WO2002092757A2 - Enzyme over-producing transgenic microorganisms - Google Patents

Abstract

A strain of Trichoderma reesei transformed to over-produce α-galactosidase. PCR primers are designed and used to link the α-galactosidase gene to the cellobiohydrolase promoter and terminator so that the α-galactosidase is expressed and expelled from the organism. Fermentation of the transformed organism produces α-galactosidase in the supernatant that is at a level of a least 20 IU per ml. Use of soybean hulls as an inducer increases the amount of α-galactosidase produced. The supernatant is filtered and concentrated to produce a liquid product that may be added to feed and food to improve conversion. The liquid concentrate may be dried, for example by spray drying, to produce a dried α-galactosidase product with an extended shelf life.

Description

ENZYME OVER-PRODUCING TRANSGENIC MICROORGANISMS

Background of the Invention The invention relates generally to transgenic microorganisms that over-produce enzymes and, more specifically, to genetic transformants of Trichoderma reesei (sometimes also identified as Trichoderma longibrachiatum) that over-produce enzymes, particularly α-galactosidase.

Supplementation of animal feeds with various hydrolytic enzymes, depending on the grains content of the feeds, has become an accepted and commercially viable practice in Europe and Asia that can result in enhanced feed conversion and increased profit after animal processing. The added enzymes can break down complex substrates in the feeds that monogastric animals don't completely digest, such as cellulose, hemicellulose, phytate and α-galactosides. Others have attempted to produce commercial quantities of certain of these enzymes by large-scale culture of transgenic organisms. Further, molecular and fermentation technologies can be applied toward development of value- added protein/enzyme products for human and animal nutrition. This application describes, inter alia, the construction of several gene expression cassettes that will be used to create transgenic microorganisms that over-express enzymes for use in such products. Many genera of eubacteria and fungi secrete hydrolytic enzymes that break down complex molecules in their environments to provide smaller, more readily assimilated substrates for growth (Coughlan, M. P. and L. G. Ljungdahl. 1988. Comparative biochemistry of fungal and bacterial cellulolytic enzyme systems, 11-30. In J.-P. Aubert, P. Beguin, and J. Millet (ed.), Biochemistry and Genetics of Cellulose Degradation. Academic Press, San Diego, USA; Leschine, S. B. 1995. Cellulose degradation in anaerobic environments. Annu. Rev. Microbiol. 49: 399-426). Exploitation of naturally occurring organisms by using classical genetic techniques to create mutants that overproduce secreted enzymes has led to the development of commercial products with applications in the food, feed, textile, baking and commodity chemicals industries (Eveleigh, D. E. and B. S. Montenecourt. 1979. Increasing yields of extracellular enzymes. Adv. Appl. Microbiol. 25: 57-74; Ghosh, A., B. K. Ghosh, H. Trimino-

Vazquez, D. E. Eveleigh and B. S. Montenecourt. 1984. Cellulase secretion from a hyper- cellulolytic mutant of Trichoderma reesei RUT-C30. Arch. Microbiol. 140: 126-133; Nieves, R. A., C. I. Ehrman, W. S. Adney, R. T. Elander and M. E. Himmel. 1998. Technical communication: survey and analysis of commercial cellulase preparations suitable for biomass conversion to ethanol. World J. Microbiol. Biotechnol. 14: 301-304). While classical genetic techniques have yielded filamentous fungal strains that can produce over 40 g protein per liter of culture broth, the profiles and proportions of secreted hydrolytic enzymes have not been changed substantially. Application of molecular genetic techniques to over-producing strains, however, has yielded transgenic strains that have substantially modified enzyme profiles and that can produce proteins from genes obtained from heterologous sources (Harkki, A., A. Mantyla, M. Penttila, S. Muttilainen, R. Bϋhler, P. Suominen, J. Knowles and H. Nevalainen. 1991. Genetic engineering of Trichoderma to produce strains with novel cellulase profiles. Enzyme Microb. Technol. 13: 227-233; van Gorcom, R. F. M., P. J. Punt and C. A. M. J. J. van den Hondel. 1994. Heterologous gene expression Aspergillus, 241-250. In K. A.

Powell et al (ed.), The Genus Aspergillus. Plenum Press, New York, USA). In particular, the strong Trichoderma reesei cbhl (cellobiohydrolase 1) gene promoter has been used extensively in a number of homologous and heterologous expression systems to drive expression of a transgene (Keranen, S. and M. Penttila. 1995. Production of recombinant proteins in the filamentous fungus Trichoderma reesei. Curr. Biol. 6: 534-537). Subtle structural features of a gene sequence, stability and translational efficiency of mRNA, codon usage biases, protein folding, protein degradation by host cell proteases and potential toxicity of an over-produced protein to the host can have unpredictable negative effects on production of a desired protein. Cell growth characteristics, gene expression levels, intracellular and secreted expression and post-translational modifications also impact the ability to over-produce biologically active recombinant proteins and cannot always be predicted and controlled. In particular, design and construction of recombinant DNA cassettes to over-express a particular protein in a specific host organism can be a major determinant of success (Makrides, S. C. 1996. Strategies for achieving high-level expression of genes in Escherichia coli. Microbiol. Rev. 60: 512-538).

The genetic elements of a recombinant gene expression cassette, chosen for expression of any particular target protein in a specific host, must include: 1) regulation of transcription (promoters and terminators); 2) ribosome binding and translation initiation sites; 3) targeting of protein expression to the cytoplasm or culture medium; and 4) a marker gene that will allow selection of genetic transformants (Makrides, ibid.). A high level of recombinant gene transcription is essential to achieve high levels of recombinant protein expression. Switching a gene's promoter from the normal, endogenous one to a different, highly active promoter will generally result in a several fold increase of expressed protein. The ability to regulate transcription is also desirable but not always necessary. In addition, placing a cleavable secretion signal sequence at the N-terminal of a recombinant protein will likely result in secretion of the protein into the culture broth and will simplify protein recovery. While antibiotic resistance genes have usually been used as genetic markers for selection of transformants, the concern about widespread dissemination of resistance genes in organisms pathogenic to humans has led to development of genetic markers based on complementation of auxotrophic mutant host strains. The expression cassettes described in this specification utilized the strong T. reesei cbhl gene promoter (with the Cbh I protein secretion signal) and terminator that were operably linked to the T. reesei agll (α-galactosidase 1) gene. The Escherichia coli hph (hygromycin phosphotransferase) gene was genetically linked to the expression cassettes for use as a selectable marker (Carroll, A. M., J. A. Sweigard and B. Valent. 1994. Improved vectors for selecting resistance to hygromycin. Fungal Genet. Newsl. 41 : 22).

