Reverse two-hybrid and one-hybrid systems to detect dissociation of protein-protein and DNA-protein interactions.

Macromolecular interactions define many biological phenomena. Although genetic methods are available to identify novel protein-protein and DNA-protein interactions, no genetic system has thus far been described to identify molecules or mutations that dissociate known interactions. Herein, we describe genetic systems that detect such events in the yeast Saccharomyces cerevisiae. We have engineered yeast strains in which the interaction of two proteins expressed in the context of the two-hybrid system or the interaction between a DNA-binding protein and its binding site in the context of the one-hybrid system is deleterious to growth. Under these conditions, dissociation of the interaction provides a selective growth advantage, thereby facilitating detection. These methods referred to as the "reverse two-hybrid system" and "reverse one-hybrid system" facilitate the study of the structure-function relationships and regulation of protein-protein and DNA-protein interactions. They should also facilitate the selection of dissociator molecules that could be used as therapeutic agents.

The yeast GAL11 protein binds to the transcription factor IIE through GAL11 regions essential for its in vivo function.

The GAL11 gene encodes an auxiliary transcription factor required for full expression of many genes in yeast. The GAL11-encoded protein (Gal11p) has recently been shown to copurify with the holoenzyme of RNA polymerase II. Here we report that Gal11p stimulates basal transcription in a reconstituted transcription system composed of recombinant or highly purified transcription factors, TFIIB, TFIIE, TFIIF, TFIIH, and TATA box-binding protein and core RNA polymerase II. We further demonstrate that each of the two domains of Gal11p essential for in vivo function respectively participates in the binding to the small and large subunits of TFIIE. The largest subunit of RNA polymerase II was coprecipitated by anti-hemagglutinin epitope antibody from crude extract of GAL11 wild type yeast expressing hemagglutinintagged small subunit of TFIIE. Such a coprecipitation of the RNA polymerase subunit was seen but in a greatly reduced amount, if extract was prepared from gal11 null yeast. In light of these findings, we suggest that Gal11p stimulates promoter activity by enhancing an association of TFIIE with the preinitiation complex in the cell.

Generation of cell-to-cell signals in quorum sensing: acyl homoserine lactone synthase activity of a purified Vibrio fischeri LuxI protein.

Many bacteria use acyl homoserine lactone signals to monitor cell density in a type of gene regulation termed quorum sensing and response. Synthesis of these signals is directed by homologs of the luxi gene of Vibrio fischeri. This communication resolves two critical issues concerning the synthesis of the V. fischeri signal. (i) The luxI product is directly involved in signal synthesis-the protein is an acyl homoserine lactone synthase; and (ii) the substrates for acyl homoserine lactone synthesis are not amino acids from biosynthetic pathways or fatty acid degradation products, but rather they are S-adenosylmethionine (SAM) and an acylated acyl carrier protein (ACP) from the fatty acid biosynthesis pathway. We purified a maltose binding protein-LuxI fusion polypeptide and showed that, when provided with the appropriate substrates, it catalyzes the synthesis of an acyl homoserine lactone. In V. fischeri, luxi directs the synthesis of N-(3-oxohexanoyl) homoserine lactone and hexanoyl homoserine lactone. The purified maltose binding protein-LuxI fusion protein catalyzes the synthesis of hexanoyl homoserine lactone from hexanoyl-ACP and SAM. There is a high level of specificity for hexanoyl-ACP over ACPs with differing acyl group lengths, and hexanoyl homoserine lactone was not synthesized when SAM was replaced with other amino acids, such as methionine, S-adenosylhomocysteine, homoserine, or homoserine lactone, or when hexanoyl-SAM was provided as the substrate. This provides direct evidence that the LuxI protein is an auto-inducer synthase that catalyzes the formation of an amide bond between SAM and a fatty acyl-ACP and then catalyzes the formation of the acyl homoserine lactone from the acyl-SAM intermediate.

Clusters of charged residues in protein three-dimensional structures.

Statistically significant charge clusters (basic, acidic, or of mixed charge) in tertiary protein structures are identified by new methods from a large representative collection of protein structures. About 10% of protein structures show at least one charge cluster, mostly of mixed type involving about equally anionic and cationic residues. Positive charge clusters are very rare. Negative (or histidine-acidic) charge clusters often coordinate calcium, or magnesium or zinc ions [e.g., thermolysin (PDB code: 3tln), mannose-binding protein (2msb), aminopeptidase (1amp)]. Mixed-charge clusters are prominent at interchain contacts where they stabilize quaternary protein formation [e.g., glutathione S-transferase (2gst), catalase (8act), and fructose-1,6-bisphosphate aldolase (1fba)]. They are also involved in protein-protein interaction and in substrate binding. For example, the mixed-charge cluster of aspartate carbamoyl-transferase (8atc) envelops the aspartate carbonyl substrate in a flexible manner (alternating tense and relaxed states) where charge associations can vary from weak to strong. Other proteins with charge clusters include the P450 cytochrome family (BM-3, Terp, Cam), several flavocytochromes, neuraminidase, hemagglutinin, the photosynthetic reaction center, and annexin. In each case in Table 2 we discuss the possible role of the charge clusters with respect to protein structure and function.

