Crystal structure of the catalytic domain of HIV-1 integrase: similarity to other polynucleotidyl transferases.

HIV integrase is the enzyme responsible for inserting the viral DNA into the host chromosome; it is essential for HIV replication. The crystal structure of the catalytically active core domain (residues 50 to 212) of HIV-1 integrase was determined at 2.5 A resolution. The central feature of the structure is a five-stranded beta sheet flanked by helical regions. The overall topology reveals that this domain of integrase belongs to a superfamily of polynucleotidyl transferases that includes ribonuclease H and the Holliday junction resolvase RuvC. The active site region is identified by the position of two of the conserved carboxylate residues essential for catalysis, which are located at similar positions in ribonuclease H. In the crystal, two molecules form a dimer with a extensive solvent-inaccessible interface of 1300 A2 per monomer.

Identification and characterization of JunD missense mutants that lack menin binding.

Menin, the product of the MEN1 tumor suppressor gene, binds to the AP1 transcription factor JunD and represses JunD transcriptional activity. The effects of human or mouse JunD missense mutations upon menin interaction were studied by random and alanine scanning mutagenesis of the menin binding region of JunD (amino acids 1-70). JunD mutant proteins were tested for menin binding in a reverse yeast two-hybrid assay, and for transcriptional regulation by menin in AP1-reporter assays. Random mutagenesis identified two different mutations that disrupted menin interaction at mouse JunD amino acid 42 (G42E and G42R). Mutation G42A generated by alanine scanning did not affect menin binding, likely reflecting the conserved nature of this amino acid substitution. Furthermore, by size exclusion chromatography menin co-migrated with wild type JunD but not with the JunD mutant tested (G42E). Alanine scanning mutagenesis of residues 30-55 revealed two different amino acids, P41 and P44, of mouse JunD that were critical for interaction with menin. Mouse JunD missense mutants P41A, G42R, G42E and P44A failed to bind menin and also escaped menin's control over their transcriptional activity. At lower amounts of transfected menin, the transcriptional effect of menin on the mutants P41A, G42R and G42E was changed from repression to activation, similar to that with c-jun. In conclusion, a small N-terminal region of JunD mediates a key difference between JunD and c-jun, and a component of this difference is dependent on JunD binding to menin.

Menin molecular interactions: insights into normal functions and tumorigenesis.

Multiple endocrine neoplasia type 1 (MEN1) is an autosomal dominant disease predisposed by heterozygous germline mutations in the MEN1 tumor suppressor gene. Biallelic loss of MEN1 resulting from small mutation and/or loss of heterozygosity occurs in a large tissue spectrum of MEN1 tumors or non-hereditary tumors. Mouse models of MEN1 underexpression or overexpression have also supported the tumor-suppressor effect of the MEN1 gene. Menin, the 610-amino-acid protein encoded by MEN1, is expressed ubiquitously and found predominantly in the nucleus. Sequence analyses do not reveal motifs of known function other than two nuclear localization sequences. Menin has been found to partner in vitro with a variety of proteins that comprise transcription factors, DNA processing factors, DNA repair proteins, and cytoskeletal proteins. The diverse functions of menin interactors suggest roles for menin in multiple biological pathways. Inactivation of menin switches its JunD partner from a downstream action of growth suppression to growth promotion. This is a plausible mechanism for menin tumorigenesis.

Principles of macromolecular X-ray crystallography.

X-ray crystallography is one of the major tools available for protein structural analysis. This unit provides an introductory review of the principles of X-ray crystallography that covers how the image is generated and analyzed. The second half of the unit describes strategies for producing crystals of protein, including methods for dealing with proteins that do not easily form crystals.

The core and carboxyl-terminal domains of the integrase protein of human immunodeficiency virus type 1 each contribute to nonspecific DNA binding.

The integrase protein of human immunodeficiency virus type 1 removes two nucleotides from the 3' ends of reverse-transcribed human immunodeficiency virus type 1 DNA (3' processing) and covalently inserts the processed ends into a target DNA (DNA strand transfer). Mutant integrase proteins that lack the amino-and/or carboxyl-terminal domains are incapable of catalyzing 3' processing and DNA strand transfer but are competent for an apparent reversal of the DNA strand transfer reaction (disintegration) in vitro. Here, we investigate the binding of integrase to DNA by UV cross-linking. Cross-linked complexes form with a variety of DNA substrates independent of the presence of divalent metal ion. Analysis with amino- and carboxyl-terminal deletion mutant proteins shows that residues 213 to 266 of the 288-residue protein are required for efficient cross-linking in the absence of divalent metal ion. Carboxyl-terminal deletion mutants that lack this region efficiently cross-link only to the branched disintegration DNA substrate, and this reaction is dependent on the presence of metal ion. Both the core and C-terminal domains of integrase therefore contribute to nonspecific DNA binding.