This application also describes the use of electroporation as the genetic transformation procedure to deliver and stably incorporate the gene cassettes into the genomes of host organisms. The development of gene transfer systems for the introduction of exogenous DNA into cells has revolutionized the field of genetics and allowed manipulation of protein expression in a wide range of microorganisms. A few eubacterial species (such as Bacillus subtilis and Streptococcus pneumoniae) are naturally competent to take up DNA, and for a few other species (such as Escherichia coli) competence can be chemically induced. The use of electric current to facilitate DNA uptake (electroporation or electrotransformation) has dramatically expanded the range of microorganisms that can be transformed and has allowed the genetic manipulation of a large number of previously recalcitrant organisms (Fiedler, S. and R. Wirth. 1988. Transformation of bacteria with plasmid DNA by electroporation. Anal. Biochem. 170: 38-44). The application of a voltage gradient to a concentrated sample of cells mixed with DNA results in transient formation of pores in the lipid membrane of the cells, which allows the exogenous DNA to passively diffuse into the cytoplasm. Incorporation of the DNA into the cell genome or maintenance by autonomous replication in the cells occurs by molecular mechanisms that have not been fully elucidated. Research into genetic transformation of filamentous fungi in the 1980's resulted in development of several reproducible procedures for Neurospora, Trichoderma and Aspergillus (Penttila, M., H. Nevalainen, M. Ratto, E. Salminen and J. Knowles. 1987. A versatile transformation system for the cellulolytic fungus Trichoderma reesei. Gene 61 : 155-164). The techniques generally involved enzymatic digestion of cell wall components in an osmotically stabilized solution to yield protoplasts. The protoplasts were then treated with DNA in a solution of CaCl₂ and polyethylene glycol and the DNA was taken up by an unknown mechanism. Finally, the cell wall was allowed to regenerate and transformed strains were grown on a selective medium. More recently, electroporation of protoplasts has also resulted in DNA uptake but has not provided increased transformation frequency in filamentous fungi (Lemke, P. A. and M. Peng. 1995. Genetic manipulation of fungi by DNA-mediated transformation, 109-139. In U. Kϋck (ed.), The Mycota II: Genetics and Biotechnology. Springer- Verlag, Berlin, Germany). Protoplast- based transformation procedures are prone to unpredictable variations due to side enzymatic activities in the solutions used to digest the cell wall and due to the fragility of protoplasts and difficulty with cell wall regeneration.

Electroporation of cells or protoplasts mixed with DNA is usually performed in a small cuvette that has metal electrodes on either side of the sample. The cells are at a high concentration (> 10⁹/ml) and are suspended in a low conductivity buffer such as water, 10% glycerol or IM sorbitol. The electrical pulse is supplied by a capacitor discharge that has an exponential decay waveform (Fiedler, et al., supra). The voltage gradient between the electrodes is the electric field (E), which is determined by V/d where V is the applied voltage, and d is the electrode gap in the sample cuvette. The electric field is the most important parameter in determining the success of electroporation and for each organism being transformed must be determined empirically. When a charge is passed through a sample the voltage across the electrodes rises quickly to a peak initial voltage (Vo) and then declines over time. The length of time it takes for the voltage to decline to 1/e (-37%) of Vo is called the time constant and is the second parameter of electroporation that can be adjusted to enhance transformation efficiency. The time constant, given in milliseconds, is the product of resistance (in ohms) and capacitance (in Farads) both of which can be set on an electroporation apparatus. The time constant is also affected by the conductivity of the sample and the resistance and capacitance settings to achieve optimum transformation efficiency must be determined empirically. We describe here the optimization of electrical parameters for genetic transformation of Trichoderma reesei, Aspergillus niger and A. awamori conidia by electroporation, and the impact of growth conditions and topology of the transforming DNA on transformation efficiency. Rather than using protoplasts for transformation we adapted conidia electroporation procedures developed for A. nidulans and N. crassa (Chakraborty, B. N., N. A. Patterson and M. Kapoor. 1991. An electroporation-based system for high-efficiency transformation of germinated conidia of filamentous fungi. Can. J. Microbiol. 37: 858-863; Sanchez, O. and J. Aguirre. 1996. Efficient transformation of Aspergillus nidulans by electroporation of germinated conidia. Fungal Genet. Newslett. 43: 48-51). Summary of the Invention

The invention consists of genetic transformants of Trichoderma reesei that overproduce enzymes, particularly α-galactosidase. Untransformed or wild-type Trichoderma typically produce between about 0.01 and 0.4 IU/ml of α-galactosidase (Zeilinger, S. et al. Conditions of formation, purification, and characterization of an α-galactosidase of Trichoderma reesei RUT C-30. Appl. Environ. Microbiol. 59: 1347-53). The genetic transformants were prepared by creating a gene expression cassette by recombinant polymerase chain reaction (PCR) and transforming the T. reesei with the gene expression cassette. The transformed T. reesei were grown in a fermenter, induced to produce the enzyme, and the enzyme was extracted from the fermentation broth. More specifically, a gene expression cassette was prepared that incorporates (a)