Identification of interdomain sequences promoting the intronless evolution of a bacterial protein family.

In the evolution of eukaryotic genes, introns are believed to have played a major role in increasing the probability of favorable duplication events, chance recombinations, and exon shuffling resulting in functional hybrid proteins. As a rule, prokaryotic genes lack introns, and the examples of prokaryotic introns described do not seem to have contributed to gene evolution by exon shuffling. Still, certain protein families in modern bacteria evolve rapidly by recombination of genes, duplication of functional domains, and as shown for protein PAB of the anaerobic bacterial species Peptostreptococcus magnus, by the shuffling of an albumin-binding protein module from group C and G streptococci. Characterization of a protein PAB-related gene in a P. magnus strain with less albumin-binding activity revealed that the shuffled module was missing. Based on this fact and observations made when comparing gene sequences of this family of bacterial surface proteins interacting with albumin and/or immunoglobulin, a model is presented that can explain how this rapid intronless evolution takes place. A new kind of genetic element is introduced: the recer sequence promoting interdomain, in frame recombination and acting as a structure-less flexibility-promoting spacer in the corresponding protein. The data presented also suggest that antibiotics could represent the selective pressure behind the shuffling of protein modules in P. magnus, a member of the indigenous bacterial flora.

Purification, molecular cloning, and expression of the mammalian sigma1-binding site.

Sigma-ligands comprise several chemically unrelated drugs such as haloperidol, pentazocine, and ditolylguanidine, which bind to a family of low molecular mass proteins in the endoplasmic reticulum. These so-called sigma-receptors are believed to mediate various pharmacological effects of sigma-ligands by as yet unknown mechanisms. Based on their opposite enantioselectivity for benzomorphans and different molecular masses, two subtypes are differentiated. We purified the sigma1-binding site as a single 30-kDa protein from guinea pig liver employing the benzomorphan(+)[3H]pentazocine and the arylazide (-)[3H]azidopamil as specific probes. The purified (+)[3H]pentazocine-binding protein retained its high affinity for haloperidol, pentazocine, and ditolylguanidine. Partial amino acid sequence obtained after trypsinolysis revealed no homology to known proteins. Radiation inactivation of the pentazocine-labeled sigma1-binding site yielded a molecular mass of 24 +/- 2 kDa. The corresponding cDNA was cloned using degenerate oligonucleotides and cDNA library screening. Its open reading frame encoded a 25.3-kDa protein with at least one putative transmembrane segment. The protein expressed in yeast cells transformed with the cDNA showed the pharmacological characteristics of the brain and liver sigma1-binding site. The deduced amino acid sequence was structurally unrelated to known mammalian proteins but it shared homology with fungal proteins involved in sterol synthesis. Northern blots showed high densities of the sigma1-binding site mRNA in sterol-producing tissues. This is also in agreement with the known ability of sigma1-binding sites to interact with steroids, such as progesterone.

Bcl-2 interacting protein, BAG-1, binds to and activates the kinase Raf-1.

The Bcl-2 protein blocks programmed cell death (apoptosis) through an unknown mechanism. Previously we identified a Bcl-2 interacting protein BAG-1 that enhances the anti-apoptotic effects of Bcl-2. Like BAG-1, the serine/threonine protein kinase Raf-1 also can functionally cooperate with Bcl-2 in suppressing apoptosis. Here we show that Raf-1 and BAG-1 specifically interact in vitro and in yeast two-hybrid assays. Raf-1 and BAG-1 can also be coimmunoprecipitated from mammalian cells and from insect cells infected with recombinant baculoviruses encoding these proteins. Furthermore, bacterially-produced BAG-1 protein can increase the kinase activity of Raf-1 in vitro. BAG-1 also activates this mammalian kinase in yeast. These observations suggest that the Bcl-2 binding protein BAG-1 joins Ras and 14-3-3 proteins as potential activators of the kinase Raf-1.

Negative electrostatic surface potential of protein sites specific for anionic ligands.

