Search

Cellulase mixtures from Hypocrea jecorina are commonly used for the saccharification of cellulose in biotechnical applications. The most abundant β-glucosidase in the mesophilic fungus Hypocrea jecorina is HjCel3A, which hydrolyzes the β-linkage between two adjacent molecules in dimers and short oligomers of glucose. It has been shown that enhanced levels of HjCel3A in H. jecorina cellulase mixtures benefit the conversion of cellulose to glucose. Biochemical characterization of HjCel3A shows that the enzyme efficiently hydrolyzes (1,4)- as well as (1,2)-, (1,3)-, and (1,6)-β-D-linked disaccharides. For crystallization studies, HjCel3A was produced in both H. jecorina (HjCel3A) and Pichia pastoris (Pp-HjCel3A). Whereas the thermostabilities of HjCel3A and Pp-HjCel3A are the same, Pp-HjCel3A has a higher degree of N-linked glycosylation. Here, we present x-ray structures of HjCel3A with and without glucose bound in the active site. The structures have a three-domain architecture as observed previously for other glycoside hydrolase family 3 β-glucosidases. Both production hosts resulted in HjCel3A structures that have N-linked glycosylations at Asn(208) and Asn(310). In H. jecorina-produced HjCel3A, a single N-acetylglucosamine is present at both sites, whereas in Pp-HjCel3A, the P. pastoris-produced HjCel3A enzyme, the glycan chains consist of 8 or 4 saccharides. The glycosylations are involved in intermolecular contacts in the structures derived from either host. Due to the different sizes of the glycosylations, the interactions result in different crystal forms for the two protein forms.

Two fungal-secreted α-fucosidases and their genes were characterized. FoFCO1 was purified from culture filtrates of Fusarium oxysporum strain 0685 grown on l-fucose and its encoding gene identified in the sequenced genome of strain 4287. FoFCO1 was active on p-nitrophenyl-α-fucoside (pNP-Fuc), but did not defucosylate a nonasaccharide (XXFG) fragment of pea xyloglucan. A putative α-fucosidase gene (FgFCO1) from Fusarium graminearum was expressed in Pichia pastoris. FgFCO1 was ∼1,800 times less active on pNP-Fuc than FoFCO1, but was able to defucosylate the XXFG nonasaccharide. Although FgFCO1 and FoFCO1 both belong to Glycosyl Hydrolase family 29, they share

α-Linked xylose is a major component of xyloglucans in the cell walls of higher plants. An α-xylosidase (AxlA) was purified from a commercial enzyme preparation from Aspergillus niger, and the encoding gene was identified. The protein is a member of glycosyl hydrolase family 31. It was active on p-nitrophenyl-α-d-xyloside, isoprimeverose, xyloglucan heptasaccharide (XXXG), and tamarind xyloglucan. When expressed in Pichia pastoris, AxlA had activity comparable to the native enzyme on pNPαX and IP despite apparent hyperglycosylation. The pH optimum of AxlA was between 3.0 and 4.0. AxlA together with β-glucosidase depolymerized xyloglucan heptasaccharide. A combination of AxlA, β-glucosidase, xyloglucanase, and β-galactosidase in the optimal proportions of 51:5:19:25 or 59:5:11:25 could completely depolymerize tamarind XG to free Glc or Xyl, respectively. To the best of our knowledge, this is the first characterization of a secreted microbial α-xylosidase. Secreted α-xylosidases appear to be rare in nature, being absent from other tested commercial enzyme mixtures and from the genomes of most filamentous fungi.

The cost of enzymes for converting plant biomass materials to fermentable sugars is a major impediment to the development of a practical lignocellulosic ethanol industry. Research on enzyme optimization with the goal of reducing the cost of converting biomass materials such as corn stover into glucose, xylose, and other sugars is being actively pursued in private industry, academia, and government laboratories. Under the auspices of the Department of Energy Great Lakes Bioenergy Research Center, we are taking several approaches to address this problem, including "bioprospecting" for superior key enzymes, protein engineering, and high-level expression in plants. A particular focus is the development of synthetic enzyme mixtures, in order to learn which of the hundreds of known enzymes are important and in what ratios. A core set comprises cellobiohydrolase, endoglucanase, β- glucosidase, endoxylanase, and β-glucosidase. Accessory enzymes include esterases, proteases, nonhydrolytic proteins, and glycosyl hydrolases that cleave the less frequent chemical linkages found in plant cell walls.