DNA that encodes the amino acid sequence for the target enzyme, (b) DNA sequences that can regulate transcription of messenger RNA from the target in a host microorganism, and (c) a second gene that encodes a selectable marker. In the preferred embodiment, Trichoderma reesei agll (α-galactosidase I), cbhl (cellobiohydrolase I), and act (actin) genes were cloned by PCR and elements of each were used to construct gene expression cassettes by recombinant PCR. The cassettes made use of the cbhl gene promoter and terminator regions to drive transcription of agll. The Escherichia coli hph gene, which encodes resistance to the antibiotic hygromycin B, was genetically linked to each expression cassette for use as a selectable marker in genetic transformation. Trichoderma reesei conidia were genetically transformed by electroporation with the α-galactosidase gene expression cassette. The conidia and a quantity of the gene expression cassette were placed in an electroporation cuvette and a voltage pulse of specified duration and strength was applied. Optimum electrical conditions for yield of Trichoderma transformants were an electrical field of 15 kV/cm applied to 40 μl samples in 0.1 cm gap cuvettes with a 15 ms time constant determined by 50 μFarad capacitance and 300 Ohms resistance settings. Both linear and circular forms of the DNA of the gene expression cassette were used in experiments. The linear transforming DNA was more efficient than circular for the yield of Trichoderma transformants. The electroporation technique was also applied to obtain transformants of Aspergillus niger conidia and Aspergillus awamori conidia, and slightly different optimum electrical conditions were found. Treatment of partially germinated Aspergillus spores with hydrolytic enzymes increased the transformation efficiency. Genetic transformation was confirmed by PCR amplification of DNA extracted from conidia or mycelium using primers specific for the transforming DNA. The transformed Trichoderma are used in the fermentative production of the target protein. The fermentation process may be batch fermentation, fed-batch fermentation or continuous fermentation, depending on the host organism and the preferred downstream processing. In the preferred embodiment, batch fermentation is used wherein the target protein is secreted by the Trichoderma into the fermentation broth. The target protein, in the preferred embodiment α-galactosidase, is recovered from the broth by first removal of the cells either by centrifugation or filtration, concentration of the liquid portion by removal of water, and spray drying to produce a dry product.

The term "gene expression cassette" in this specification means a linear or circular structural gene with DNA that encodes the amino acid sequence for the target protein (either homologous or heterologous to the host microorganism), DNA sequences that can regulate transcription of messenger RNA from the target in the host microorganism, and a second gene construct that encodes a selectable marker.

Brief Description of the Drawings Fig. 1 is a photograph of agarose gel electrophoresis of PCR products amplified from T. reesei RUT-C30 total DNA, wherein lane 1 is DNA size markers, lane 2 is the amplified cbhl gene (-4.2 kbp) and lane 3 is the amplified agll gene (-1.6 kbp).

Fig. 2 is a schematic drawing of the process whereby recombinant PCR is used to change a gene regulatory sequence that is operably linked to a structural gene.

Fig. 3 is a photograph of agarose gel electrophoresis of PCR products amplified from cloned T. reesei genes Fig. 4 is a photograph of agarose gel electrophoresis of T. reesei agll gene expression cassette assembled by recombinant PCR

Fig. 5 is a schematic drawing of the α-galactosidase expression cassette, plasmid pKBE2001, shown in detail including junctions regions created by recombinant PCR. Fig. 6 is a photograph of the agarose gel electrophoresis pattern obtained using

PCR products amplified from genomic DNA of putative T. reesei transgenic strains following electroporation of T. reesei RUT-C30 with pKBE2001 and selection for resistance to hygromycin B.

Fig. 7 is a photograph of the agarose gel electrophoresis pattern obtained using PCR products amplified from genomic DNA of putative A. awamori ATCC 11358 and A. niger KASNl transgenic strains following electroporation with pKBE2002 and selection for resistance to hygromycin B.

Fig. 8 is a flowchart of the fermentation process for producing commercial quantities of α-galactosidase produced by the transformed organism of the present invention.

Detailed Description of a Preferred Embodiment Experiment 1 - Preparation of a Cassette A. Fungal and bacterial strains, media and culture conditions Trichoderma reesei ATCC 56765 (RUT-C30) was obtained from the American Type Culture Collection, Manassas, VA. Trichoderma strains were normally maintained as spore cultures on potato dextrose agar slants (PDA; Sigma, St. Louis, MO). T. reesei RUT-C30 was grown in Gaugy's PM medium, for isolation of total DNA, which contained, per liter, 40.0 g glucose, 2.0 g yeast extract, 3.0 g NaNO₃, 0.5 g KC1, 0.5 g MgSO₄7 H₂O, 10 mg FeSO .7H₂O and 1.0 g KH₂PO₄ (Gaugy, D. and M. Fevre. 1985. Regeneration and reversion of protoplasts from different species of Penicillium. Microbios 44: 285-293). Trichoderma strains were also grown on V8 agar slants that contained, per liter, 200 ml V8 juice (Campbell Soup Company, Camden, NJ), 1.5 g CaCO₃ and 15 g Bacto™ agar (Becton Dickinson, Co., Sparks, MD). Fungal strains were also grown on InterLink Biotechnologies, L.L.C. ISP2 medium which contained, per liter, 10.0 g malt extract, 5.0 g yeast extract, 1.0 g Instant Ocean (Aquarium Systems, Mentor, OH), 10.0 g potato flour, 5.0 g glucose and 20.0 g Bacto™ agar. Fungal strains were grown at 29° C and broth cultures were shaken at 180-200 rpm.

E. coli strains XL 1 -blue MRF' (Stratagene, La Jolla, CA) and DH5α were routinely used for all cloning, vector construction and plasmid preparation procedures. E. coli strains were grown in Luria-Burtani broth (LB) which contained, per liter, 10.0 g NaCl, 10.0 g tryptone and 5.0 g yeast extract. LB was solidified to make plates by including 15 g Bacto™ agar per liter. Bacterial strains were grown at 37° C and broth cultures were shaken at 300 rpm.

