SMART, a simple modular architecture research tool: Identification of signaling domains

Accurate multiple alignments of 86 domains that occur in signaling proteins have been constructed and used to provide a Web-based tool (SMART: simple modular architecture research tool) that allows rapid identification and annotation of signaling domain sequences. The majority of signaling proteins are multidomain in character with a considerable variety of domain combinations known. Comparison with established databases showed that 25% of our domain set could not be deduced from SwissProt and 41% could not be annotated by Pfam. SMART is able to determine the modular architectures of single sequences or genomes; application to the entire yeast genome revealed that at least 6.7% of its genes contain one or more signaling domains, approximately 350 greater than previously annotated. The process of constructing SMART predicted (i) novel domain homologues in unexpected locations such as band 4.1-homologous domains in focal adhesion kinases; (ii) previously unknown domain families, including a citron-homology domain; (iii) putative functions of domain families after identification of additional family members, for example, a ubiquitin-binding role for ubiquitin-associated domains (UBA); (iv) cellular roles for proteins, such predicted DEATH domains in netrin receptors further implicating these molecules in axonal guidance; (v) signaling domains in known disease genes such as SPRY domains in both marenostrin/pyrin and Midline 1; (vi) domains in unexpected phylogenetic contexts such as diacylglycerol kinase homologues in yeast and bacteria; and (vii) likely protein misclassifications exemplified by a predicted pleckstrin homology domain in a Candida albicans protein, previously described as an integrin.

A TRP homolog in Saccharomyces cerevisiae forms an intracellular Ca2+-permeable channel in the yeast vacuolar membrane

The molecular identification of ion channels in internal membranes has made scant progress compared with the study of plasma membrane ion channels. We investigated a prominent voltage-dependent, cation-selective, and calcium-activated vacuolar ion conductance of 320 pS (yeast vacuolar conductance, YVC1) in Saccharomyces cerevisiae. Here we report on a gene, the deduced product of which possesses significant homology to the ion channel of the transient receptor potential (TRP) family. By using a combination of gene deletion and re-expression with direct patch clamping of the yeast vacuolar membrane, we show that this yeast TRP-like gene is necessary for the YVC1 conductance. In physiological conditions, tens of micromolar cytoplasmic Ca2+ activates the YVC1 current carried by cations including Ca2+ across the vacuolar membrane. Immunodetection of a tagged YVC1 gene product indicates that YVC1 is primarily localized in the vacuole and not other intracellular membranes. Thus we have identified the YVC1 vacuolar/lysosomal cation-channel gene. This report has implications for the function of TRP channels in other organisms and the possible molecular identification of vacuolar/lysosomal ion channels in other eukaryotes.

Modular organization of intrinsic connections associated with spectral tuning in cat auditory cortex

Many response properties in primary auditory cortex (AI) are segregated spatially and organized topographically as those in primary visual cortex. Intensive study has not revealed an intrinsic, anatomical organizing principle related to an AI functional topography. We used retrograde anatomic tracing and topographic physiologic mapping of acoustic response properties to reveal long-range (≥1.5 mm) convergent intrinsic horizontal connections between AI subregions with similar bandwidth and characteristic frequency selectivity. This suggests a modular organization for processing spectral bandwidth in AI.

Rad52 forms DNA repair and recombination centers during S phase

Maintenance of genomic integrity and stable transmission of genetic information depend on a number of DNA repair processes. Failure to faithfully perform these processes can result in genetic alterations and subsequent development of cancer and other genetic diseases. In the eukaryote Saccharomyces cerevisiae, homologous recombination is the major pathway for repairing DNA double-strand breaks. The key role played by Rad52 in this pathway has been attributed to its ability to seek out and mediate annealing of homologous DNA strands. In this study, we find that S. cerevisiae Rad52 fused to green fluorescent protein (GFP) is fully functional in DNA repair and recombination. After induction of DNA double-strand breaks by γ-irradiation, meiosis, or the HO endonuclease, Rad52-GFP relocalizes from a diffuse nuclear distribution to distinct foci. Interestingly, Rad52 foci are formed almost exclusively during the S phase of mitotic cells, consistent with coordination between recombinational repair and DNA replication. This notion is further strengthened by the dramatic increase in the frequency of Rad52 focus formation observed in a pol12-100 replication mutant and a mec1 DNA damage checkpoint mutant. Furthermore, our data indicate that each Rad52 focus represents a center of recombinational repair capable of processing multiple DNA lesions.

A modular misexpression screen in Drosophila detecting tissue-specific phenotypes.