Identification of the biosynthetic enzymes involved in cell wall biosynthesis remains one of the major unsolved problems of plant biology. Of the major polysaccharides of the plant cell wall, pectins and hemicelluloses are synthesized in the Golgi, and callose and cellulose are synthesized at the

Two new beta-glucanase-encoding genes, EXG2 and MLG2, were isolated from the plant-pathogenic fungus Cochliobolus carbonum using polymerase chain reaction based on amino acid sequences from the purified proteins. EXG2 encodes a 46.6-kDa exo-beta1,3-glucanase and is located on the same 3.5-Mb chromosome that contains the genes of HC-toxin biosynthesis. MLG2 encodes a 26.8-kDa mixed-linked (beta1,3-beta1,4) glucanase with low activity against beta1,4-glucan and no activity against beta1,3-glucan. Specific mutants of EXG2 and MLG2 were constructed by targeted gene replacement. Strains with multiple mutations (genotypes exg1/mlg1, exg2/mlg1, mlg1/mlg2, and exg1/exg2/mlg1/mlg2) were also constructed by sequential disruption and by crossing. Total mixed-linked glucanase activity in culture filtrates of mlg1/mlg2 and exg1/exg2/mlg1/mlg2 mutants was reduced by approximately 73%. Total beta1,3-glucanase activity was reduced by 10, 54, and 96% in exg2, mlg1, and exg1/exg2/mlg1/mlg2 mutants, respectively. The quadruple mutant showed only a modest decrease in growth on beta1,3-glucan or mixed-linked glucan. None of the mutants showed any decrease in virulence.

Dichloroacetamide safeners protect maize (Zea mays L.) against injury from chloroacetanilide and thiocarbamate herbicides. Etiolated maize seedlings have a high-affinity cytosolic-binding site for the safener [3H](R,S)-3-dichloroacetyl-2,2,5-trimethyl-1,3-oxazol-idine ([3H]Saf), and this safener-binding activity (SafBA) is competitively inhibited by the herbicides. The safener-binding protein (SafBP), purified to homogeneity, has a relative molecular weight of 39,000, as shown by sodium dodecyl sulfate-polyacrylamide gel electrophoresis, and an isoelectric point of 5.5. Antiserum raised against purified SafBP specifically recognizes a 39-kD protein in etiolated maize and sorghum (Sorghum bicolor L.), which have SafBA, but not in etiolated wheat (Triticum aestivum L.), oat (Avena sativa L.), barley (Hordeum vulgare L.), tobacco (Nicotiana tabacum L.), or Arabidopsis, which lack SafBA. SafBP is most abundant in the coleoptile and scarcest in the leaves, consistent with the distribution of SafBA. SBP1, a cDNA encoding SafBP, was cloned using polymerase chain reaction primers based on purified proteolytic peptides. Extracts of Escherichia coli cells expressing SBP1 have strong [3H]Saf binding, which, like binding to the native maize protein, is competitively inhibited by the safener dichlormid and the herbicidesS-ethyl dipropylthiocarbamate, alachlor, and metolachlor. SBP1 is predicted to encode a phenolicO-methyltransferase, but SafBP does notO-methylate catechol or caffeic acid. The acquisition of its encoding gene opens experimental approaches for the evaluation of the role of SafBP in response to the relevant safeners and herbicides.

