either box B or the tail destabilizes this domain. As HMO1-boxAB, lacking only the lysine-rich C-terminal segment, exhibits two melting transitions at 46.0 and 63.3 degrees C, we conclude that the destabilization of the box A domain seen in full-length HMO1 is due primarily to its interaction with the lysine-rich tail. Determination of DNA substrate specificity using electrophoretic mobility shift assays shows unexpectedly that the lysine-rich tail does not increase DNA binding affinity but instead is required for DNA bending by full-length HMO1; HMO1-boxBC, lacking the box A domain, also fails to bend DNA. In contrast, both HMO1 and HMO1-boxAB, but not the individual HMG domains, exhibit preferred binding to constrained DNA minicircles. Taken together, our data suggest that interactions between box A and the C-terminal tail induce a conformation that is required for DNA bending.     

Cyclin-dependent kinase 1 depolymerizes nuclear lamin filaments by disrupting the head-to-tail interaction of the lamin central rod domain

Nuclear lamins maintain the nuclear envelope structure by forming long linear filaments via two alternating molecular arrangements of coiled-coil dimers, known as A11 and A22 binding modes. The A11 binding mode is characterized by the antiparallel interactions between coil 1b domains, whereas the A22 binding mode is facilitated by interactions between the coil 2 domains of lamin. The junction between A11- and A22-interacting dimers in the lamin tetramer produces another parallel head–tail interaction between coil 1a and the C-terminal region of coil 2, called the ACN interaction. During mitosis, phosphorylation in the lamin N-terminal head region by the cyclin-dependent kinase (CDK) complex triggers depolymerization of lamin filaments, but the associated mechanisms remain unknown at the molecular level. In this study, we revealed using the purified proteins that phosphorylation by the CDK1 complex promotes disassembly of lamin filaments by directly abolishing the ACN interaction between coil 1a and the C-terminal portion of coil 2. We further observed that this interaction was disrupted as a result of alteration of the ionic interactions between coil 1a and coil 2. Combined with molecular modeling, we propose a mechanism for CDK1-dependent disassembly of the lamin filaments. Our results will help to elucidate the cell cycle–dependent regulation of nuclear morphology at the molecular level.     

Epstein-Barr virus nuclear protein 2 transactivation of the latent membrane protein 1 promoter is mediated by J kappa and PU.1.

Expression of the Epstein-Barr virus (EBV) latent membrane protein 1 (LMP-1) oncogene is regulated by the EBV nuclear protein 2 (EBNA-2) transactivator. EBNA-2 is known to interact with the cellular DNA-binding protein J kappa and is recruited to promoters containing the GTGGGAA J kappa recognition sequence. The minimal EBNA-2-responsive LMP-1 promoter includes one J kappa-binding site, and we now show that mutation of that site, such that J kappa cannot bind, reduces EBNA-2 responsiveness by 60%. To identify other factors which interact with the LMP-1 EBNA-2 response element (E2RE), a -236/-145 minimal E2RE was used as a probe in an electrophoretic mobility shift assay. The previously characterized factors J kappa, PU.1, and AML1 bind to the LMP-1 E2RE, along with six other unidentified factors (LBF2 to LBF7). Binding sites were mapped for each factor. LBF4 is B- and T-cell specific and recognizes the PU.1 GGAA core sequence as shown by methylation interference. LBF4 has a molecular mass of 105 kDa and is probably unrelated to PU.1. LBF2 was found only in epithelial cell lines, whereas LBF3, LBF5, LBF6, and LBF7 were not cell type specific. Mutations of the AML1- or LBF4-binding sites had no effect on EBNA-2 transactivation, whereas mutation of the PU.1-binding site completely eliminated EBNA-2 responses. A gst-EBNA-2 fusion protein specifically depleted PU.1 from nuclear extracts and bound in vitro translated PU.1, providing biochemical evidence for a direct EBNA-2-PU.1 interaction. Thus, EBNA-2 transactivation of the LMP-1 promoter is dependent on interaction with at least two distinct sequence-specific DNA-binding proteins, J kappa and PU.1. LBF3, LBF5, LBF6, or LBF7 may also be involved, since their binding sites also contribute to EBNA-2 responsiveness.     