Plasmid pCB1003, which contains the cloned Escherichia coli hph (hygromycin phosphotransferase) gene operably linked to the Aspergillus nidulans trpC gene promoter, was obtained from the Fungal Genetics Stock Center (University of Kansas, Kansas City, Kansas) (Carroll, et al., supra). Antibiotics were added to growth media as required (10 μg tetracycline/ml or 50 μg ampicillin/ml for E. coli; 100 μg hygromycin B/ml for fungal strains) and all media and reagents were sterilized by autoclaving. B. Cloning of Genes

The E. coli plasmid vector pBluescript^® II KS⁺ (Stratagene, La Jolla, CA) was routinely used for cloning and for assembly of expression vectors. DNA restriction digestions, ligations, agarose gel purification and quantification used standard molecular biology procedures (Sambrook, J., E. F. Fritsch and T. Maniatis. 1989. Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York, USA). Commercially available kits were used for plasmid purification, cleanup of PCR and DNA modification reactions, and recovery of DNA from agarose gels (Qiagen, Inc., Valencia, CA). The Microchemical Facility at Iowa State University performed DNA sequencing and sequences were assembled and analyzed using the Lasergene suite of software programs (DNASTAR, Inc., Madison, WI). The OLIGO^® software program (Molecular Biology Insights, Inc., Cascade, CO) was used to design DNA primers for PCR and for DNA sequencing, and to determine thermal cycling conditions for PCR. DNA primer design was based on gene sequence data obtained from the GenBank public database (http://www.ncbi.nlm.nih.gov/) or on data obtained by contract with the Microchemical Facility at Iowa State University.

Thermal stable DNA polymerases were purchased from several different sources and were used according to the manufacturers recommendations with reaction buffers provided by the manufacturers. The polymerases included: cloned Pfu, PfuTurbo™, TaqPlus Precision^® and Taq2000™ (Stratagene, La Jolla, CA); and JumpStart™ Taq

DNA Polymerase (Sigma, St. Louis, MO). Fifty μl PCR reactions typically included a total of 250 ng template DNA, 25 pmol of each primer, 10 nmol of each deoxynucleotide and 1.25 to 2.5 units of polymerase; DMSO (5%) was sometimes added to aid in template denaturation. Thermal cycling conditions were typically: 20 to 35 cycles of denaturation at 92-95° C for 10 s to 1 min, primer annealing at an appropriate temperature for 15 s to 1

min, and extension at 72° C for 1 min/kb of amplified DNA (Taq polymerases) or 2 min/kb of amplified DNA (Pfu polymerases). PCR products were separated by electrophoresis through 0.6-2.0% agarose gels and were visualized and photographed by UV (305 nm wavelength) transillumination after staining with ethidium bromide. C. Isolation of Total DNA from Trichoderma reesei RUT-C30 Total DNA was isolated from an overnight broth culture of T. reesei RUT-C30. Fifty ml of Gaugy's PM medium was inoculated with about 10⁷ conidia and grown for 72 h at 29° C, 200 rpm. One ml of the 72 h culture was inoculated into 50 ml of fresh

Gaugy's PM and incubated for 22 h at 29° C, 200 rpm. The mycelium was recovered from 25 ml of the overnight culture by centrifugation and was resuspended in 5 ml of 1.2 M sorbitol, 100 mM potassium phosphate buffer (pH 5.6) that contained 5 mg Novozyme 234/ ml (Sigma, St. Louis, MO), 5 mg Driselase/ml (Sigma, St. Louis, MO) and 100 μg

RNase/ ml (Sigma, St. Louis, MO). After incubation for 2 h at 30° C, 200 rpm the protoplasts were pelleted by centrifugation at 4,000 x g for 10 min. The protoplasts were then resuspended in 1 ml of 1.2 M sorbitol, 50 mM Tris buffer (pH 8), 50 mM EDTA and 50 μl of 20 mg proteinase K/ml (Roche Molecular Biochemicals, Indianapolis, IN) was

added. The sample was divided into two equal parts (-600 μl each) and 60 μl of 20% SDS was gently mixed into each part by inverting 15 times. The samples were extracted twice with 600 μl of phenol (pH 7.9), and 100 μl of 10 mM Tris buffer (pH 8), 0.1 mM EDTA (TE) was added to each before the second extraction as required for proper phase separation. The samples were then extracted twice with phenol :CHCl₃:isoamyl alcohol (25:24:1) and DNA was precipitated from the aqueous phase by adding 1/10* volume (-75 μl) of 3 M sodium acetate (pH 5.2) and 2 volumes of 95% ethanol. The precipitated DNA was spooled onto the end of a Pasteur pipette, whose end had been heat-sealed and formed into a hook, and allowed to air-dry for about 1 min. The DNA was dissolved in

250 μl sterile water, precipitated, and spooled a second time. After a third precipitation

and spooling the DNA was precipitated a fourth time and stored overnight at -20° C. The

DNA was pelleted by centrifugation for 20 min at 14,000 x g, washed once using 500 μl

of 70% ethanol, air dried for 10 min and resuspended in 400 μl TE. Absorbance, at 260 nm, of a 1/200 dilution of the DNA sample revealed that a total of 112 μg of total DNA (genomic and mitochondrial) had been recovered and that the concentration was 280 ng T. reesei RUT-C30 DNA/μl.

D. Cloning of Genes The gene for our target enzyme, agll (α-galactosidase), was cloned by PCR amplification of T. reesei RUT-C30 total DNA. Primer pairs that incorporated restriction enzyme sites in their 5' termini were used for amplification, which allowed for directional cloning into an E. coli vector (Scharf, S. J. 1990. Cloning with PCR, 84-91. In M. A. Innis, D. H. Gelfand, J. J. Sninsky, and T. J. White (ed.), PCR Protocols: A Guide to Methods and Applications. Academic Press, San Diego, USA). The T. reesei cbhl (cellobiohydrolase I) and act (actin) genes were also cloned and were subsequently cannibalized for gene regulatory sequences which were used in our expression cassettes to drive transcription of the structural genes. The sequences of PCR primer pairs used for cloning T. reesei agll, cbhl and act are shown in Table 1.

TABLE 1 : DNA PRIMER PAIRS USED FOR CLONING OF PCR AMPLIFIED

GENES

Genes were amplified from T. reesei RUT-C30 total DNA by PCR.

GenBank DNA sequence data were accessed at the National Center for Biotechnology Information web site http://www.ncbi.nlm.nih.gov/.

Underlined sequences are restriction enzymes sites designed into the primers to facilitate directional cloning: GGATCC, Kpn I; TCTAGA, Xba I; AAGCTT, Hind III; GAATTC, Eco R I; GGATCC,

BamH I.

Optimal temperature for the primers to anneal to the T. reesei RUT-C30 template DNA in the PCR amplification.

Predicted size of the successful PCR amplification product.

