Interaction Trap/Two‐Hybrid System to Identify Interacting Proteins

Regulating repression: roles for the sir4 N-terminus in linker DNA protection and stabilization of epigenetic states

Silent information regulator proteins Sir2, Sir3, and Sir4 form a heterotrimeric complex that represses transcription at subtelomeric regions and homothallic mating type (HM) loci in budding yeast. We have performed a detailed biochemical and genetic analysis of the largest Sir protein, Sir4. The N-terminal half of Sir4 is dispensable for SIR–mediated repression of HM loci in vivo, except in strains that lack Yku70 or have weak silencer elements. For HM silencing in these cells, the C-terminal domain (Sir4C, residues 747–1,358) must be complemented with an N-terminal domain (Sir4N; residues 1–270), expressed either independently or as a fusion with Sir4C. Nonetheless, recombinant Sir4C can form a complex with Sir2 and Sir3 in vitro, is catalytically active, and has sedimentation properties similar to a full-length Sir4-containing SIR complex. Sir4C-containing SIR complexes bind nucleosomal arrays and protect linker DNA from nucleolytic digestion, but less effectively than wild-type SIR complexes. Consistently, full-length Sir4 is required for the complete repression of subtelomeric genes. Supporting the notion that the Sir4 N-terminus is a regulatory domain, we find it extensively phosphorylated on cyclin-dependent kinase consensus sites, some being hyperphosphorylated during mitosis. Mutation of two major phosphoacceptor sites (S63 and S84) derepresses natural subtelomeric genes when combined with a serendipitous mutation (P2A), which alone can enhance the stability of either the repressed or active state. The triple mutation confers resistance to rapamycin-induced stress and a loss of subtelomeric repression. We conclude that the Sir4 N-terminus plays two roles in SIR–mediated silencing: it contributes to epigenetic repression by stabilizing the SIR–mediated protection of linker DNA; and, as a target of phosphorylation, it can destabilize silencing in a regulated manner.

Three Silent Information Regulator (SIR) proteins Sir2, Sir3, and Sir4 are involved in the epigenetic gene silencing of the homothallic mating (HM) loci and of telomere-proximal genes in budding yeast. They bind as a heterotrimeric complex to chromatin, repressing the underlying genes. Sir2 has an essential histone deacetylase activity, and Sir3 binds nucleosomes, with a high specificity for unmodified histones. We explored Sir4, whose role had largely remained a mystery. We report here that Sir4 N- and C-terminal domains have distinct functions: The Sir4 C-terminus binds all proteins essential for SIR–mediated silencing and is sufficient to repress HM loci, but surprisingly it is not sufficient to efficiently repress at telomeres. The Sir4 N-terminus binds DNA, which strengthens the SIR–chromatin interaction and helps target Sir4 to telomeric loci. In addition the Sir4 N-terminus binds sequence-specific factors that recruit Sir4 to sites of repression. We find that the Sir4 N-terminus is a target of mitotic phosphorylation. Mutation of the phosphoacceptor sites indicates that they help fine-tune subtelomeric repression. We propose therefore that phosphorylation of the Sir4 N-terminal domain modulates epigenetic repression at telomeres in response to cell cycle and/or stress situations.

Proteomic analysis of pluripotent stem cells

Mass spectrometry (MS)-based proteomics has become one of the most powerful tools for identifying expressed proteins, providing quick insights into molecular and cellular biology. Traditionally, proteins isolated by either one- or two-dimensional gel electrophoresis are digested with a site specific protease. The resulting peptides are subject to one of two forms of analysis: (1) matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) MS, where a "mass fingerprint" of all the peptides in a sample is generated, or (2) electrospray ionization tandem MS (ESI-MS/MS), where a mass fragmentation spectra is generated for each peptide in a sample. The resulting mass information is then compared to that of a theoretical database created with available genomic sequence information. This unit provides protocols for this type of assessment in embryonic stem cells (ESCs).

A HECT E3 ubiquitin-protein ligase with sequence similarity to E6AP does not target p53 for degradation in the softshell clam (Mya arenaria)

Numerous reports have raised the level of national concern that chemicals found in the environment may have adverse effects on the health of humans and wildlife. Environmental exposure to pollutants, such as dioxin, has been implicated in gonadal tumor formation in Maine softshell clams (Mya arenaria). Prevalence of these tumors is as high as 40% in some populations. Although their etiology is still unknown, investigations into the mechanisms of tumor formation have revolved around a hypothesis of dioxin-induced toxicity. The aryl hydrocarbon receptor (AHR) was initially investigated, but was later determined to not bind the prototypical ligand, 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD), suggesting that dioxin toxicity is mediated through an AHR-independent pathway. An alternative mechanism of tumor formation has been investigated, involving a protein with significant sequence similarity to mammalian E6AP, a HECT (homologous to E6AP carboxy terminus) E3 ubiquitin-protein ligase. E6AP, in association with the high-risk human papillomavirus (HPV) E6 protein, is involved in the abnormal degradation of the p53 tumor suppressor protein in human cervical cancer. Tumorigenic clam reproductive tissue revealed higher M. arenaria E3 (MaE3) protein levels concomitant with lower M. arenaria p53 (Map53) levels. While the function of MaE3 as a HECT E3 was verified, results from three methods agree that MaE3 does not associate with Map53. However, alteration in Map53 levels may still play a role in clam gonadal tumorigenesis. Due to upregulation of MaE3 in neoplastic reproductive tissue, further investigations will focus on determining the proteolytic targets of MaE3. In conjunction with our previous findings that dioxin toxicity in the softshell clam is not mediated by AHR, the results from our current investigation suggest a complex etiology for the clam germinomas.