Heterogeneity in recombinant HIV-1 integrase corrected by site-directed mutagenesis: the identification and elimination of a protease cleavage site.

Purified recombinant human immunodeficiency virus type 1 (HIV-1) integrase and certain deletion mutants exhibit heterogeneity consistent with proteolysis at a site close to the C-terminus. Electrospray ionization mass spectrometric analysis indicated that proteolytic cleavage generated a protein missing five residues from the C-terminus. PCR mutagenesis of amino acids on either side of the cleavage site identified two changes which were subsequently shown to prevent clipping when proteins were expressed and purified from Escherichia coli: the substitution of Arg284, the residue on the C-terminal side of the cleavage site, by either glycine or lysine. The introduction of either of these mutations into full-length integrase did not affect in vitro 3' processing or strand transfer activities. Thus, the incorporation of either of these mutations is likely to be beneficial when homogeneity of HIV-1 integrase is a concern, as in crystallographic or nuclear magnetic resonance spectroscopic experiments.

GCN5-related N-acetyltransferases: a structural overview.

Hundreds of acetyltransferases exist. All use a common acetyl donor--acetyl coenzyme A--and each exhibits remarkable specificity for acetyl acceptors, which include small molecules and proteins. Analysis of the primary sequences of these enzymes indicates that they can be sorted into several superfamilies. This review covers the three-dimensional structures of members of one of these superfamilies, now referred to in the literature as the GCN5-related N-acetyltransferases (GNAT), reflecting the importance of one functional category, the histone acetyltransferases. Despite the diversity of substrate specificities, members of the GNAT superfamily demonstrate remarkable similarity in protein topology and mode of acetyl coenzyme A binding, likely reflecting a conserved catalytic mechanism.

Melatonin biosynthesis: the structure of serotonin N-acetyltransferase at 2.5 A resolution suggests a catalytic mechanism.

Conversion of serotonin to N-acetylserotonin, the precursor of the circadian neurohormone melatonin, is catalyzed by serotonin N-acetyltransferase (AANAT) in a reaction requiring acetyl coenzyme A (AcCoA). AANAT is a globular protein consisting of an eight-stranded beta sheet flanked by five alpha helices; a conserved motif in the center of the beta sheet forms the cofactor binding site. Three polypeptide loops converge above the AcCoA binding site, creating a hydrophobic funnel leading toward the cofactor and serotonin binding sites in the protein interior. Two conserved histidines not found in other NATs are located at the bottom of the funnel in the active site, suggesting a catalytic mechanism for acetylation involving imidazole groups acting as general acid/base catalysts.

The structural basis of ordered substrate binding by serotonin N-acetyltransferase: enzyme complex at 1.8 A resolution with a bisubstrate analog.

Serotonin N-acetyltransferase, a member of the GNAT acetyltransferase superfamily, is the penultimate enzyme in the conversion of serotonin to melatonin, the circadian neurohormone. Comparison of the structures of the substrate-free enzyme and the complex with a bisubstrate analog, coenzyme A-S-acetyltryptamine, demonstrates that acetyl coenzyme A (AcCoA) binding is accompanied by a large conformational change that in turn leads to the formation of the serotonin-binding site. The structure of the complex also provides insight into how the enzyme may facilitate acetyl transfer. A water-filled channel leading from the active site to the surface provides a pathway for proton removal following amine deprotonation. Furthermore, structural and mutagenesis results indicate an important role for Tyr-168 in catalysis.

A new method for extraction of iron-molybdenum cofactor (FeMoco) from nitrogenase adsorbed to DEAE-cellulose. 2. Solubilization of FeMoco in a wide range of organic solvents.

While the iron-molybdenum cofactor (FeMoco) of nitrogenase, a constituent of the active site for nitrogen reduction, can be extracted into N-methylformamide (NMF) and pyrrollidinone, the inability to solubilize it in any other organic solvents has hampered further understanding of its structure and chemical properties. A method to solubilize FeMoco, prepared in N,N-dimethylformamide (DMF) with Bu4N+ as counterion [McLean, P. A., Wink, D. A., Chapman, S. K., Hickman, A. B., McKillop, D. M., & Orme-Johnson, W. H. (1989) Biochemistry (preceding paper in this issue)], in acetonitrile, acetone, methylene chloride, tetrahydrofuran, and benzene is reported. FeMoco evaporated to dryness in vacuo dissolves readily in good yield (55-100%) and with no significant loss in specific activity. In addition, FeMoco can be extracted directly into these solvents from MoFe protein bound to a DEAE-Sepharose column if the protein is pretreated with DMF. Methods have also been developed to extract fully active FeMoco into acetone and acetonitrile in the absence of any amide solvents (NMF or DMF). Extraction of FeMoco into acetone (30% yield) involves only pretreatment of column-bound protein with methanol, while extraction into acetonitrile (22% yield) requires pretreatment with methanol followed by THF. We conclude that the presence of a suitable soluble cation confers solubility to the cofactor in many common organic solvents and that the solubility of FeMoco in a given solvent may be independent of the ability of that solvent to extract the cofactor from column-bound protein.