NMR studies of interactions between C-terminal tail of Kir2.1 channel and PDZ1,2 domains of PSD95

Control of surface expression of inwardly rectifying potassium (Kir) channels is important for regulating membrane excitability. Kir2 channels have been shown to interact directly with PDZ-containing proteins in the postsynaptic density (PSD). These scaffold proteins, such as PSD95, bind to Kir2.1 channels via a PDZ-binding motif (T/S-x-Phi) in the C-terminal tail (SEI428). By utilizing a multidimensional solution NMR approach, we show that the previously unresolved structure of Kir2.1 tail (residues 372-428) is highly flexible. Using in vitro binding assays, we determined that shortening the flexible tail of Kir2.1 preceding the C-terminal region (residues 414-428) does not significantly disrupt PDZ binding. We also investigated which amino acids in the Kir2.1 tail associated with PSD95 PDZ1,2 by NMR spectroscopy, revealing that a stretch of 12 C-terminal amino acids is involved in interaction with both PDZ domains (residues 417-428). Deletion of the 11 amino acids preceding the C-terminal tail, Delta414-424, completely disrupts binding to PSD95 PDZ1,2. Therefore, the molecular interfaces formed between PDZ domains and Kir2.1 tail involve regions outside the previously identified binding motif (SEI428) and may be important for additional channel-specific interactions with associating PDZ-containing proteins.     

Contribution of the C-terminal tail of U1A RBD1 to RNA recognition and protein stability.

The N-terminal RNA binding domain (RBD1) of the human U1A protein interacts specifically with a short RNA hairpin containing the U1 snRNA stem/loop II sequence. Previous RNA binding studies have suggested that the C-terminal tail of RBD1 contributes to RNA recognition in addition to interactions on the beta-sheet surface of the protein. To evaluate the contributions of these C-terminal residues in RBD1 to RNA binding affinity and specificity, as well as to study the thermodynamic stability of RBDs, a number of RBD1 mutants with truncated tails, with single amino acid substitutions, and with both a truncation and an amino acid substitution, have been constructed. The thermodynamic stabilities of these mutants have been measured and compared by GdnHCI unfolding experiments. The RNA binding affinity and specificity of these mutant proteins have been assessed by measuring the binding of each protein to the wild-type RNA hairpin and to selected RNA mutants with nucleotide substitutions in the RNA loop. The results demonstrate first that, although the C-terminal tail of RBD1 makes significant contributions to RNA binding affinity, it is not required for RNA binding, and second, its contributions to binding specificity are mediated only through selected nucleotides in the RNA loop, for in the absence of the tail, the protein continues to use other nucleotides to discriminate among RNAs. In these truncated proteins, the secondary structure intrinsic to the C-terminal tail is absent, yet their affinity and discrimination for RNAs are not lost. Thus, a structured tail is not required for RNA recognition.     

Interactions of the basic N-terminal and the acidic C-terminal domains of the maize chromosomal HMGB1 protein

Maize HMGB1 is a typical member of the family of plant chromosomal HMGB proteins, which have a central high-mobility group (HMG)-box DNA-binding domain that is flanked by a basic N-terminal region and a highly acidic C-terminal domain. The basic N-terminal domain positively influences various DNA interactions of the protein, while the acidic C-terminal domain has the opposite effect. Using DNA-cellulose binding and electrophoretic mobility shift assays, we demonstrate that the N-terminal basic domain binds DNA by itself, consistent with its positive effects on the DNA interactions of HMGB1. To examine whether the negative effect of the acidic C-terminal domain is brought about by interactions with the basic part of HMGB1 (N-terminal region, HMG-box domain), intramolecular cross-linking in combination with formic acid cleavage of the protein was used. These experiments revealed that the acidic C-terminal domain interacts with the basic N-terminal domain. The intramolecular interaction between the two oppositely charged termini of the protein is enhanced when serine residues in the acidic tail of HMGB1 are phosphorylated by protein kinase CK2, which can explain the negative effect of the phosphorylation on certain DNA interactions. In line with that, covalent cross-linking of the two terminal domains resulted in a reduced affinity of HMGB1 for linear DNA. Comparable to the finding with maize HMGB1, the basic N-terminal and the acidic C-terminal domains of the Arabidopsis HMGB1 and HMGB4 proteins interact, indicating that these intramolecular interactions, which can modulate HMGB protein function, generally occur in plant HMGB proteins.     

Molecular basis of the head‐to‐tail assembly of giant muscle proteins obscurin‐like 1 and titin

Large filament proteins in muscle sarcomeres comprise many immunoglobulin‐like domains that provide a molecular platform for self‐assembly and interactions with heterologous protein partners. We have unravelled the molecular basis for the head‐to‐tail interaction of the carboxyl terminus of titin and the amino‐terminus of obscurin‐like‐1 by X‐ray crystallography. The binary complex is formed by a parallel intermolecular β‐sheet that presents a novel immunoglobulin‐like domain‐mediated assembly mechanism in muscle filament proteins. Complementary binding data show that the assembly is entropy‐driven rather than dominated data by specific polar interactions. The assembly observed leads to a V‐shaped zipper‐like arrangement of the two filament proteins.