Determination of the crystal structure of an "open" unliganded active mutant (T141D) form of the Escherichia coli phosphate receptor for active transport has allowed calculation of the electrostatic surface potential for it and two other comparably modeled receptor structures (wild type and D137N). A discovery of considerable implication is the intensely negative potential of the phosphate-binding cleft. We report similar findings for a sulfate transport receptor, a DNA-binding protein, and, even more dramatically, redox proteins. Evidently, for proteins such as these, which rely almost exclusively on hydrogen bonding for anion interactions and electrostatic balance, a noncomplementary surface potential is not a barrier to binding. Moreover, experimental results show that the exquisite specificity and high affinity of the phosphate and sulfate receptors for unions are insensitive to modulations of charge potential, but extremely sensitive to conditions that leave a hydrogen bond donor or acceptor unpaired.

The embryonic transcription factor stage specific activator protein contains a potent bipartite activation domain that interacts with several RNA polymerase II basal transcription factors.

Stage specific activator protein (SSAP) is a member of a newly discovered class of transcription factors that contain motifs more commonly found in RNA-binding proteins. Previously, we have shown that SSAP specifically binds to its recognition sequence in both the double strand and the single strand form and that this DNA-binding activity is localized to the N-terminal RNA recognition motif domain. Three copies of this recognition sequence constitute an enhancer element that is directly responsible for directing the transcriptional activation of the sea urchin late histone H1 gene at the midblastula stage of embryogenesis. Here we show that the remainder of the SSAP polypeptide constitutes an extremely potent bipartite transcription activation domain that can function in a variety of mammalian cell lines. This activity is as much as 3 to 5 times stronger than VP16 at activating transcription and requires a large stretch of amino acids that contain glutamine-glycine rich and serine-threonine-basic amino acid rich regions. We present evidence that SSAP's activation domain shares targets that are also necessary for activation by E1a and VP16. Finally, SSAP's activation domain is found to participate in specific interactions in vitro with the basal transcription factors TATA-binding protein, TFIIB, TFIIF74, and dTAF(II) 110.

Functional complementation of xeroderma pigmentosum complementation group E by replication protein A in an in vitro system.

Xeroderma pigmentosum (XP) is caused by a defect in nucleotide excision repair. Patients in the complementation group E (XP-E) have the mildest form of the disease and the highest level of residual repair activity. About 20% of the cell strains derived from XP-E patients lack a damaged DNA-binding protein (DDB) activity that binds to ultraviolet-induced (6-4) photoproducts with high affinity. We report here that cell-free extracts prepared from XP-E cell strains that either lacked or contained DDB activity were severely defective in excising DNA damage including (6-4) photoproducts. However, this excision activity defect was not restored by addition of purified DDB that, in fact, inhibited removal of (6-4) photoproducts by the human excision nuclease reconstituted from purified proteins. Extensive purification of correcting activity from HeLa cells revealed that the correcting activity is inseparable from the human replication/repair protein A [RPA (also known as human single stranded DNA binding protein, HSSB)]. Indeed, supplementing XP-E extracts with recombinant human RPA purified from Escherichia coli restored excision activity. However, no mutation was found in the genes encoding the three subunits of RPA in an XP-E (DDB-) cell line. It is concluded that RPA functionally complements XP-E extracts in vitro, but it is not genetically altered in XP-E patients.

G protein subunits and the stimulation of phospholipase C by Gs-and Gi-coupled receptors: Lack of receptor selectivity of Galpha(16) and evidence for a synergic interaction between Gbeta gamma and the alpha subunit of a receptor activated G protein.

Stimulatory guanine nucleotide binding protein (Gs)-coupled receptors activated by luteinizing hormone, vasopressin, and the catecholamine isoproterenol (luteinizing hormone receptor, type 2 vasopressin receptor, and types 1 and 2 beta-adrenergic receptors) and the Gi-coupled M2 muscarinic receptor (M2R) were expressed transiently in COS cells, alone and in combination with Gbeta gamma dimers, their corresponding Galphas (Galpha(s), or Galpha(i3)) and either Galpha(q) or Galpha(16). Phospholipase C (PLC) activity, assessed by inositol phosphate production from preincorporated myo[3H]inositol, was then determined to gain insight into differential coupling preferences among receptors and G proteins. The following were observed: (i) All receptors tested were able to stimulate PLC activity in response to agonist occupation. The effect of the M2R was pertussis toxin sensitive. (ii) While, as expected, expression of Galpha(q) facilitated an agonist-induced activation of PLC that varied widely from receptor to receptor (400% with type 2 vasopressin receptor and only 30% with M2R), expression of Galpha(16) facilitated about equally well the activation of PLC by any of the tested receptors and thus showed little if any discrimination for one receptor over another. (iii) Gbeta gamma elevated basal (agonist independent) PLC activity between 2- and 4-fold, confirming the proven ability of Gbeta gamma to stimulate PLCbeta. (iv) Activation of expressed receptors by their respective ligands in cells coexpressing excess Gbeta gamma elicited agonist stimulated PLC activities, which, in the case of the M2R, was not blocked by pertussis toxin (PTX), suggesting mediation by a PTX-insensitive PLC-stimulating Galpha subunit, presumably, but not necessarily, of the Gq family. (v) The effects of Gbeta gamma and the PTX-insensitive Galpha elicited by M2R were synergistic, suggesting the possibility that one or more forms of PLC are under conditional or dual regulation of G protein subunits such that stimulation by one sensitizes to the stimulation by the other.