COI1 links jasmonate signalling and fertility to the SCF ubiquitin-ligase complex in Arabidopsis

Jasmonates (JAs) regulate Arabidopsis thaliana wound and defence responses, pollen development, and stress-related growth inhibition. Significantly, each of these responses requires COI1, an F-box protein. Other F-box proteins interact with SKP1 and cullin proteins to form SCF complexes that selectively recruit regulatory proteins targeted for ubiquitination. To determine whether COI1 also functions in an SCF complex, we have characterized Arabidopsis proteins that bind to COI1. An Arabidopsis cDNA expression library was screened in yeast for clones that produce proteins which can bind to COI1. We recovered two SKP1 homologues and a histone deacetylase. The Arabidopsis F-box protein TIR1 interacted with SKP1 proteins, but not with the histone deacetylase. Mutant COI1 proteins revealed that the F-box is required for interaction with SKP1s, but that sequences in leucine-rich repeat domains are required for interaction with the histone deacetylase. Epitope-tagged COI1 was introduced into Arabidopsis plants and cell cultures. Co-immunoprecipitation experiments confirmed the interaction in planta of COI1 with SKP1-like proteins and histone deacetylase, and also indicated that COI1 interacted with cullin. These results suggest that COI1 forms an SCFCOI1 complex in vivo. COI1 is therefore expected to form a functional E3-type ubiquitin ligase in plants and to regulate expression of jasmonate responsive genes, possibly by targeted ubiquitination of a histone deacetylase.

Identification and characterization of novel substrates of Trk receptors in developing neurons

Neurotrophins influence growth and survival of specific populations of neurons through activation of Trks, members of the receptor tyrosine kinase (RTK) family. In this report, we describe the identification and characterization of two substrates of Trk kinases, rAPS and SH2-B, which are closely related Src homolog 2 (SH2) domain-containing signaling molecules. rAPS and SH2-B are substrates of TrkB and TrkC in cortical neurons and SH2-B is a substrate of TrkA in sympathetic neurons. Moreover, rAPS and SH2-B bind to Grb2, and both are sufficient to mediate NGF induction of Ras, MAP kinase (MAPK), and morphological differentiation of PC12 cells. Lastly, antibody perturbation and transient transfection experiments indicate that SH2-B, or a closely related molecule, is necessary for NGF-dependent signaling in neonatal sympathetic neurons. Together, these observations indicate that rAPS and SH2-B mediate Trk signaling in developing neurons.

Ribonuclease E organizes the protein interactions in the Escherichia coli RNA degradosome

The Escherichia coli RNA degradosome is the prototype of a recently discovered family of multiprotein machines involved in the processing and degradation of RNA. The interactions between the various protein components of the RNA degradosome were investigated by Far Western blotting, the yeast two-hybrid assay, and coimmunopurification experiments. Our results demonstrate that the carboxy-terminal half (CTH) of ribonuclease E (RNase E) contains the binding sites for the three other major degradosomal components, the DEAD-box RNA helicase RhlB, enolase, and polynucleotide phosphorylase (PNPase). The CTH of RNase E acts as the scaffold of the complex upon which the other degradosomal components are assembled. Regions for oligomerization were detected in the amino-terminal and central regions of RNase E. Furthermore, polypeptides derived from the highly charged region of RNase E, containing the RhlB binding site, stimulate RhlB activity at least 15-fold, saturating at one polypeptide per RhlB molecule. A model for the regulation of the RhlB RNA helicase activity is presented. The description of RNase E now emerging is that of a remarkably complex multidomain protein containing an amino-terminal catalytic domain, a central RNA-binding domain, and carboxy-terminal binding sites for the other major components of the RNA degradosome.

Delineation of the interaction domains of Agrobacterium tumefaciens VirB7 and VirB9 by use of the yeast two-hybrid assay

The Agrobacterium tumefaciens VirB proteins are postulated to form a transport pore for the transfer of T-DNA. Formation of the transport pore will involve interactions among the VirB proteins. A powerful genetic method to study protein-protein interaction is the yeast two-hybrid assay. To test whether this method can be used to study interactions among the VirB membrane proteins, we studied the interaction of VirB7 and VirB9 in yeast. We recently demonstrated that VirB7 and VirB9 form a protein complex linked by a disulfide bond between cysteine 24 of VirB7 and cysteine 262 of VirB9 (L. Anderson, A. Hertzel, and A. Das, Proc. Natl. Acad. Sci. USA 93:8889-8894, 1996). We now demonstrate that VirB7 and VirB9 interact in yeast, and this interaction does not require the cysteine residues essential for the disulfide linkage. By using defined segments in fusion constructions, we mapped the VirB7 interaction domain of VirB9 to residues 173 to 275. In tumor formation assays, both virB7C24S and virB9C262S expressed from a multicopy plasmid complemented the respective deletion mutation, indicating that the cysteine residues may not be essential for DNA transfer.