Tertiary and quaternary structures of photoreactive Fe-type nitrile hydratase from Rhodococcus sp. N-771: roles of hydration water molecules in stabilizing the structures and the structural origin of the substrate specificity of the enzyme.

The crystal structure analysis of the Fe-type nitrile hydratase from Rhodococcus sp. N-771 revealed the unique structure of the enzyme composed of the alpha- and beta-subunits and the unprecedented structure of the non-heme iron active center [Nagashima, S., et al. (1998) Nat. Struct. Biol. 5, 347-351]. A number of hydration water molecules were identified both in the interior and at the exterior of the enzyme. The study presented here investigated the roles of the hydration water molecules in stabilizing the tertiary and the quaternary structures of the enzyme, based on the crystal structure and the results from a laser light scattering experiment for the enzyme in solution. Seventy-six hydration water molecules between the two subunits significantly contribute to the alphabeta heterodimer formation by making up the surface shape, forming extensive networks of hydrogen bonds, and moderating the surface charge of the beta-subunit. In particular, 20 hydration water molecules form the extensive networks of hydrogen bonds stabilizing the unique structure of the active center. The amino acid residues hydrogen-bonded to those hydration water molecules are highly conserved among all known nitrile hydratases and even in the homologous enzyme, thiocyanate hydrolase, suggesting the structural conservation of the water molecules in the NHase family. The crystallographic asymmetric unit contained two heterodimers connected by 50 hydration water molecules. The heterotetramer formation in crystallization was clearly explained by the concentration-dependent aggregation state of NHase found in the light scattering measurement. The measurement proved that the dimer-tetramer equilibrium shifted toward the heterotetramer dominant state in the concentration range of 10(-2)-1.0 mg/mL. In the tetramer dominant state, 50 water molecules likely glue the two heterodimers together as observed in the crystal structure. Because NHase exhibits a high abundance in bacterial cells, the result suggests that the heterotetramer is physiologically relevant. In addition, it was revealed that the substrate specificity of this enzyme, recognizing small aliphatic substrates rather than aromatic ones, came from the narrowness of the entrance channel from the bulk solvent to the active center. This finding may give a clue for changing the substrate specificity of the enzyme. Under the crystallization condition described here, one 1,4-dioxane molecule plugged the channel. Through spectroscopic and crystallographic experiments, we found that the molecule prevented the dissociation of the endogenous NO molecule from the active center even when the crystal was exposed to light.     

Mass spectrometry of purified amyloid beta protein in Alzheimer's disease.

The amyloid beta-protein (A beta) that is progressively deposited in Alzheimer's disease (AD) arises from proteolysis of the integral membrane protein, beta-amyloid precursor protein (beta APP). Although A beta formation appears to play a seminal role in AD, only a few studies have examined the chemical structure of A beta purified from brain, and there are discrepancies among the findings. We describe a new method for the rapid extraction and purification of A beta that minimizes artifactual proteolysis. A beta purified by two-dimensional reverse-phase HPLC was analyzed by combined amino acid sequencing and mass spectrometry after digestion with a lysylendopeptidase. The major A beta peptide in the cerebral cortex of all five AD brains examined was aspartic acid 1 to valine 40. A minor species beginning at glutamic acid 3 but blocked by conversion to pyroglutamate was also found in all cases. A species ending at threonine 43 was detected, varying from approximately 5 to 25% of total A beta COOH-terminal fragments. Peptides ending with valine 39, isoleucine 41, or alanine 42 were not detected, except for one brain with a minor peptide ending at valine 39. Our findings suggest that A beta 1-40 is the major species of beta-protein in AD cerebral cortex. A beta 1-40 and A beta 1-43 peptides could arise independently from beta APP, or A beta 1-43 could be the initial excised fragment, followed by digestion to yield A beta 1-40. These analyses of native A beta in AD brain recommend the use of synthetic A beta 1-40 peptide to model amyloid fibrillogenesis and toxicity in vitro.     