Margolles -Clark, E., M. Tenkanen, E. Luonteri and M. Penttila. 1996. Three α-galactosidase genes of

Trichoderma reesei cloned by expression in yeast. Eur. J. Biochem . 240: 104-111.

Takashima, S., H. Iikura, A. Nakamura, H. Masaki and T. Uozumi. 1996. Analysis of Crel binding sites in the Trichoderma reesei cbhl upstream region. FEMS Microbiol. Lett . 145: 361-366.

Wey, T. T., T. H. Hseu and L. Huang. 1994. Molecular cloning and sequence analysis of the cellobiolhydrolase I gene from Trichoderma koningii G-39. Curr. Microbiol . 28: 31 -39.

Matheucci Jr., E., F. Henrique -Silva, S. El-Gogary, C. H. Rossini, A. Leite, J. E. Vera, J. C. Urioste, O.

Crivellaro and H. El -Dorry. 1995. Structure, organization and promoter expression of the actin - encoding gene in Trichoderma reesei. Gene 161 : 103-106.

Agarose gel electrophoresis of PCR products amplified from T. reesei RUT-C30 total DNA is shown in Figure 1. After PCR, 2 μl samples were loaded into wells of a 0.6% agarose gel and electrophoresed for 25 min at 20 Volts/cm in 0.5% TAE buffer. Gels were stained with 1.5 μg ethidium bromide/ml and nucleic acid bands were visualized by UV transillumination. In Fig. 1, lane 1 is DNA size markers, lane 2 is the amplified cbhl gene (-4.2 kbp) and lane 2 is the amplified agll gene (-1.6 kbp). Cloning of all genes and regulatory elements was confirmed by DNA sequencing. E. Construction of target enzyme expression vectors The DNA cassette for expression of T. reesei Agll (α-galactosidase) was assembled by recombinant DNA amplification using long overlapping PCR primers that were designed to create precise, novel junctions between gene regulatory sequences and the structural gene (Tables 2 and 3; Fig. 2) (Higuchi, R. 1990. Recombinant PCR, 177- 183. In M. A. Innis, D. H. Gelfand, J. J. Sninsky, and T. J. White (ed.), PCR Protocols: A Guide to Methods and Applications. Academic Press, San Diego, USA). The T. reesei cbhl gene promoter and terminator were operably linked to the agll gene by a series of PCR reactions that replaced the endogenous promoter and terminator sequences.

One PCR reaction used primers cbagl with cbag2 to add DNA sequences to the 3' end of the promoter Ycbhl that were identical to the 5' end of the agll structural gene. The Ycbhl fragment encompassed about 500 nucleotides upstream of the translation initiation codon and the first 51 nucleotides of the cbhl structural gene, which encode the Cbh I signal sequence. Another PCR reaction used the primers cbag5 with cbagό to add DNA sequences to the 5' end of the terminator Υcbhl that were identical to the 3' end of the agll structural gene . The first PCR reaction also added a restriction enzyme site to the 5' end of Pcbhl and to the 3' end of Υcbhl to facilitate linking of the final construct to the hygromycin B resistance marker gene in a plasmid vector. Likewise, DNA sequences were added to the 5' end of the agll structural gene that were identical to the 3' end of Fcbhl, and to the 3' end of agll that were identical to the 5' end of Υcbhl (primers cbag3 and cbag4). The nucleotide sequences of the cbagl-6 primers set out in Table 2 and Figure 3 are identified as SEQ ID NO. 1-6, respectively. In Fig. 3, agarose gel electrophoresis of PCR products amplified from cloned T. reesei genes is illustrated. After PCR, samples were loaded into wells of a 0.6% agarose gel and electrophoresed for 25 min at 20 Volts/cm in 0.5% TAE buffer. Gels were stained with 1.5 μg ethidium bromide/ml and nucleic acid bands were visualized by UV transillumination. Lane 1 is DNA size markers. Lanes 2 and 3 are, respectively, 1 and 5 μl samples of the amplification reaction that used the cloned cbhl gene as template with primers cbag 1 and cbag 2 to yield the 568 bp Ϋcbhl fragment. Lanes 4 and 5 are, respectively, 1 and 5 μl samples of the amplification reaction that used the cloned agll gene as template with primers cbag 3 and cbag 4 to yield the 1.62 kbp agll structural gene fragment. Lanes 6 and 7 are, respectively, 1 and 5 μl samples of the amplification reaction that used the cloned cbhl gene as template with primers cbag 5 and cbag 6 to yield the 662 bp Υcbhl fragment.

TABLE 2: DNA PRIMERS USED FOR RECOMBINANT PCR TO ASSEMBLE THE T. reesei agll EXPRESSION CASSETTE

¹ Underlined sequences are restriction enzymes sites designed into the primers to facilitate directional cloning: CTGCAG, Pst I; CGATCG, Pvu I; GGATCC, BamH I. Sequences in bold are for the agll structural gene.

² Optimal temperature for the primers to anneal to the DNA template in the PCR amplification.

³ For each primer pair, the first junction given is the newly created DNA sequence at the 5' end of the PCR product and the second junction is at the 3' end.

⁴ Predicted size of the successful PCR amplification product.

The products of the first PCR reactions were purified by agarose gel electrophoresis and samples of the promoter, structural gene and terminator fragments were combined in a 1 : 1 : 1 molar ratio and used as templates in a second set of PCR reactions. The 5' and 3' sequences that had been added to the individual genetic elements served as 'adapters' to create the novel, specific junctions between the structural genes and the regulatory sequences. For example, the Ycbhl, agll, and Ycbhl products from the first set of PCR reactions were mixed and amplified using the cbagl and cbagό primers to assemble the 2,795 bp agll cassette (Fig. 4).

Fig. 4 is an agarose gel electrophoresis of the T. reesei agll gene expression cassette assembled by recombinant PCR. The Ycbhl, agll, and Ycbhl gene fragments recovered from previous amplification reactions (Fig. 3) were purified and used together as templates with the cbagl and cbagό primers. After PCR, samples were loaded into wells of a 0.6% agarose gel and electrophoresed for 25 min at 20 Volts/cm in 0.5% TAE buffer. Gels were stained with 1.5 μg ethidium bromide/ml and nucleic acid bands were visualized by UV transillumination. Lane 1 is DNA size markers. Lanes 2 and 3, respectively, are 1 μl and 5 μl samples of the PCR reaction. The assembled cassette is clearly seen migrating to about 2.8 kbp. In a similar manner, Pact (the actin gene promoter) and Ycbhl were operably linked to the hph gene contained in plasmid pCB1003 and restriction enzyme sites were added (Table 3).

