Glucose polyester biosynthesis. Purification and characterization of a glucose acyltransferase.

Glandular trichomes of the wild tomato species Lycopersicon pennellii secrete 2,3,4-O-tri-acyl-glucose (-Glc), which contributes to insect resistance. A Glc acyltransferase catalyzes the formation of diacyl-Glc by disproportionating two equivalents of 1-O-acyl-beta-Glc, a high-energy molecule formed by a UDP-Glc dependent reaction. The acyltransferase was purified 4,900-fold from L. pennellii leaves by polyethylene glycol fractionation, diethylaminoethyl chromatography, concanavalin A affinity chromatography, and chromatofocusing. The acyltransferase possesses an isoelectric point of 4.8, a relative molecular mass around 110 kD, and is composed of 34- and 24-kD polypeptides as a heterotetramer. The 34- and 24-kD proteins were partially sequenced. The purified enzyme catalyzes both the disproportionation of 1-O-acyl-beta-Glcs to generate 1,2-di-O-acyl-beta-Glc and anomeric acyl exchange between 1-O-acyl-beta-Glc and Glc.

Regulation of Triacylglucose Fatty Acid Composition (Uridine Diphosphate Glucose:Fatty Acid Glucosyltransferases with Overlapping Chain-Length Specificity).

UDP-glucose (UDP-Glc):fatty acid glucosyltransferases catalyze the UDP-Glc-dependent activation of fatty acids as 1-O-acyl-[beta]-glucoses. 1-O-Acyl-[beta]-glucoses act as acyl donors in the biosynthesis of 2,3,4-tri-O-acylglucoses secreted by wild tomato (Lycopersicon pennellii) glandular trichomes. The acyl composition of L. pennellii 2,3,4-tri-O-acylglucoses is dominated by branched short-chain acids (4:0 and 5:0; approximately 65%) and straight and branched medium-chain-length fatty acids (10:0 and 12:0; approximately 35%). Two operationally soluble UDP-Glc:fatty acid glucosyltransferases (I and II) were separated and partially purified from L. pennellii (LA1376) leaves by polyethylene glycol precipitation followed by DEAE-Sepharose and Cibacron Blue 3GA-agarose chromatography. Whereas both transferases possessed similar affinity for UDP-Glc, glucosyltransferase I showed higher specificity toward short-chain fatty acids (4:0) and glucosyltransferase II showed higher specificity toward medium-chain fatty acids (8:0 and 12:0). The overlapping specificity of UDP-Glc:fatty acid glucosyltransferases for 4:0 to 12:0 fatty acid chain lengths suggests that the mechanism of 6:0 to 9:0 exclusion from acyl substituents of 2,3,4-tri-O-acylglucoses is unlikely to be controlled at the level of fatty acid activation. UDP-Glc:fatty acid glucosyltransferases are also present in cultivated tomato (Lycopersicon esculentum), and activities toward 4:0, 8:0, and 12:0 fatty acids do not appear to be primarily epidermal when assayed in interspecific periclinal chimeras.

1-O-acyl-beta-D-glucoses as fatty acid donors in transacylation reactions.

Glandular trichome exudates of the wild tomato, Lycopersicon pennellii Corr. (D'Arcy), are composed of 2,3,4-tri-O-acylglucoses possessing short to medium chain length (C4-C12) straight- and branched-chain fatty acids. These fatty acids are activated via a UDP-glucose-dependent reaction to form their respective high-energy 1-O-acyl-beta-D-glucopyranose derivatives. The activated 1-O-acyl moieties are then transacylated to other glucose and/or partially acylated glucose derivatives. Here we demonstrate that 1-O-acyl-beta-D-glucopyranoses participate in two types of glucose transacylation reactions. The first is anomeric exchange of the acyl moiety from 1-O-acyl-beta-D-glucopyranose to beta-glucose. The second is disproportionation between two molecules of 1-O-acyl-beta-D-glucopyranose resulting in formation of diacylglucose and glucose. Leaf extracts from cultivated tomato (Lycopersicon esculentum), which does not accumulate epicuticular acylglucoses, possess low levels of both the acyl exchange and disproportionation activities.

UDPglucose: fatty acid transglucosylation and transacylation in triacylglucose biosynthesis.