Isolation to homogeneity and partial characterization of a histo-blood group A defined Fuc alpha 1----2Gal alpha 1----3-N-acetylgalactosaminyltransferase from human lung tissue

The soluble histo-blood group A glycosyltransferase (Fuc alpha 1----Gal alpha 1----3-N-acetylgalactosaminyltransferase) was purified approximately 600,000-fold to homogeneity from human lung tissue. The enzyme was solubilized in 1% Triton X-100, partially purified by affinity chromatography on Sepharose 4B, and eluted with UDP. Final purification was obtained by twice repeated fast protein liquid chromatography ion exchange (Mono STM) with NaCl gradient elution and reverse-phase chromatography (proRPC) with acetonitrile gradient elution. Identity of the purified protein was established by (i) demonstration of the putative A transferase protein only in affinity-purified extracts of A but not O individuals, and (ii) specific immunoprecipitation of enzyme activity and putative protein with monoclonal antibodies. Sodium dodecyl sulfate electrophoresis revealed a single protein band with apparent Mr of approximately 40,000 under both reducing and nonreducing conditions. Digestion with N-glycanase yielded a reduction in Mr of approximately 6,000 (estimated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis), suggesting that the A transferase is a glycoprotein with N-linked carbohydrate chains. Amino acid composition and N-terminal amino acid sequence of the intact transferase, as well as of peptides released by endolysyl peptidase digest or cyanogen bromide cleavage, are presented.     

Amino acid sequence at the ATP-binding site of cGMP-dependent protein kinase

The amino acid sequence at the ATP-binding site on the cGMP-dependent protein kinase has been determined. For this determination the enzyme was labeled covalently by 5'-p-fluorosulfonyl[14C]benzoyladenosine and fragmented using cyanogen bromide or digested by trypsin after succinylation. The 14C-labeled peptides were purified by gel filtration and high performance liquid chromatography. The amino acid sequence around the site was found to be: -Val-Glu-Leu-Val-Gln-Leu-Lys-Ser-Glu-Glu-Ser-Lys-Thr-Phe-Ala-Met-*Lys-Ile-Leu-Lys--Lys-Arg-His-Ile-Val-Asp-Thr-Arg-Gln-Gln-Glu-His-Ile-Arg-Ser-Glu-Lys-, in which *Lys is the lysine residue that was modified by the affinity reagent. When this sequence was compared with that of the ATP-binding site of the catalytic subunit of cAMP-dependent protein kinase, a high degree of structural homology was observed for this site in the two proteins.     

Horse liver aldehyde dehydrogenase. Purification and characterization of two isozymes.

Two isozymes of horse liver aldehyde dehydrogenase (aldehyde, NAD oxidoreductase (EC 1.2.1.3)), F1 and F2, have been purified to homogeneity using salt fractionation followed by ion exchange and gel filtration chromatography. The specific activities of the two isozymes in a pH 9.0 system with propionaldehyde as substrate were approximately 0.35 and 1.0 mumol of NADH/min/mg of protein for the F1 and F2 isozymes, respectively. The multiporosity polyacrylamide gel electrophoresis molecular weights of the F1 and F2 isozymes were approximately 230,000 and 240,000 respectively. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis gave subunit molecular weight estimates of 52,000 and 53,000 for the F1 and F2 isozymes, respectively. The amino acid compositions of the two isozymes were found to be similar; the ionizable amino acid contents being consistent with the electrophoretic and chromatographic behavior of the two isozymes. Both isozymes exhibited a broad aldehyde specificity, oxidizing a wide variety of aliphatic and aromatic aldehydes and utilized NAD as coenzyme, but at approximately 300-fold higher coenzyme concentration could use NADP. The F1 isozyme exhibited a very low Km for NAD (3 muM) and a higher Km for acetaldehyde (70 muM), while the F2 isozyme was found to have a higher Km for NAD (30 muM) and a low Km for acetaldehyde (0.2 muM). The two isozymes showed similar chloral hydrate and p-chloromercuribenzoate inhibition characteristics, but the F1 isozyme was found to be several orders of magnittude more sensitive to disulfiram, a physiological inhibitor of acetaldehyde oxidation. Based on its disulfiram inhibition characteristics, it has been suggested that the F1 isozyme may be the primary enzyme for oxidizing the acetyldehyde produced during ethanol oxidation in vivo.     