Amatoxins, including α-amanitin, are bicyclic octapeptides found in mushrooms (Agaricomycetes, Agaricales) of certain species in the genera Amanita, Galerina, Lepiota, and Conocybe. Amatoxins and the chemically similar phallotoxins are synthesized on ribosomes in Amanita bisporigera, Amanita phalloides, and Amanita ocreata. In order to determine if amatoxins are synthesized by a similar mechanism in another, distantly related mushroom, we obtained genome survey sequence data from a monokaryotic isolate of Galerina marginata, which produces α-amanitin. The genome of G. marginata contains two copies of the α-amanitin gene (GmAMA1-1 and GmAMA1-2). The α-amanitin proprotein sequences of G. marginata (35 amino acids) are highly divergent from AMA1 of A. bisporigera except for the toxin region itself (IWGIGCNP in single-letter amino acid code) and the amino acids immediately upstream (N[A/S]TRLP). G. marginata does not contain any related toxin-encoding sequences besides GmAMA1-1 and GmAMA1-2. DNA from two other α-amanitin-producing isolates of Galerina (G. badipes and G. venenata) hybridized to GmAMA1, whereas DNA from the toxin non-producing species Galerina hybrida did not. Expression of the GmAMA1 genes was induced by growth on low carbon. RNASeq evidence indicates that both copies of GmAMA1 are expressed approximately equally. A prolyl oligopeptidase (POP) is strongly implicated in processing of the cyclic peptide toxins of A. bisporigera and Conocybe apala. G. marginata has two predicted POP genes; one, like AbPOPB of A. bisporigera, is present only in the toxin-producing isolates of Galerina and the other, like AbPOPA of A. bisporigera, is present in all species. Our results indicate that G. marginata biosynthesizes amatoxins on ribosomes by a pathway similar to Amanita species, involving a genetically encoded proprotein of 35 amino acids that is post-translationally processed by a POP. However, due to the high degree of divergence, the evolutionary relationship between AMA1 in the genera Amanita and Galerina is unclear.

Background Pretreatment is a critical step in the conversion of lignocellulose to fermentable sugars. Although many pretreatment processes are currently under investigation, none of them are entirely satisfactory in regard to effectiveness, cost, or environmental impact. The use of hydrogen peroxide at pH 11.5 (alkaline hydrogen peroxide (AHP)) was shown by Gould and coworkers to be an effective pretreatment of grass stovers and other plant materials in the context of animal nutrition and ethanol production. Our earlier experiments indicated that AHP performed well when compared against two other alkaline pretreatments. Here, we explored several key parameters to test the potential of AHP for further improvement relevant to lignocellulosic ethanol production. Results The effects of biomass loading, hydrogen peroxide loading, residence time, and pH control were tested in combination with subsequent digestion with a commercial enzyme preparation, optimized mixtures of four commercial enzymes, or optimized synthetic mixtures of pure enzymes. AHP pretreatment was performed at room temperature (23°C) and atmospheric pressure, and after AHP pretreatment the biomass was neutralized with HCl but not washed before enzyme digestion. Standard enzyme digestion conditions were 0.2% glucan loading, 15 mg protein/g glucan, and 48 h digestion at 50°C. Higher pretreatment biomass loadings (10% to 20%) gave higher monomeric glucose (Glc) and xylose (Xyl) yields than the 2% loading used in earlier studies. An H2O2 loading of 0.25 g/g biomass was almost as effective as 0.5 g/g, but 0.125 g/g was significantly less effective. Optimized mixtures of four commercial enzymes substantially increased post-AHP-pretreatment enzymatic hydrolysis yields at all H2O2 concentrations compared to any single commercial enzyme. At a pretreatment biomass loading of 10% and an H2O2 loading of 0.5 g/g biomass, an optimized commercial mixture at total protein loadings of 8 or 15 mg/g glucan gave monomeric Glc yields of 83% or 95%, respectively. Yields of Glc and Xyl after pretreatment at a low hydrogen peroxide loading (0.125 g H2O2/g biomass) could be improved by extending the pretreatment residence time to 48 h and readjusting the pH to 11.5 every 6 h during the pretreatment. A Glc yield of 77% was obtained using a pretreatment of 15% biomass loading, 0.125 g H2O2/g biomass, and 48 h with pH adjustment, followed by digestion with an optimized commercial enzyme mixture at an enzyme loading of 15 mg protein/g glucan. Conclusions Alkaline peroxide is an effective pretreatment for corn stover. Particular advantages are the use of reagents with low environmental impact and avoidance of special reaction chambers. Reasonable yields of monomeric Glc can be obtained at an H2O2 concentration one-quarter of that used in previous AHP research. Additional improvements in the AHP process, such as peroxide stabilization, peroxide recycling, and improved pH control, could lead to further improvements in AHP pretreatment.