Glandular trichomes of the wild tomato Lycopersicon pennellii Corr. (D'Arcy) secrete large amounts of 2,3,4-tri-O-acylglucoses possessing straight- and branched-chain fatty acids of short to medium chain length (C4-C12). Although previous biosynthetic studies suggested that glucose acylation proceeded via acyl CoA intermediates, repeated attempts to demonstrate isobutyryl-CoA-dependent glucose acylation were unsuccessful. When [14C]isobutyrate is administered to detached L. pennellii leaves, the label is readily converted to 1-O-isobutyryl-beta-D-glucose. This is immediately followed by the appearance of di- and triacylated glucose esters. L. pennellii extracts catalyzed the formation of 1-O-isobutyryl-beta-D-glucose from isobutyrate and UDPglucose, and detached L. pennellii trichomes catalyzed transfer of the isobutyryl moiety from synthetic 1-O-isobutyryl-beta-D-glucose to D-glucose. Detached L. pennellii trichomes also catalyzed the formation of diacylglucose and triacylglucose via transfer of the isobutyryl moiety from 1-O-[14C]isobutyryl-beta-D-glucose to mono- or diacylglucoses, respectively. These studies suggest a multistep mechanism in which activation of fatty acids to their respective high-energy 1-O-acyl-beta-D-glucopyranose derivatives is followed by transfer of the 1-O-acyl moiety to non-anomeric positions of other glucose and/or partially acylated glucose molecules. This appears to be the primary mechanism of activation and fatty acid esterification to glucose in L. pennellii trichomes. Cultivated tomato, L. esculentum Mill., also activates free fatty acids to their 1-O-acyl-beta-D-glucose derivatives but lacks the acyl transfer mechanism for synthesizing polyacylated sugars.

DNA sequences of three beta-1,4-endoglucanase genes from Thermomonospora fusca.

The DNA sequences of the Thermomonospora fusca genes encoding cellulases E2 and E5 and the N-terminal end of E4 were determined. Each sequence contains an identical 14-bp inverted repeat upstream of the initiation codon. There were no significant homologies between the coding regions of the three genes. The E2 gene is 73% identical to the celA gene from Microbispora bispora, but this was the only homology found with other cellulase genes. E2 belongs to a family of cellulases that includes celA from M. bispora, cenA from Cellulomonas fimi, casA from an alkalophilic Streptomyces strain, and cellobiohydrolase II from Trichoderma reesei. E4 shows 44% identity to an avocado cellulase, while E5 belongs to the Bacillus cellulase family. There were strong similarities between the amino acid sequences of the E2 and E5 cellulose binding domains, and these regions also showed homology with C. fimi and Pseudomonas fluorescens cellulose binding domains.

Cloning of a Thermomonospora fusca xylanase gene and its expression in Escherichia coli and Streptomyces lividans.

Thermomonospora fusca chromosomal DNA was partially digested with EcoRI to obtain 4- to 14-kilobase fragments, which were used to construct a library of recombinant phage by ligation with EcoRI arms of lambda gtWES. lambda B. A recombinant phage coding for xylanase activity which contained a 14-kilobase insert was identified. The xylanase gene was localized to a 2.1-kilobase SalI fragment of the EcoRI insert by subcloning onto pBR322 and derivatives of pBR322 that can also replicate in Streptomyces lividans. The xylanase activity produced by S. lividans transformants was 10- to 20-fold higher than that produced by Escherichia coli transformants but only one-fourth the level produced by induced T. fusca. A 30-kilodalton peptide with activity against both Remazol brilliant blue xylan and xylan was produced in S. lividans transformants that carried the 2.1-kilobase SalI fragment of T. fusca DNA and was not produced by control transformants. T. fusca cultures were found to contain a xylanase of a similar size that was induced by growth on xylan or Solka Floc. Antiserum directed against supernatant proteins isolated from a Solka Floc-grown T. fusca culture inhibited the xylanase activity of S. lividans transformants. The cloned T. fusca xylanase gene was expressed at about the same level in S. lividans grown in minimal medium containing either glucose, cellobiose, or xylan. The xylanase bound to and hydrolyzed insoluble xylan. The cloned xylanase appeared to be the same as the major protein in xylan-induced T. fusca culture supernatants, which also contained at least three additional minor proteins with xylanase activity and having apparent molecular masses of 43, 23, and 20 kilodaltons.

Cloning of the Thermomonospora fusca Endoglucanase E2 Gene in Streptomyces lividans: Affinity Purification and Functional Domains of the Cloned Gene Product.