Genetic screens in Drosophila have lead to the discovery of many genes important for patterning and signal transduction in diverse organisms. Traditionally, the phenotypic effects of loss-of-function mutations are analyzed. As an alternative way to link genes and function, I have developed a versatile misexpression screen in Drosophila, the first such screen in higher eukaryotes. The screen identifies genes that, when over- or misexpressed in a pattern of interest, give a specific phenotype or modulate an existing mutant phenotype. It is based on Gal4 transactivation of a mobile enhancer and promoter that "targets" random endogenous genes for expression. The modular design of the screen allows directed expression in any temporal or spatial pattern. When activated in the developing eye, 4% of target inserts gave dominant phenotypes. One insertion was in the gene encoding Ras GTPase-activating protein; its overexpression phenotype was strongly enhanced by a mutation in Ras1. Thus, biologically relevant phenotypes and genetic interactions are identified using this method. The screen is a powerful new tool for developmental genetics; similar approaches can also be developed for other organisms.

Alternatively spliced forms in the cytoplasmic domain of the human growth hormone (GH) receptor regulate its ability to generate a soluble GH-binding protein.

The mechanism underlying the generation of soluble growth hormone binding protein (GHBP) probably differs among species. In rats and mice, it involves an alternatively spliced mRNA, whereas in rabbits, it involves limited proteolysis of the membrane-bound growth hormone receptor (GHR). In humans, this latter mechanism is favored, as no transcript coding for a soluble GHR has been detected so far. To test this hypothesis, we analyzed COS-7 cells transiently expressing the full-length human (h) GHR and observed specific GH-binding activity in the cell supernatants. Concomitantly, an alternatively spliced form in the cytoplasmic domain of GHR, hGHR-tr, was isolated from several human tissues. hGHR-tr is identical in sequence to hGHR, except for a 26-bp deletion leading to a stop codon at position 280, thereby truncating 97.5% of the intracellular domain of the receptor protein. When compared with hGHR, hGHR-tr showed a significantly increased capacity to generate a soluble GHBP. Interestingly, this alternative transcript is also expressed in liver from rabbits, mice, and rats, suggesting that, in these four species, proteolysis of the corresponding truncated transmembrane GHR is a common mechanism leading to GHBP generation. These findings support the hypothesis that GHBP may at least partly result from alternative splicing of the region encoding the intracellular domain and that the absence of a cytoplasmic domain may be involved in increased release of GHBP.

Two new forms of gonadotropin-releasing hormone in a protochordate and the evolutionary implications.

The neuropeptide gonadotropin-releasing hormone (GnRH) is the major regulator of reproduction in vertebrates. Our goal was to determine whether GnRH could be isolated and identified by primary structure in a protochordate and to examine its location by immunocytochemistry. The primary structure of two novel decapeptides from the tunicate Chelyosoma productum (class Ascidiacea) was determined. Both show significant identity with vertebrate GnRH. Tunicate GnRH-I (pGlu-His-Trp-Ser-Asp-Tyr-Phe-Lys-Pro-Gly-NH2) has 60% of its residues conserved, compared with mammalian GnRH, whereas tunicate GnRH-II (pGlu-His-Trp-Ser-Leu-Cys-His-Ala-Pro-Gly-NH2) is unusual in that it was isolated as a disulfide-linked dimer. Numerous immunoreactive GnRH neurons lie within blood sinuses close to the gonoducts and gonads in both juveniles and adults, implying that the neuropeptide is released into the bloodstream. It is suggested that in ancestral chordates, before the evolution of the pituitary, the hormone was released into the bloodstream and acted directly on the gonads.

Modular cis-regulatory organization of developmentally expressed genes: two genes transcribed territorially in the sea urchin embryo, and additional examples.

The cis-regulatory systems that control developmental expression of two sea urchin genes have been subjected to detailed functional analysis. Both systems are modular in organization: specific, separable fragments of the cis-regulatory DNA each containing multiple transcription factor target sites execute particular regulatory subfunctions when associated with reporter genes and introduced into the embryo. The studies summarized here were carried out on the CyIIIa gene, expressed in the embryonic aboral ectoderm and on the Endo16 gene, expressed in the embryonic vegetal plate, archenteron, and then midgut. The regulatory systems of both genes include modules that control particular aspects of temporal and spatial expression, and in both the territorial boundaries of expression depend on a combination of negative and positive functions. In both genes different regulatory modules control early and late embryonic expression. Modular cis-regulatory organization is widespread in developmentally regulated genes, and we present a tabular summary that includes many examples from mouse and Drosophila. We regard cis-regulatory modules as units of developmental transcription control, and also of evolution, in the assembly of transcription control systems.

Multiple forms of complement C3 in trout that differ in binding to complement activators.