Background Enzymes for plant cell wall deconstruction are a major cost in the production of ethanol from lignocellulosic biomass. The goal of this research was to develop optimized synthetic mixtures of enzymes for multiple pretreatment/substrate combinations using our high-throughput biomass digestion platform, GENPLAT, which combines robotic liquid handling, statistical experimental design and automated Glc and Xyl assays. Proportions of six core fungal enzymes (CBH1, CBH2, EG1, β-glucosidase, a GH10 endo-β1,4-xylanase, and β-xylosidase) were optimized at a fixed enzyme loading of 15 mg/g glucan for release of Glc and Xyl from all combinations of five biomass feedstocks (corn stover, switchgrass, Miscanthus, dried distillers' grains plus solubles [DDGS] and poplar) subjected to three alkaline pretreatments (AFEX, dilute base [0.25% NaOH] and alkaline peroxide [AP]). A 16-component mixture comprising the core set plus 10 accessory enzymes was optimized for three pretreatment/substrate combinations. Results were compared to the performance of two commercial enzymes (Accellerase 1000 and Spezyme CP) at the same protein loadings. Results When analyzed with GENPLAT, corn stover gave the highest yields of Glc with commercial enzymes and with the core set with all pretreatments, whereas corn stover, switchgrass and Miscanthus gave comparable Xyl yields. With commercial enzymes and with the core set, yields of Glc and Xyl were highest for grass stovers pretreated by AP compared to AFEX or dilute base. Corn stover, switchgrass and DDGS pretreated with AFEX and digested with the core set required a higher proportion of endo-β1,4-xylanase (EX3) and a lower proportion of endo-β1,4-glucanase (EG1) compared to the same materials pretreated with dilute base or AP. An optimized enzyme mixture containing 16 components (by addition of α-glucuronidase, a GH11 endoxylanase [EX2], Cel5A, Cel61A, Cip1, Cip2, β-mannanase, amyloglucosidase, α-arabinosidase, and Cel12A to the core set) was determined for AFEX-pretreated corn stover, DDGS, and AP-pretreated corn stover. The optimized mixture for AP-corn stover contained more exo-β1,4-glucanase (i.e., the sum of CBH1 + CBH2) and less endo-β1,4-glucanase (EG1 + Cel5A) than the optimal mixture for AFEX-corn stover. Amyloglucosidase and β-mannanase were the two most important enzymes for release of Glc from DDGS but were not required (i.e., 0% optimum) for corn stover subjected to AP or AFEX. As a function of enzyme loading over the range 0 to 30 mg/g glucan, Glc release from AP-corn stover reached a plateau of 60-70% Glc yield at a lower enzyme loading (5-10 mg/g glucan) than AFEX-corn stover. Accellerase 1000 was superior to Spezyme CP, the core set or the 16-component mixture for Glc yield at 12 h, but the 16-component set was as effective as the commercial enzyme mixtures at 48 h. Conclusion The results in this paper demonstrate that GENPLAT can be used to rapidly produce enzyme cocktails for specific pretreatment/biomass combinations. Pretreatment conditions and feedstock source both influence the Glc and Xyl yields as well as optimal enzyme proportions. It is predicted that it will be possible to improve synthetic enzyme mixtures further by the addition of additional accessory enzymes.