TABLE 3: DNA PRIMERS USED FOR RECOMBINANT PCR TO CLONE AND ASSEMBLE THE SELECTABLE H H MARKER

Underlined sequences are restriction enzymes sites designed into the primers to facilitate directional cloning: GGATCC, BamH I; TCTAGA, Xba I; CGATCG, Pvu I; GGATCC, BamH I. Sequences in bold are for the hph structural gene.

Optimal temperature fo r the primers to anneal to the DNA template in the PCR amplification.

For each primer pair, the first junction given is the newly created DNA sequence at the 5' end of the

PCR product and the second junction is at the 3' end.

Predicted size of the successful PCR amplification product.

The assembled agll and hph cassettes were cloned separately, using the engineered Pst I and BamH I, and BamH I and Xba I restriction sites, respectively, and correct formation of the novel junctions was confirmed by DNA sequencing. The agll gene cassette was then subcloned into the plasmid that contained the hph marker cassette using an intervening, engineered BamH I site. The target gene and marker gene cassettes were linked such that the genes would be concurrently transcribed and the target gene was located upstream of the hph marker gene. Fig. 5 depicts plasmid pKBE2001 which contains the T. reesei agll expression cassette linked to the hph marker gene cassette by an intervening BamH I site and the entire construct contained in pBluescript II KS⁺ between the Pst I and Xba I sites. The genetic components of two expression cassettes that were constructed are given in Table 4.

TABLE 4: GENETIC COMPONENTS OF SEVERAL GENETICALLY ENGINEERED

EXPRESSION CASSETTES

1 Expression cassettes were assembled in the pBluescript^® II S⁺ (Stratagene, La Jolla, CA) multiple cloning site.

Egll is the T. reesei endocellulase 1 enzyme and Agll is the T. reesei α-galactosidase 1 enzyme.

Individual components of the expression cassettes were cloned by PCR and assembled by recombinant

PCR using overlapping 'adapter' primers as described.

4 Total size of each fully assembled, genetically engineered expression cassette.

F. Discussion We designed and constructed a DNA gene cassette to over-express the T. reesei Agl 1 (α-galactosidase) enzymes in a T. reesei, or other filamentous fungus, host. We chose T. reesei as the host organism for genetic transformation and protein expression because of its long history of use to safely produce several different proteins for commercial applications, the possibility of achieving high levels of protein expression, and the availability of genetic tools and transformation procedures. Aspergillus species are also widely used as hosts for homologous and heterologous protein expression.

The promoter sequence used to drive recombinant gene expression in a transgenic host is a critical element. In general, filamentous fungal gene promoters are subject to complex global and specific gene regulation networks that are not clearly described or understood. Molecular mechanisms of gene induction and repression probably include proteins that bind to specific nucleotide motifs in promoters, and transcription factors that interact with the RNA transcription holoenzyme complex. The T. reesei cbhl promoter contains nucleotide sequence motifs that are involved in binding proteins that repress and induce gene transcription (Ilmen, M., M. -L. Onnela, S. Klemsdal, S. Keranen and M. Penttila. 1996. Functional analysis of the cellobiohydrolase I promoter of the filamentous fungus Trichoderma reesei. Mol. Gen. Genet. 253: 303-314; van Gorcom, et al., supra). At least six binding sites for the catabolite repressor Crel are present between 700 and 1500 bp upstream of the protein coding region, and sequences within 200 bp upstream of the translation initiation site are required for induction (Henrique-Silva, F., S. El-Gogary, J. C. Carle-Urioste, E. Matheucci Jr., O. Crivellaro and H. El-Dorry. 1996. Two regulatory regions controlling basal and cellulose-induced expression of the gene encoding cellobiohydrolase I of Trichoderma reesei are adjacent to its TATA box. Biochem. Biophys. Res. Commun. 228: 229-237). The cbhl promoter fragment used in the present expression cassette does not include any of the catabolite repressor binding sites. The cbhl promoter fragment used in the present expression cassette includes all sites known to be required for induction of transcription. Binding sites involved in nitrogen metabolism regulation and pH regulation might also be present and very likely other, as yet unidentified, sites will be found that regulate cbhl transcription in a variety of growth conditions. Using the cbhl promoter to drive recombinant gene expression must take into consideration culture conditions that will not repress gene transcription and that will induce transcription. In contrast, very little is known about the biological significance of primary or secondary structures found in gene terminator sequences in filamentous fungi. The α-galactosidase expression cassette described are used in genetic transformation experiments to make transgenic filamentous fungi that over-produce α- galactosidase.

Experiment 2 - Electroporation A. Fungal and bacterial strains, media and culture conditions

Trichoderma reesei ATCC 56765 (RUT-C30) and Aspergillus awamori ATCC 11358 were both obtained from the American Type Culture Collection, Manassas, VA. Aspergillus niger strain KASNl was isolated by the biochemistry group of Kemin Biotechnology, L.C. Trichoderma strains were normally maintained as conidia cultures on potato dextrose agar slants (PDA; Sigma, St. Louis, MO). Aspergillus strains were maintained as conidia cultures and Trichoderma strains were also grown on V8 agar slants which contained, per liter, 200 ml V8 juice (Campbell Soup Company, Camden, NJ), 1.5 g CaCO₃ and 15 g Bacto™ agar (Becton Dickinson, Co., Sparks, MD). Fungal strains were also grown on InterLink Biotechnologies, L.L.C., ISP2 medium which contained, per liter, 10.0 g malt extract, 5.0 g yeast extract, 1.0 g Instant Ocean