Thermomonospora fusca YX grown in the presence of cellulose produces a number of beta-1-4-endoglucanases, some of which bind to microcrystalline cellulose. By using a multicopy plasmid, pIJ702, a gene coding for one of these enzymes (E2) was cloned into Streptomyces lividans and then mobilized into both Escherichia coli and Streptomyces albus. The gene was localized to a 1.6-kilobase PvuII-ClaI segment of the originally cloned 3.0-kilobase SstI fragment of Thermomonospora DNA. The culture supernatants of Streptomyces transformants contain a major endoglucanase that cross-reacts with antibody against Thermomonospora cellulase E2 and has the same molecular weight (43,000) as T. fusca E2. This protein binds quickly and tightly to Avicel, from which it can be eluted with guanidine hydrochloride but not with water. It also binds to filter paper but at a slower rate than to Avicel. Several large proteolytic degradation products of this enzyme generated in vivo lose the ability to bind to Avicel and have higher activity on carboxymethyl cellulose than the native enzyme. Other smaller products bind to Avicel but lack activity. A weak cellobiose-binding site not observed in the native enzyme was present in one of the degradation products. In E. coli, the cloned gene produced a cellulase that also binds tightly to Avicel but appeared to be slightly larger than T. fusca E2. The activity of intact E2 from all organisms can be inactivated by Hg ions. Dithiothreitol protected against Hg inactivation and reactivated both unbound and Avicel-bound Hg-inhibited E2, but at different rates.

Expression of a Thermomonospora fusca Cellulase Gene in Streptomyces lividans and Bacillus subtilis.

A cellulase gene from Thermomonospora fusca coding for endocellulase E(5) was introduced into Streptomyces lividans by using shuttle plasmids that can replicate in either S. lividans or Escherichia coli. Plasmid DNA isolated from E. coli was used to transform S. lividans, selecting for thiostrepton resistance. The transformants expressed and excreted the endocellulase, but the ability to produce the endocellulase was unstable. This instability was shown to result from deletion of the endocellulase gene from the plasmid. Plasmid DNA prepared from a culture in which plasmid modification had occurred was used to transform E. coli, selecting for Amp cells, and all of the transformants were cellulase positive, showing that pBR322 and T. fusca DNA were deleted together. When a plasmid was constructed containing only T. fusca DNA in plasmid pIJ702, the transformants were more stable, and the level of endocellulase activity produced in the culture supernatant after growth on 0.2% glucose was close to the level produced by T. fusca cultures grown on 0.2% cellulose. About 50% of the total protein in the culture supernatant of the S. lividans transformant was endocellulase E(5). The enzyme produced by the S. lividans transformant was identical to pure T. fusca E(5) in its electrophoretic mobility and was completely inhibited by antiserum to E(5). Shuttle plasmids containing the E(5) gene that could replicate in Bacillus subtilis and E. coli were also constructed and used to transform B. subtilis. Again there was extensive deletion of the plasmid DNA during transformation and growth in B. subtilis. There was no evidence of E(5) activity, even in those B. subtilis transformants that retained the E(5) gene.

Isolation and characterization of the Salmonella typhimurium LT2 xylose regulon.

Salmonella DNA was partially digested with EcoRI, and the digest was fractionated to obtain fragments larger than 8 kilobases (kb). These were ligated into EcoRI-cut pBR322, and the mixture was used to transform Salmonella Xyl- cells selecting for ampR xyl+ transformants. A 21- and a 27-kb plasmid were isolated, both of which contained the entire xylose regulon. The xylose regulon was localized to a 6.3-kb segment of a 13.5-kb EcoRI fragment. Subclones were constructed which contained either the genes for D-xylose isomerase and D-xylulokinase or the genes for the D-xylose transport and the D-xylose regulatory factors. The gene order determined by the subcloning experiments is consistent with that determined by genetic mapping. The spots corresponding to D-xylose isomerase and D-xylulokinase subunits were identified in two-dimensional gels of several xylose-induced strains. Each of them had a molecular weight of 45,000 and an isoelectric point of 6.2 +/- 0.1.

Analysis of HGPRT- CRM+ human lymphoblast mutants.

Three 6-thioguanine-resistant mutants of the human diploid lymphoblast line MGL-8 were studied. The inactivation by heat of both HGPRT activity and antigenicity of the HGPRT immunologically cross-reacting material of the A30 mutant cells were not protected by PRPP, indicating that the HGPRT in A30 cells has an altered PRPP binding site, leading to lack of stabilization and rapid degradation of the enzyme. Two dimensional separations of the immunoprecipitates from extracts of the parental and mutant cell lines showed that the A35 mutant CRM has a more acidic isoelectric pH, while the A30 CRM has a more basic isoelectric pH and that the A30 protein has a faster rate of degradation than the wild-type HGPRT. The A30 CRM also has a smaller molecular size than the wild-type enzyme.