The high cost of enzymes is a major bottleneck preventing the development of an economically viable lignocellulosic ethanol industry. Commercial enzyme cocktails for the conversion of plant biomass to fermentable sugars are complex mixtures containing more than 80 proteins of suboptimal activities and relative proportions. As a step toward the development of a more efficient enzyme cocktail for biomass conversion, we have developed a platform, called GENPLAT, that uses robotic liquid handling and statistically valid experimental design to analyze synthetic enzyme mixtures. Commercial enzymes (Accellerase 1000 +/- Multifect Xylanase, and Spezyme CP +/- Novozyme 188) were used to test the system and serve as comparative benchmarks. Using ammonia-fiber expansion (AFEX) pretreated corn stover ground to 0.5 mm and a glucan loading of 0.2%, an enzyme loading of 15 mg protein/g glucan, and 48 h digestion at 50 degrees C, commercial enzymes released 53% and 41% of the available glucose and xylose, respectively. Mixtures of three, five, and six pure enzymes of Trichoderma species, expressed in Pichia pastoris, were systematically optimized. Statistical models were developed for the optimization of glucose alone, xylose alone, and the average of glucose + xylose for two digestion durations, 24 and 48 h. The resulting models were statistically significant (P < 0.0001) and indicated an optimum composition for glucose release (values for optimized xylose release are in parentheses) of 29% (5%) cellobiohydrolase 1, 5% (14%) cellobiohydrolase 2, 25% (25%) endo-beta1,4-glucanase 1, 14% (5%) beta-glucosidase, 22% (34%) endo-beta1,4-xylanase 3, and 5% (17%) beta-xylosidase in 48 h at a protein loading of 15 mg/g glucan. Comparison of two AFEX-treated corn stover preparations ground to different particle sizes indicated that particle size (100 vs. 500 microm) makes a large difference in total digestibility. The assay platform and the optimized "core" set together provide a starting point for the rapid testing and optimization of alternate core enzymes from other microbial and recombinant sources as well as for the testing of "accessory" proteins for development of superior enzyme mixtures for biomass conversion.

A high throughput enzyme assay platform, called GENPLAT, was used to guide the development of an optimized mixture of individual purified enzymes from ten "accessory" and six "core" enzymes. Enzyme mixtures were optimized for release of Glu, Xyl, or a combination of the two from corn stover pretreated by ammonia-fiber expansion (AFEX). Assay conditions were a fixed enzyme loading of 15 mg/g glucan, 48 h digestion, and 50 degrees C. Five of the ten tested accessory proteins enhanced Glu or Xyl yield compared to the core set alone, and five did not. An 11-component mixture containing the core set and five accessory enzymes optimized for Glu released 52.1% of the available Glu, compared to 38.5% with the core set alone. A mixture optimized for Xyl released 39.9% of the Xyl, compared to 26.4% with the core set alone. We predict that there is still considerable opportunity for further improvement of synthetic mixtures. Furthermore, the strategy described here is applicable to the development of more efficient enzyme cocktails for any pretreatment/biomass combination and for detecting enzymes that make a heretofore unrecognized contribution to lignocellulose deconstruction.

Race 1 of Cochliobolus carbonum, a fungal plant pathogen, owes its exceptional virulence on certain genotypes of maize to the production of HC-toxin, a cyclic tetrapeptide. Production of HC-toxin is controlled by a single known gene, TOX2. Race 1, but not races that do not make HC-toxin, contains two copies of a 22-kilobase (kb) region of chromosomal DNA that is required for HC-toxin biosynthesis and hence virulence. We have sequenced this 22-kb region and here show that it contains an open reading frame of 15.7 kb that encodes a multifunctional cyclic peptide synthetase of potential M(r)574,620. This gene, called HTS1, apparently contains no introns. The predicted gene product, HC-toxin synthetase (HTS), contains four amino acid-binding (adenylate-forming) domains that are highly similar to those found in other cyclic peptide synthetases and other adenylate-binding enzymes. The DNA sequence encodes tryptic peptides derived from two HC-toxin biosynthetic enzymes, HC-toxin synthetase 1 (HTS-1) and HC-toxin synthetase 2 (HTS-2), indicating that these two enzymes exist in vivo as part of a single polypeptide. Consistent with this, in some enzyme preparations antibodies against the enzyme HTS-2, which was originally purified as a protein with a subunit M(r) of 160,000, recognize a protein with an estimated subunit M(r) greater than 480,000.