Biophysical and enzymatic properties of the catalytic domain of HIV-1 integrase.

A deletion derivative of the integrase protein from human immunodeficiency virus type-1 (HIV-1) consisting of the central core domain (amino acids 50-212) has been characterized biophysically and biochemically. This deletion mutant is of particular interest for structural studies as it can carry out the disintegration reaction suggesting the presence of an active site and, under certain conditions, is more soluble than full-length integrase. The circular dichroism and fluorescence of the deletion mutant and the 288-residue full-length integrase were similar, indicating that the core residues maintain similar overall conformations in both proteins. The deletion mutant is approximately 10% more alpha-helical than the full-length protein. Analytical centrifugation demonstrated that both proteins undergo monomer-dimer association although the truncated protein showed slightly less tendency to dimerize; the dissociation constants were 2.5 x 10(-5) M for the full-length protein and 8.0 x 10(-5) M for the truncated protein. The disintegration activity of both proteins was also compared. Although a higher concentration of the truncation mutant was required for optimal activity, the mutant did not have altered pH or Mn2+ requirements relative to the full-length protein. The combined biophysical and enzymatic studies suggest that this truncated form of HIV-1 integrase is likely to be useful for structural studies.

Molecular organization in site-specific recombination: the catalytic domain of bacteriophage HP1 integrase at 2.7 A resolution.

HP1 integrase promotes site-specific recombination of the HP1 genome into that of Haemophilus influenzae. The isolated C-terminal domain (residues 165-337) of the protein interacts with the recombination site and contains the four catalytic residues conserved in the integrase family. This domain represents a novel fold consisting principally of well-packed alpha helices, a surface beta sheet, and an ordered 17-residue C-terminal tail. The conserved triad of basic residues and the active-site tyrosine are contributed by a single monomer and occupy fixed positions in a defined active-site cleft. Dimers are formed by mutual interactions of the tail of one monomer with an adjacent monomer; this orients active-site clefts antiparallel to each other.

Unexpected structural diversity in DNA recombination: the restriction endonuclease connection.

Transposition requires a coordinated series of DNA breakage and joining reactions. The Tn7 transposase contains two proteins: TnsA, which carries out DNA breakage at the 5' ends of the transposon, and TnsB, which carries out breakage and joining at the 3' ends of the transposon. TnsB is a member of the retroviral integrase superfamily whose hallmark is a conserved DDE motif. We report here the structure of TnsA at 2.4 A resolution. Surprisingly, the TnsA fold is that of a type II restriction endonuclease. Thus, Tn7 transposition involves a collaboration between polypeptides, one containing a DDE motif and one that does not. This result indicates that the range of biological processes that utilize restriction enzyme-like folds also includes DNA transposition.

Three new structures of the core domain of HIV-1 integrase: an active site that binds magnesium.

HIV-1 integrase is an essential enzyme in the life cycle of the virus, responsible for catalyzing the insertion of the viral genome into the host cell chromosome; it provides an attractive target for antiviral drug design. The previously reported crystal structure of the HIV-1 integrase core domain revealed that this domain belongs to the superfamily of polynucleotidyltransferases. However, the position of the conserved catalytic carboxylic acids differed from those observed in other enzymes of the class, and attempts to crystallize in the presence of the cofactor, Mg2+, were unsuccessful. We report here three additional crystal structures of the core domain of HIV-1 integrase mutants, crystallized in the presence and absence of cacodylate, as well as complexed with Mg2+. These three crystal forms, containing between them seven independent core domain structures, demonstrate the unambiguous extension of the previously disordered helix alpha4 toward the amino terminus from residue M154 and show that the catalytic E152 points in the general direction of the two catalytic aspartates, D64 and D116. In the vicinity of the active site, the structure of the protein in the absence of cacodylate exhibits significant deviations from the previously reported structures. These differences can be attributed to the modification of C65 and C130 by cacodylate, which was an essential component of the original crystallization mixture. We also demonstrate that in the absence of cacodylate this protein will bind to Mg2+, and could provide a satisfactory platform for binding of inhibitors.

Catalytic domain of human immunodeficiency virus type 1 integrase: identification of a soluble mutant by systematic replacement of hydrophobic residues.