(Aquarium Systems, Mentor, OH), 10.0 g potato flour, 5.0 g glucose and 20.0 g Bacto™ agar. Media used for selection of fungal transformants and for viability counts following electroporation also contained 0.1% Triton X-100 to restrict colony growth. Fungal strains were grown at 29° C and broth cultures were shaken at 180-200 rpm. E. coli strains XL1 -blue MRF' (Stratagene, La Jolla, CA) and DH5α were routinely used for all cloning, vector construction and plasmid preparation procedures. E. coli strains were grown in Luria-Burtani broth (LB) which contained, per liter, 10.0 g NaCl, 10.0 g tryptone and 5.0 g yeast extract. LB was solidified to make plates by including 15 g Bacto™ agar per liter. Bacterial strains were grown at 37° C and broth cultures were shaken at 300 rpm. Plasmid pCB1003, which contains the cloned Escherichia coli hph (hygromycin phosphotransferase) gene operably linked to the Aspergillus nidulans trpC gene promoter, was obtained from the Fungal Genetics Stock Center (University of Kansas, Kansas City, KS) (Sweigard, J., F. Chumley, A. Carroll, L. Farrall and B. Valent. 1997. A series of vectors for fungal transformation. Fungal Genet. Newslett. 44: 52-53). Antibiotics were added to growth media as required (10 μg tetracycline/ml or 50 μg ampicillin/ml for E. coli; 100 μg hygromycin B/ml for fungal strains) and all media and reagents were sterilized by autoclaving.

B. Preparation of target enzyme expression cassettes for electroporation A DNA cassette for expression of T. reesei Agl 1 (α-galactosidase) was assembled as discussed above by recombinant DNA amplification using long overlapping PCR primers that were designed to create precise, novel junctions between gene regulatory sequences and the structural gene. The E. coli plasmid vector pBluescript II KS⁺ (Stratagene, La Jolla, CA) was used for cloning of the assembled cassettes. All DNA manipulations and analyses used standard procedures and commercially available kits were used for plasmid purification, cleanup of PCR and DNA modification reactions, and recovery of DNA from agarose gels (Qiagen, Inc., Valencia, CA) (Sambrook, et al., supra). The genetic components of two expression cassettes that were made are given in Table 4. Prior to electroporation of T. reesei, the expression cassette was linearized and released from pKBE2001 by digestion with the restriction enzymes Pst I and Xba I. Intact circular pKBE2002 was used for electroporation of Aspergillus.

C. Electroporation of Trichoderma reesei RUT-C30 Conidia T. reesei RUT-C30 conidia, either freshly harvested or from frozen stocks, were inoculated onto V8 agar or ISP2 agar slants and grown for 14 days at 30° C. The conidia were harvested by adding 5 ml of sterile water to the slant and gently rubbing the surface of the culture with a pipette tip. Large debris was removed from the conidia suspension by filtration through Miracloth (Calbiochem, San Diego, CA) and a 1/100 dilution was counted using a hemacytometer. Conidia collected from two agar slants (10 ml suspension) were centrifuged for 10 min at 3500 x g and were washed twice with cold 1.2 M sorbitol. After the final wash the conidia were resuspended to 2.5 x 10⁹ conidia/ml in cold electroporation buffer (1.2 M sorbitol, 1% PEG) and 40 μl samples were put in 1.5

ml microcentrifuge tubes on ice. Four to 8 μg of linearized pKBE2001 was mixed with each sample, the samples were transferred to 0.1 cm gap electroporation cuvettes and the samples were kept on ice for 5 min prior to electroporation. The effects of applied voltage and pulse duration, supplied by a Bio-Rad Gene Pulser II and Pulse Controller Plus (Bio-Rad, Hercules, CA), on conidia transformation were tested using multiple samples in a number of different experiments. Following electroporation, conidia were washed from the electroporation cuvettes using 1 ml of cold 1.2 M sorbitol, and 500 μl samples were

spread onto PDA plates that contained 75 μg to 100 μg hygromycin B/ml to select for

transformants. Duplicate 100 μl samples of a 10^"6 dilution of non-electroporated conidia were spread onto PDA plates and resulting colony counts were used with hemacytometer counts of conidia to calculate percent germination ((colony count/hemacytometer count) x 100 = % germination). Duplicate 100 μl samples of a 10^"5 dilution of electroporated conidia were spread onto PDA plates and resulting colony counts were used with those of non-electroporated conidia to determine percentage killing ((1 - (electroporated colony count non-electroporated colony count)) x 100 = % killing). Colonies were usually visible on selection plates after three days of incubation at 30° C and new colonies appeared up to 11 days of incubation. D. Electroporation of Aspergillus Conidia A. niger KASNl or A. awamori ATCC 11358 conidia, either freshly harvested or from frozen stocks, were inoculated onto V8 agar slants and grown for 5 to 31 days at 30° C. The conidia were harvested by adding 5 ml of sterile 0.1% Tween 80 to the slant and gently rubbing the surface of the culture with a pipette tip. Clumps of conidia were disrupted by vortexing at full power and a 1/1000 dilution was counted using a hemacytometer. Freshly harvested conidia were inoculated into sterile PD broth, that contained 25 μg 2-deoxy-glucose/ml and 10 μg tetracycline/ml, in an Erlenmeyer flask to give a final density of 5 x 10⁶ spores/ml; the liquid volume was about 1/5 the volume of the culture flask. The broth culture was incubated at 30° C for 2 h and 20 min with shaking at 180-200 rpm. At that point the conidia appeared slightly swollen. A sterile solution of beta-glucuronidase (Sigma, St. Louis, MO) and Driselase (Sigma, St. Louis, MO) was added to give final concentrations of 248 U/ml and 610 μg/ml, respectively, and incubation was continued for 2 more hours at 30° C, 180-200 rpm. The conidia suspension was centrifuged at 6,000 to 8,000 x g for 15 min and resuspended in 100 ml of cold 1.2 M sorbitol. The sample was centrifuged two more times and washed, successively, with 35 ml and 10 ml of cold 1.2 M sorbitol. After a final centrifugation the conidia were resuspended in enough cold 1.2 M sorbitol to yield a concentration of 2.5 x 10⁹ conidia/ml (typically, 100-200 μl of cold 1.2 M sorbitol was added to the conidia pellet to achieve that concentration). Two to 4 μg of plasmid pKBE2002 (5 μl maximum volume) was mixed with 40 μl of conidia suspension (10 total conidia). The samples were transferred to 0.1 cm gap electroporation cuvettes and were kept on ice for 5 min prior to electroporation. The effects of applied voltage and pulse duration, supplied by a