Recombinant human uteroglobin suppresses cellular invasiveness via a novel class of high-affinity cell surface binding site.

The mechanism(s) that regulates invasion of trophoblasts through the uterine epithelium during embryo implantation and nidation in hemochorial placental mammals is poorly understood. While limited trophoblast invasion is essential for the establishment of normal pregnancy, dysregulation of this process may contribute to the pathogenesis of choriocarcinoma, a highly invasive and lethal form of cancer arising from the trophoblasts. We have previously demonstrated that rabbit uteroglobin (UG), a cytokine-like, antiinflammatory protein, produced by the endometrial epithelium during pregnancy, has a potent antichemotactic effect on neutrophils and monocytes in vitro. Here, we report that recombinant human UG (hUG) dramatically suppresses invasion of human trophoblasts and NIH 3T3 cells through an artificial basement membrane (Matrigel) in vitro but has no effect on that of human choriocarcinoma cells. We identified a previously unreported high-affinity, high molecular weight (approximately 190 kDa), nonglycosylated hUG-binding protein, readily detectable on human trophoblasts and NIH 3T3 cells but totally lacking on choriocarcinoma cells. Taken together, these results raise the possibility that (i) hUG plays a critical role in regulating cellular invasiveness, at least in part, via its previously unrecognized cell surface binding site, and (ii) some of the numerous biological activities of proteins of the UG family, reported so far, may be mediated via this binding site.

In vivo selection of basic region-leucine zipper proteins with altered DNA-binding specificities.

A transcription interference assay was used to generate mutant basic region-leucine zipper proteins with altered DNA-binding specificities. A library of mutants of a CCAAT/enhancer binding protein was constructed by randomizing five DNA-contacting amino acids in the basic region Asn-18, Ala-15, Val-14, Ser-11, and Arg-10. These mutants were then selected for their ability to bind mutant recognition sequences containing substitutions at the 2 and 3 positions of the wild-type sequence 5'-A5T4T3G2C1G1'C2'A3A4'T5'-3'. Mutants containing the sequence Leu-18Tyr-15Xaa-14Tyr-11Arg-10, in which four of the five contact residues are altered, were identified that recognize the palindromic sequence 5'-ATCYCGY'GAT-3' (Xaa = asparagine when Y = G; Xaa = methionine when Y = A). Moreover, in a selection against the sequence 5'-ATTACGTAAT-3', mutants were obtained containing substitutions not only in the basic region but also in the hinge region between the basic and leucine zipper regions. The mutant proteins showed high specificity in a functional transcription interference assay. A model for the interaction of these mutants with the target DNA sequences is discussed.

Protein kinase cross-talk: membrane targeting of the beta-adrenergic receptor kinase by protein kinase C.

The beta-adrenergic receptor kinase (betaARK) is the prototypical member of the family of cytosolic kinases that phosphorylate guanine nucleotide binding-protein-coupled receptors and thereby trigger uncoupling between receptors and guanine nucleotide binding proteins. Herein we show that this kinase is subject to phosphorylation and regulation by protein kinase C (PKC). In cell lines stably expressing alpha1B- adrenergic receptors, activation of these receptors by epinephrine resulted in an activation of cytosolic betaARK. Similar data were obtained in 293 cells transiently coexpressing alpha1B- adrenergic receptors and betaARK-1. Direct activation of PKC with phorbol esters in these cells caused not only an activation of cytosolic betaARK-1 but also a translocation of betaARK immunoreactivity from the cytosol to the membrane fraction. A PKC preparation purified from rat brain phospborylated purified recombinant betaARK-1 to a stoichiometry of 0.86 phosphate per betaARK-1. This phosphorylation resulted in an increased activity of betaARK-1 when membrane-bound rhodopsin served as its substrate but in no increase of its activity toward a soluble peptide substrate. The site of phosphorylation was mapped to the C terminus of betaARK-1. We conclude that PKC activates betaARK by enhancing its translocation to the plasma membrane.

Protein-protein interactions in eukaryotic transcription initiation: structure of the preinitiation complex.

We have used alanine scanning to analyze protein-protein interactions by human TATA-element binding protein (TBP) within the transcription preinitiation complex. The results indicate that TBP interacts with RNA polymerase II and general transcription factors IIA, IIB, and IIF within the functional transcription preinitiation complex and define the determinants of TBP for each of these interactions. The results permit construction of a model for the structure of the preinitiation complex.