The integrase protein of human immunodeficiency virus type 1 is necessary for the stable integration of the viral genome into host DNA. Integrase catalyzes the 3' processing of the linear viral DNA and the subsequent DNA strand transfer reaction that inserts the viral DNA ends into host DNA. Although full-length integrase is required for 3' processing and DNA strand transfer activities in vitro, the central core domain of integrase is sufficient to catalyze an apparent reversal of the DNA strand transfer reaction, termed disintegration. This catalytic core domain, as well as the full-length integrase, has been refractory to structural studies by x-ray crystallography or NMR because of its low solubility and propensity to aggregate. In an attempt to improve protein solubility, we used site-directed mutagenesis to replace hydrophobic residues within the core domain with either alanine or lysine. The single substitution of lysine for phenylalanine at position 185 resulted in a core domain that was highly soluble, monodisperse in solution, and retained catalytic activity. This amino acid change has enabled the catalytic domain of integrase to be crystallized and the structure has been solved to 2.5-A resolution [Dyda, F., Hickman, A. B., Jenkins, T. M., Engelman, A., Craigie, R. & Davies, D. R. (1994) Science 266, 1981-1986]. Systematic replacement of hydrophobic residues may be a useful strategy to improve the solubility of other proteins to facilitate structural and biochemical studies.

A new method for extraction of iron-molybdenum cofactor (FeMoco) from nitrogenase adsorbed to DEAE-cellulose. 1. Effects of anions, cations, and preextraction treatments.

A convenient and rapid method of obtaining the cofactor of nitrogenase (FeMoco) with a low and apparently limiting Fe/Mo ratio has been developed. FeMoco can be extracted from the MoFe protein bound to DEAE-cellulose. The cofactor is eluted in either N-methylformamide (NMF), N,N-dimethylformamide (DMF), or mixtures of these solvents by use of salts such as Et4NBr,Bu4NBr,Ph4PCl, and Ph4AsCl. The method is simple, is rapid (45 min), yields concentrated cofactor, and, unlike the original method [Shah, V. K., & Brill, W. J. (1977) Proc. Natl. Acad. Sci. U.S.A. 74, 3249-3253] which requires anaerobic centrifugation, is easily scaled up. Furthermore, it gives yields of cofactor in excess of 70%. Its disadvantages are a high Fe:Mo ratio when DMF is the extracting solvent and a high salt concentration in the resultant FeMoco solution. These disadvantages are easily overcome by removing excess Fe by pretreating the cofactor with bipyridyl while still on the column. This gives Fe:Mo ratios of (6 +/- 1):1 (11 trials) with specific activities ranging from 170 to 220 nmol of C2H4/[min.(nmol of Mo)]. Chromatography on Sephadex LH-20 removes ca. 99% of the excess salt. The adsorption of MoFe protein to DEAE-cellulose seems to facilitate denaturation by organic solvents so that pretreatment of the protein with acid, used in earlier methods, is unnecessary. There is an apparent dependence on the charge density of the anion employed for elution of FeMoco bound to DEAE-cellulose, such that Cl- greater than Br- much greater than I-, PF6- is the order of effectiveness of the Bu4N+ salts of these anions.(ABSTRACT TRUNCATED AT 250 WORDS)

Solution structure of the DNA binding domain of HIV-1 integrase.

The solution structure of the DNA binding domain of HIV-1 integrase (residues 220-270) has been determined by multidimensional NMR spectroscopy. The protein is a dimer in solution, and each subunit is composed of a five-stranded beta-barrel with a topology very similar to that of the SH3 domain. The dimer is formed by a stacked beta-interface comprising strands 2, 3, and 4, with the two triple-stranded antiparallel beta-sheets, one from each subunit, oriented antiparallel to each other. One surface of the dimer, bounded by the loop between strands beta 1 and beta 2, forms a saddle-shaped groove with dimensions of approximately 24 x 23 x 12 A in cross section. Lys264, which has been shown from mutational data to be involved in DNA binding, protrudes from this surface, implicating the saddle-shaped groove as the potential DNA binding site.

Role of a pineal cAMP-operated arylalkylamine N-acetyltransferase/14-3-3-binding switch in melatonin synthesis.

The daily rhythm in melatonin levels is controlled by cAMP through actions on the penultimate enzyme in melatonin synthesis, arylalkylamine N-acetyltransferase (AANAT; serotonin N-acetyltransferase, EC ). Results presented here describe a regulatory/binding sequence in AANAT that encodes a cAMP-operated binding switch through which cAMP-regulated protein kinase-catalyzed phosphorylation [RRHTLPAN --> RRHpTLPAN] promotes formation of a complex with 14-3-3 proteins. Formation of this AANAT/14-3-3 complex enhances melatonin production by shielding AANAT from dephosphorylation and/or proteolysis and by decreasing the K(m) for 5-hydroxytryptamine (serotonin). Similar switches could play a role in cAMP signal transduction in other biological systems.