Two ANGUSTIFOLIA genes regulate gametophore and sporophyte development in Physcomitrella patens

In Arabidopsis thaliana the ANGUSTIFOLIA (AN) gene regulates the width of leaves by controlling the diffuse growth of leaf cells in the medio‐lateral direction. In the genome of the moss Physcomitrella patens, we found two normal ANs (PpAN1‐1 and 1‐2). Both PpAN1 genes complemented the A. thaliana an‐1 mutant phenotypes. An analysis of spatiotemporal promoter activity of each PpAN1 gene, using transgenic lines that contained each PpAN1‐promoter– uidA (GUS) gene, showed that both promoters are mainly active in the stems of haploid gametophores and in the middle to basal region of the young sporophyte that develops into the seta and foot. Analyses of the knockout lines for PpAN1‐1 and PpAN1‐2 genes suggested that these genes have partially redundant functions and regulate gametophore height by controlling diffuse cell growth in gametophore stems. In addition, the seta and foot were shorter and thicker in diploid sporophytes, suggesting that cell elongation was reduced in the longitudinal direction, whereas no defects were detected in tip‐growing protonemata. These results indicate that both PpAN1 genes in P. patens function in diffuse growth of the haploid and diploid generations but not in tip growth. To visualize microtubule distribution in gametophore cells of P. patens, transformed lines expressing P. patens α‐tubulin fused to sGFP were generated. Contrary to expectations, the orientation of microtubules in the tips of gametophores in the PpAN1‐1/1‐2 double‐knockout lines was unchanged. The relationships among diffuse cell growth, cortical microtubules and AN proteins are discussed.     

Site-specific incorporation of an unnatural amino acid into proteins in mammalian cells

A suppressor tRNA(Tyr) and mutant tyrosyl-tRNA synthetase (TyrRS) pair was developed to incorporate 3-iodo-L-tyrosine into proteins in mammalian cells. First, the Escherichia coli suppressor tRNA(Tyr) gene was mutated, at three positions in the D arm, to generate the internal promoter for expression. However, this tRNA, together with the cognate TyrRS, failed to exhibit suppressor activity in mammalian cells. Then, we found that amber suppression can occur with the heterologous pair of E.coli TyrRS and Bacillus stearothermophilus suppressor tRNA(Tyr), which naturally contains the promoter sequence. Furthermore, the efficiency of this suppression was significantly improved when the suppressor tRNA was expressed from a gene cluster, in which the tRNA gene was tandemly repeated nine times in the same direction. For incorporation of 3-iodo-L-tyrosine, its specific E.coli TyrRS variant, TyrRS(V37C195), which we recently created, was expressed in mammalian cells, together with the B.stearothermophilus suppressor tRNA(Tyr), while 3-iodo-L-tyrosine was supplied in the growth medium. 3-Iodo-L-tyrosine was thus incorporated into the proteins at amber positions, with an occupancy of >95%. Finally, we demonstrated conditional 3-iodo-L-tyrosine incorporation, regulated by inducible expression of the TyrRS(V37C195) gene from a tetracycline-regulated promoter.     

The growth inhibitory factor that is deficient in the Alzheimer's disease brain is a 68 amino acid metallothionein-like protein

We have purified and characterized the growth inhibitory factor (GIF) that is abundant in the normal human brain, but greatly reduced in the Alzheimer's disease (AD) brain. GIF inhibited survival and neurite formation of cortical neurons in vitro. Purified GIF is a 68 amino acid small protein, and its amino acid sequence is 70% identical to that of human metallothionein II with a 1 amino acid insert and a unique 6 amino acid insert in the NH2-terminal and the COOH-terminal portions, respectively. The antibodies to the unique sequence of GIF revealed a distinct subset of astrocytes in the gray matter that appears to be closely associated with neuronal perikarya and dendrites. In the AD cortex, the number of GIF-positive astrocytes was drastically reduced, suggesting that GIF is down-regulated in the subset of astrocytes during AD.