Hypoxanthine phosphoribosyltransferase: two-dimensional gels from normal and Lesch-Nyhan hemolyzates.

Immunoprecipitated hypoxanthine phosphoribosyltransferase (HPRT) from hemolyzates displays two major spots after two-dimensional polyacrylamide gel electrophoresis. HeLa cells or human lymphoblasts display only a single HPRT spot located at the same position as the most basic of the hemolyzate HPRT spots. This suggests that the most basic spot is the form initially synthesized, and the more acidic hemolyzate HPRT spot (a pseudoisozyme) is probably derived from the first by an age-related modification (for example, deamidation). The HPRT pattern of the hemolyzate from a Lesch-Nyhan patient was shifted to a more basic isoelectric pH, implying the mutation of a structural gene.

Analysis of HeLa cell hypoxanthine phosphoribosyltransferase mutants and revertants by two-dimensional polyacrylamide gel electrophoresis: evidence for silent gene activation.

The spot corresponding to hypoxanthine phosphoribosyltransferase (HPRT; IMP:pyrophosphate phosphoribosyltransferase, EC 2.4.2.8) has been identified in two-dimensional polyacrylamide gels of HeLa cell extracts. This spot is absent in gels of 24 HPRT dificient mutants. A missense mutant displays a new HPRT spot at the same molecular weight but different isoelectric focusing position. Five independently isolated revertants of the missense mutant display spots corresponding to both the wild-type and mutant proteins indicating that they synthesize HPRT from two separate genes. If the missense protein is synthesized from a mutated form of the initially active HPRT gene, then wild-type HPRT protein in the revertants must be snythesized from a newly activated but prevously silent wild-type gene. The newly activated gene in the revertants of the missense mutation appears unstable producing a high frequency of spontaneous HPRT mutants.

Radioimmune determination of hypoxanthine phosphoribosyltransferase crossreacting material in erythrocytes of Lesch-Nyhan patients.

We have developed a sensitive radioimmunoassay capable of detecting and quantitating 20 ng of hypoxanthine phosphoribosyltransferase (EC 2.4.2.8; IMP:pyrophosphate phosphoribosyltransferase) protein. For this assay, hypoxanthine phosphoribosyltransferase from human erythrocytes was iodinated with 125I under mild conditions using hydrogen peroxide and lactoperoxidase attached to Sepharose-4B. Antisera prepared against homogeneous human hypoxanthine phosphoribosyltransferase precipitates the iodinated enzyme as effectively as the unlabeled enzyme. The radioimmunoassay has been used to look for hypoxanthine phosphoribosyltransferase crossreacting material in hemolysates from sixteen different patients with a marked genetic deficiency of this enzyme characteristic of the Lesch-Nyhan syndrome. Fifteen hemolysates contained no detectable (less than 1% of normal) crossreacting material. One hemolysate contained a normal amount of crossreacting material. Hypoxanthine phosphoribosyltransferase from this patient (E.S.) has been shown to be a Km mutant enzyme.

Specific hydrolysis of the cohesive ends of bacteriophage lambda DNA by three single strand-specific nucleases.

Procedures have been worked out for Aspergillus nuclease S1 and mung been nuclease to quantitatively cleave off both of the 12-nucleotide long, single-stranded cohesive ends of lambdaDNA. This cleavage is indicated by the almost complete elimination of the repair incorporation of radioactive nucleotides by DNA polymerase into the digested DNA. With S1 nuclease, cleavage was complete at 10 degrees as well as at 30 degrees. Under the conditions for quantitative cleavage of the single-stranded regions there was no digestion of the double-stranded lambdaDNA. The mung bean nuclease cleaved off the cohesive ends completely at 30 degrees but at 5 degrees, the cleavage was not complete even at high enzyme concentration. The nearest neighbor analysis of the repaired DNA indicates that at 5 degrees about four nucleotides remained undigested. The mung bean nuclease also introduced, under the conditions used, some nicks into double-stranded DNA as determined by the repair incorporation. The Escherichia coli exonuclease VII cleaved off part of the cohesive ends of lambdaDNA, leaving two nucleotides on each end as single-stranded tails.