High-throughput MS/MS was used to identify proteins secreted by Fusarium graminearum (Gibberella zeae) during growth on 13 media in vitro and in planta during infection of wheat heads. In vitro secreted proteins were collected from the culture filtrates, and in planta proteins were collected by vacuum infiltration. A total of 289 proteins (229 in vitro and 120 in planta) were identified with high statistical confidence. Forty-nine of the in planta proteins were not found in any of the in vitro conditions. The majority (91-100%) of the in vitro proteins had predicted signal peptides, but only 56% of the in planta proteins. At least 13 of the nonsecreted proteins found only in planta were single-copy housekeeping enzymes, including enolase, triose phosphate isomerase, phosphoglucomutase, calmodulin, aconitase, and malate dehydrogenase. The presence of these proteins in the in planta but not in vitro secretome might indicate that significant fungal lysis occurs during pathogenesis. On the other hand, several of the proteins lacking signal peptides that were found in planta have been reported to be potent immunogens secreted by animal pathogenic fungi, and therefore could be important in the interaction between F. graminearum and its host plants.

The CREA gene has been implicated in glucose repression in several fungi. The product of this gene, CreA, binds to the promoter region of several enzymes and down-regulates gene expression. An ortholog of CREA was isolated and characterized from the maize pathogenic fungus, Cochliobolus carbonum. The deduced amino acid sequence of the C. carbonum CREA gene is very similar to the CreA proteins of Aspergillus niger, Gibberella fujikuroi, Sclerotinia sclerotiorum and Trichoderma reesei, as well as the Mig1 protein of Saccharomyces cerevisiae. And like the other fungal proteins, C. carbonum CreA has two zinc finger regions and a nuclear localization signal. Putative CreA binding sequences were also identified in the 5' region of three C. carbonum cell wall degrading enzyme genes suggesting that the protein may play a role in the regulatory process that controls these enzymes expression.

The expression of four Cochliobolus carbonum endo-1,4-b-xylanase genes (XYL1, XYL2, XYL3, XYL4), and an exo-1,4-b-xylosidase gene (XYP1) was studied following the growth of the fungus in minimal medium containing glucose, sucrose, xylose, xylan, pectin, or cellulose. The XYL1 and XYL2 genes were expressed only when the culture medium contained xylan or cellulose. Both XYL3 and XYL4 are induced by xylose and xylan, and XYP1 expression is induced by xylose, xylan, pectin and cellulose. None of these genes is expressed in glucose or sucrose media. The differential expression of these enzymes may provide means for the fungus to adapt to different conditions. (African Journal of Biotechnology: 2002 1(2): 64-66)

The production of cell wall–degrading enzymes (wall depolymerases) by plant pathogenic fungi is under catabolite (glucose) repression. In Saccharomyces cerevisiae , the SNF1 gene is required for expression of catabolite-repressed genes when glucose is limiting. An ortholog of SNF1 , ccSNF1 , was isolated from the maize pathogen Cochliobolus carbonum , and ccsnf1 mutants of HC toxin–producing (Tox2 + ) and HC toxin–nonproducing (Tox2 – ) strains were created by targeted gene replacement. Growth in vitro of the ccsnf1 mutants was reduced by 50 to 95% on complex carbon sources such as xylan, pectin, or purified maize cell walls. Growth on simple sugars was affected, depending on the sugar. Whereas growth on glucose, fructose, or sucrose was normal, growth on galactose, galacturonic acid, maltose, or xylose was somewhat reduced, and growth on arabinose was strongly reduced. Production of HC toxin was normal in the Tox2 + ccsnf1 mutant, as were conidiation, conidial morphology, conidial germination, and in vitro appressorium formation. Activities of secreted β-1,3-glucanase, pectinase, and xylanase in culture filtrates of the Tox2 + ccsnf1 mutant were reduced by 53, 24, and 65%, respectively. mRNA expression was downregulated under conditions that induced the following genes encoding secreted wall-degrading enzymes: XYL1 , XYL2 , XYL3 , XYL4 , XYP1 , ARF1 , MLG1 , EXG1 , PGN1 , and PGX1 . The Tox2 + ccsnf1 mutant was much less virulent on susceptible maize, forming fewer spreading lesions; however, the morphology of the lesions was unchanged. The Tox2 – ccsnf1 mutant also formed fewer non-spreading lesions, which also retained their normal morphology. The results indicate that ccSNF1 is required for biochemical processes important in pathogenesis by C. carbonum and suggest that penetration is the single most important step at which ccSNF1 is required. The specific biochemical processes controlled by ccSNF1 probably include, but are not necessarily restricted to, the ability to degrade polymers of the plant cell wall and to take up and metabolize the sugars produced.