In all other species analyzed to date, the functionally active form of complement component C3 exists as the product of a single gene. We have now identified and characterized three functional C3 proteins (C3-1, C3-3, and C3-4) in trout that are the products of at least two distinct C3 genes. All three proteins are composed of an alpha-and a beta-chain and contain a thioester bond in the alpha-chain. However, they differ in their electrophoretic mobility, glycosylation, reactivity with monospecific C3 antibodies, and relative ability to bind to various surfaces (zymosan, Escherichia coli, erythrocytes). A comparison of the partial amino acid sequences of the three proteins showed that the amino acid sequence identity/similarity of C3-3 to C3-4 is 87/91%, while that of C3-3 and C3-4 to C3-1 is 51.5/65.5% and 60/73% respectively. Thus, trout possess multiple forms of functional C3 that represent the products of several distinct genes and differ in their ability to bind covalently to various complement activators.

Molecular basis for dysfunction of some mutant forms of methylmalonyl-CoA mutase: deductions from the structure of methionine synthase.

Inherited defects in the gene for methylmalonyl-CoA mutase (EC 5.4.99.2) result in the mut forms of methylmalonic aciduria. mut- mutations lead to the absence of detectable mutase activity and are not corrected by excess cobalamin, whereas mut- mutations exhibit residual activity when exposed to excess cobalamin. Many of the mutations that cause methylmalonic aciduria in humans affect residues in the C-terminal region of the methylmalonyl-CoA mutase. This portion of the methylmalonyl-CoA mutase sequence can be aligned with regions in other B12 (cobalamin)-dependent enzymes, including the C-terminal portion of the cobalamin-binding region of methionine synthase. The alignments allow the mutations of human methylmalonyl-CoA mutase to be mapped onto the structure of the cobalamin-binding fragment of methionine synthase from Escherichia coli (EC 2.1.1.13), which has recently been determined by x-ray crystallography. In this structure, the dimethylbenzimidazole ligand to the cobalt in free cobalamin has been displaced by a histidine ligand, and the dimethylbenzimidazole nucleotide "tail" is thrust into a deep hydrophobic pocket in the protein. Previously identified mut0 and mut- mutations (Gly-623 --> Arg, Gly-626 --> Cys, and Gly-648 --> Asp) of the mutase are predicted to interfere with the structure and/or stability of the loop that carries His-627, the presumed lower axial ligand to the cobalt of adenosylcobalamin. Two mutants that lead to severe impairment (mut0) are Gly-630 --> Glu and Gly-703 --> Arg, which map to the binding site for the dimethylbenzimidazole nucleotide substituent of adenosylcobalamin. The substitution of larger residues for glycine is predicted to block the binding of adenosylcobalamin.

Pea formaldehyde-active class III alcohol dehydrogenase: common derivation of the plant and animal forms but not of the corresponding ethanol-active forms (classes I and P).

A plant class III alcohol dehydrogenase (or glutathione-dependent formaldehyde dehydrogenase) has been characterized. The enzyme is a typical class III member with enzymatic parameters and substrate specificity closely related to those of already established animal forms. Km values with the pea enzyme are 6.5 microM for NAD+, 2 microM for S-hydroxymethylglutathione, and 840 microM for octanol versus 9, 4, and 1200 microM, respectively, with the human enzyme. Structurally, the pea/human class III enzymes are closely related, exhibiting a residue identity of 69% and with only 3 of 23 residues differing among those often considered in substrate and coenzyme binding. In contrast, the corresponding ethanol-active enzymes, the long-known human liver and pea alcohol dehydrogenases, differ more (47% residue identities) and are also in functionally important active site segments, with 12 of the 23 positions exchanged, including no less than 7 at the usually much conserved coenzyme-binding segment. These differences affect functionally important residues that are often class-distinguishing, such as those at positions 48, 51, and 115, where the plant ethanol-active forms resemble class III (Thr, Tyr, and Arg, respectively) rather than the animal ethanol-active class I forms (typically Ser, His, and Asp, respectively). Calculations of phylogenetic trees support the conclusions from functional residues in subgrouping plant ethanol-active dehydrogenases and the animal ethanol-active enzymes (class I) as separate descendants from the class III line. It appears that the classical plant alcohol dehydrogenases (now called class P) have a duplicatory origin separate from that of the animal class I enzymes and therefore a paralogous relationship with functional convergence of their alcohol substrate specificity. Combined, the results establish the conserved nature of class III also in plants, and contribute to the molecular and functional understanding of alcohol dehydrogenases by defining two branches of plant enzymes into the system.

Electrical microstimulation suggests two different forms of representation of head-centered space in the intraparietal sulcus of rhesus monkeys.