The filamentous fungus Cochliobolus carbonum produces endo-α1,4-polygalacturonase (endoPG), exo-α1,4-polygalacturonase (exoPG), and pectin methylesterase when grown in culture on pectin. Residual activity in a pgn1 mutant (lacking endoPG) was due to exoPG activity, and the responsible protein has now been purified. After chemical deglycosylation, the molecular mass of the purified protein decreased from greater than 60 to 45 kDa. The gene that encodes exoPG, PGX1 , was isolated with PCR primers based on peptide sequences from the protein. The product of PGX1 , Pgx1p, has a predicted molecular mass of 48 kDa, 12 potential N-glycosylation sites, and 61% amino acid identity to an exoPG from the saprophytic fungus Aspergillus tubingensis . Strains of C. carbonum mutated in PGX1 were constructed by targeted gene disruption and by gene replacement. Growth of pgx1 mutant strains on pectin was reduced by ca. 20%, and they were still pathogenic on maize. A double pgn1/pgx1 mutant strain was constructed by crossing. The double mutant grew as well as the pgx1 single mutant on pectin and was still pathogenic despite having less than 1% of total wild-type PG activity. Double mutants retained a small amount of PG activity with the same cation-exchange retention time as Pgn1p and also pectin methylesterase and a PG activity associated with the mycelium. Continued growth of the pgn1/pgx1 mutant on pectin could be due to one or more of these residual activities.

Although best known for elaboration of a host-specific toxin, HC-toxin, the fungal maize pathogen Cochliobolus carbonum also produces a large number of cell wall degrading enzymes (CWDEs) capable of depolymerizing various components of the host cell wall. C. carbonum genes have been cloned that encode endo-polygalacturonase, exo-polygalacturonase, pectin methylesterase, four endo-xylanases, β-xylosidase, cellobiohydrolase, endo-β-1,4 glucanase, two exo-β-1,3-glucanases, two β-1,3-1,4 glucanases, two alkaline proteases and cutinase. Using a transformation-mediated gene disruption strategy, many of these genes have been mutated singly in pathogenic wild-type C. carbonum strains. In every case, the disruption of the CWDE gene and the concomitant loss of CWDE activity did not cause any detectable change in the pathogenicity or virulence of the mutant strain. Strains of C. carbonum that contain multiple mutations in related CWDEs are being constructed in order to examine whether the inability to degrade a particular cell wall component will affect disease development.

As part of our research to determine phylogenetic relationships of organisms within the phytobacterial species Xanthomonas campestris, we have examined the use of the random amplified polymorphic DNA (RAPD) technique. The objective of this aspect of our research was to determine if a valid cladistic character analysis could be carried out by direct comparison of RAPD products separated on ethidium bromide-stained agarose gels. RAPD products were amplified from 47 Xanthomonas campestris DNA templates using a single oligonucleotide primer. These RAPD products were compared and variation was characterized by Southern analysis of both RAPD products and genomic DNA of the 47 bacterial strains using two cloned RAPD products as probes. Analysis of the data set revealed that the RAPD products were not necessarily homologous or independent, crucial prerequisites for characters to be analyzed in a cladistic phylogenetic analysis. It has been commonly assumed that RAPD variation occurs due to insertion/deletion events or alterations in the primer binding site. Within our data set, we demonstrate absence phenotypes arising from the apparent absence of corresponding loci and also due to the preferred synthesis of alternative RAPD products from unrelated loci. These different types of variation are a reflection of different types of genotypic variation, and direct examination of RAPD products did not allow us to distinguish by which mechanism a particular absence phenotype arose. Although this may not be important for phenetic analyses, for analyses of homologous characters using a cladistic approach it is critical. We also detected unrelated, co-migrating RAPD products and multiple related RAPD products within reaction mixtures. These could both contribute to errors in estimates of similarity, important in any phylogenetic analysis. All of these characteristics of RAPD products should be taken into consideration when RAPD products are used for phylogenetic comparisons.