We examined the effects of eye position on saccades evoked by electrical stimulation of the intraparietal sulcus (IPS) of rhesus monkeys. Microstimulation evoked saccades from sites on the posterior bank, floor, and the medial bank of the IPS. The size and direction of the eye movements varied as a function of initial eye position before microstimulation. At many stimulation sites, eye position affected primarily the amplitude and not the direction of the evoked saccades. These "modified vector saccades" were characteristic of most stimulation-sensitive zones in the IPS, with the exception of a narrow strip located mainly on the floor of the sulcus. Stimulation in this "intercalated zone" evoked saccades that moved the eyes into a particular region in head-centered space, independent of the starting position of the eyes. This latter response is compatible with the stimulation site representing a goal zone in head-centered coordinates. On the other hand, the modified vector saccades observed outside the intercalated zone are indicative of a more distributed representation of head-centered space. A convergent projection from many modified vector sites onto each intercalated site may be a basis for a transition from a distributed to a more explicit representation of space in head-centered coordinates.

Core binding factor beta-smooth muscle myosin heavy chain chimeric protein involved in acute myeloid leukemia forms unusual nuclear rod-like structures in transformed NIH 3T3 cells.

Patients with the M4Eo subtype of acute myeloid leukemia almost invariably are found to have an inversion of chromosome 16 in their leukemic cells, which results in a gene fusion between the transcription factor called core binding factor beta (CBFbeta) on 16q and a smooth muscle myosin heavy chain (SMMHC) gene on 16p. Subcellular localizations of the wild-type CBFbeta and the CBFbeta-SMMHC fusion protein were determined by immunofluorescence of NIH 3T3 cells that overexpress wild-type or fusion protein. Normal CBFbeta showed an unexpected perinuclear pattern consistent with primary localization in the Golgi complex. The CBFbeta-SMMHC fusion protein had a very different pattern. Nuclear staining included rod-like crystalline structures as long as 11 microm. The heterodimeric partner of CBFbeta, CBFalpha, formed part of this complex. Cytoplasmic staining included stress fibers that colocalized with actin, probably as a consequence of the myosin heavy chain component of the fusion protein. Deletion of different regions of the CBFbeta portion of the fusion protein showed that binding to CBFalpha was not required for nuclear translocation. However, deletion of parts of the SMMHC domain of the fusion protein involved in myosin-mediated filament formation resulted in proteins that did not form rod-like structures. These observations confirm previous indirect evidence that the CBFbeta-SMMHC fusion protein is capable of forming macromolecular nuclear aggregates and suggests possible models for the mechanism of leukemic transformation.

Sequence-selective carbohydrate-DNA interaction: dimeric and monomeric forms of the calicheamicin oligosaccharide interfere with transcription factor function.

The synthetic oligosaccharide moiety of the antibiotic calicheamicin and the head-to-head dimer of this oligosaccharide are known to bind to the minor groove of DNA in a sequence-selective manner preferring distinct target sequences. We tested these carbohydrates for their ability to interfere with transcription factor function. The oligosaccharides inhibit binding of transcription factors to DNA in a sequence-selective manner, probably by inducing a conformational change in DNA structure. They also interfere with transcription by polymerase II in vitro. The effective concentrations of the oligosaccharides for inhibition of transcription factor binding and for transcriptional inhibition are in the micromolar range. The dimer is a significantly more active inhibitor than is the monomer.

Vpr protein of human immunodeficiency virus type 1 forms cation-selective channels in planar lipid bilayers.

A small (96-aa) protein, virus protein R (Vpr), of human immunodeficiency virus type 1 contains one hydrophobic segment that could form a membrane-spanning helix. Recombinant Vpr, expressed in Escherichia coli and purified by affinity chromatography, formed ion channels in planar lipid bilayers when it was added to the cis chamber and when the trans chamber was held at a negative potential. The channels were more permeable to Na+ than to Cl- ions and were inhibited when the trans potential was made positive. Similar channel activity was caused by Vpr that had a truncated C terminus, but the potential dependence of channel activity was no longer seen. Antibody raised to a peptide mimicking part of the C terminus of Vpr (AbC) inhibited channel activity when added to the trans chamber but had no effect when added to the cis chamber. Antibody to the N terminus of Vpr (AbN) increased channel activity when added to the cis chamber but had no effect when added to the trans chamber. The effects of potential and antibodies on channel activity are consistent with a model in which the positive C-terminal end of dipolar Vpr is induced to traverse the bilayer membrane when the opposite (trans) side of the membrane is at a negative potential. The C terminus of Vpr would then be available for interaction with AbC in the trans chamber, and the N terminus would be available for interaction with AbN in the cis chamber. The ability of Vpr to form ion channels in vitro suggests that channel formation by Vpr in vivo is possible and may be important in the life cycle of human immunodeficiency virus type 1 and/or may cause changes in cells that contribute to AIDS-related pathologies.