Specificity in many plant-pathogen interactions is determined by single genes in pathogen and host. The single locus for host-selective pathogenicity (TOX2) in the fungus Cochliobolus carbonum governs production of a cyclic tetrapeptide named HC-toxin. We have isolated a chromosomal region, 22 kilobases (kb) long, that contains a 15.7-kb open reading frame (HTS1) encoding a multifunctional cyclic peptide synthetase. The 22-kb chromosomal region is duplicated in toxin-producing isolates of the fungus but is completely absent from the genomes of toxin-nonproducing isolates. Mutants of the fungus with disruptions in both copies of HTS1, at either of two different sites within HTS1, were engineered by DNA-mediated transformation. Disruption of both copies at either site resulted in loss of ability to produce HC-toxin and loss of host-selective pathogenicity, but the mutants displayed different biochemical phenotypes depending on the site of disruption. The results demonstrate that TOX2 encodes, at least in part, a large, multifunctional biosynthetic enzyme and that the evolution of host range in C. carbonum involved the insertion or deletion of a large piece of chromosomal DNA.

Five regions of the Bradyrhizobium japonicum genome, which are transcribed at high levels in nitrogen-fixing soybean (Glycine max) nodules, were identified. None of these regions contained previously identified genes (e.g., nif, nod, and fix genes) that are known to be essential for development of functional nitrogen-fixing nodules. To assess the role of these regions in the development of the B. japonicum-soybean symbiosis, we cloned and used them to construct B. japonicum strains, in which large DNA segments (2.0-6.8 kilobases) containing the highly transcribed regions were deleted. The deletion strains were examined for symbiotic effectiveness and were found to be indistinguishable from the wild-type strain. Transcription of the cloned regions under a variety of physiological conditions and in several defined mutant B. japonicum strains was also examined. The transcriptional start sites for one pair of divergent transcripts were determined; the promoters do not contain any of the conserved sequences found in B. japonicum genes involved in symbiosis or nitrogen metabolism.

DNA sequences have been isolated that are expressed at high levels in bacteroids, the differentiated form of the soybean microsymbiont, Bradyrhizobium japonicum. Random-primed cDNA was synthesized using total RNA isolated from purified B. japonicum bacteroids or from cells grown in culture. When used directly to screen bacteriophage lambda libraries, these cDNA probes produced a high background hybridization signal due to sequence similarity between B. japonicum and E. coli ribosomal DNA (rDNA) operons. To reduce this background signal, the rDNA operon of B. japonicum was cloned and the rDNA plasmid DNA used in subtractive hybridization with the cDNA probes and as a competitor in hybridization solutions. This method greatly reduced the background signal in screening of genomic libraries and thus permitted the identification of twelve unique recombinant phage which contained sequences that are expressed at higher levels in B. japonicum bacteroids than in cells grown in culture.

A gene (PGN1) encoding extracellular endopolygalacturonase was isolated from the fungal maize pathogen Cochliobolus carbonum race 1. A probe was synthesized by polymerase chain reaction using oligonucleotides based on the endopolygalacturonase amino acid sequence. Genomic and cDNA copies of the gene were isolated and sequenced. The corresponding mRNA was present in C. carbonum grown on pectin but not on sucrose as carbon source. The single copy of PGN1 in C. carbonum was disrupted by homologous integration of a plasmid containing an internal fragment of the gene. Polygalacturonase activity in one transformant chosen for further analysis was 10% or 35% of the wild-type activity based on viscometric or reducing sugar assays, respectively. End product analysis indicated that the residual activity in the mutant was due to an exopolygalacturonase. Pathogenicity on maize of the mutant lacking endopolygalacturonase activity was qualitatively indistinguishable from the wild-type strain, indicating that in this disease interaction endopolygalacturonase is not required. Either pectin degradation is not critical to this interaction or exopolygalacturonase alone is sufficient.