Bio-Rad Gene Pulser^® II and Pulse Controller Plus (Bio-Rad, Hercules, CA), on conidia transformation were tested using multiple samples in a number of different experiments. Following electroporation, conidia were washed from the electroporation cuvettes using 1 ml of cold PD broth, 1.2 M sorbitol and were transferred to 1.5 ml microcentrifuge tubes. The conidia were pelleted by microcentrifugation for one minute at full speed and pellets were resuspended in 200 μl PD broth, 1.2 M sorbitol. One hundred μl samples were

spread onto ISP2 plates that contained 200 μg hygromycin B/ml to select for

transformants. Duplicate 100 μl samples of a 10^"5 dilution of non-electroporated conidia were spread onto ISP2 plates and resulting colony counts were used with hemacytometer counts to calculate percent germination. Duplicate 100 μl samples of a 10^"4 dilution of electroporated conidia were spread onto ISP2 plates and resulting colony counts were used with those of non-electroporated conidia to determine percentage killing. Colonies were usually visible on selection plates after three days of incubation at 30° C and new colonies appeared up to 14 days of incubation.

E. Detection of genetic transformation of Trichoderma and Aspergillus Colonies that grew on hygromycin B selection plates, following electroporation of either Trichoderma or Aspergillus, were isolated using sterile straws to remove a portion of each colony and were transferred to fresh PDA plates. After growth and formation of conidia, a sample was removed by scraping the culture with a sterile toothpick. The samples were placed in separate, sterile, 1.5 ml microcentrifuge tubes and were microwaved for 3 min at 80% power (1000 Watt microwave). Thirty μl of TE (10 mM Tris buffer, pH 8, 0.1 mM EDTA) was added to each sample and the samples were dispersed by vortexing at full power for about 30 seconds. Ninety μl of Qiagen gel extraction reagent (Qiagen, Inc., Valencia, CA) was added to each sample and total DNA was recovered following the manufacturer's gel extraction procedure to yield purified total DNA in 50 μl EB (10 mM Tris buffer, pH 8). Total DNA was also extracted from the T. reesei RUT-C30, A. niger KASNl and A. awamori ATCC 11358 host strains. One- half to 2.5 μl of each purified DNA sample was used as template in PCR reactions with hph gene-specific primers (Table 6). T. reesei actin (act) gene-specific primers and plasmid pCB1003 (1 ng), the source of the hph gene in the expression cassettes, were used separately as positive controls (Frohman, M. A. and G. R. Martin. 1990. Detection of homologous recombinants, 228-236. In M. A. Innis, D. H. Gelfand, J. J. Sninsky and T. J. White (ed.), PCR Protocols: A Guide to Methods and Applications. Academic Press,

San Diego, USA). The 25 μl PCR reactions contained, in addition to the template, lx Taq

2000^® buffer (Strategene, LaJolla, CA), 5% DMSO, 25 pmol of each primer, 10 nmol of each deoxynucleotide, and 1.25 Units of Taq 2000^® DNA polymerase. Thermal cycling conditions were: 1 min of preliminary denaturation at 94° C followed by 40 cycles of 40 s at 93°, 1 min at 58° C and 1 min 10 s at 72° C.

TABLE 6: PRIMERS USED FOR PCR SCREENING OF PUTATIVE GENETIC

TRANSFORMANTS

Optimal temperature for the primers to anneal to the template DNA in the PCR amplification.

Predicted size of the correct PCR product.

T Thhee sseeqquueennccee ooff tthhee EE.. ccoollii hhpphh ggeennee wwaass < determined by DNA sequencing of the cloned gene contained iinn ppllaassmmiidd ppCCBB11000033..

The sequence of the T. reesei act gene was obtained from GenBank accession number X75421.

F. Optimization of Electroporation Procedure for Trichoderma reesei Changes in various parameters of the electroporation procedure were examined for the effects on transformation of Trichoderma reesei RUT-C30 conidia. The applied voltage and its duration were critical for recovery of transformants, as was the physical state of the transforming DNA. Media used to grow RUT-C30 for production of conidia and the use of different osmotic stabilizers during electroporation had less effect on transformation.

Suspensions of RUT-C30 conidia mixed with the plasmid pKBE2001 (α- galactosidase expression cassette) were pulsed using field strengths that ranged from 12.5 kV/cm to 22.5 kV/cm while not changing the time constant (approx. 15 ms at 50 μF and 300 Ohms). Table 7 shows the transformation yields and percentage killing that resulted from the different applied voltages. Optimum yield of 30 transformants was obtained at a field strength of 15 kV/cm, which resulted in killing of about 65% of the conidia.

TABLE 7: EFFECT OF ELECTRICAL FIELD STRENGTH ON TRANSFORMATION OF T. reesei RUT-C30 CONIDIA

Capacitance was 50 μF and resistance was 300 Ohms using a 0.1 cm gap electroporation cassette

² Percent killing was determined from plate counts of germinated, growing conidia before and after electroporation .

³ Total number of phenotypic transformants for each experiment.

Using the field strength of 15 kV/cm, the time constant was varied by changing the capacitance and the resistance. Table 8 shows the transformation yields and percentage killing that resulted from the different conditions. Optimum yield of 52 transformants was obtained with a time constant of about 14 ms (50 μF and 300 Ohms) at a field strength of 15 kV/cm, which resulted in killing of about 60% of the conidia. The optimum electrical parameters of 15 kV/cm and 14 ms time constant were used in all subsequent experiments.

TABLE 8: EFFECT OF TIME CONSTANT ON TRANSFORMATION OF T. REESEI YUJY-CS0 CONIDIA

Field strength was 15 kV/cm in each experiment.

² Percent killing was determined from plate counts of germinated, growing conidia before and after electroporation.

³ Total number of colonies growing on selectio n plates for each expeπment.

Circular plasmid pKBE2001 and plasmid that had been linearized by digestion with Pst I and Xba I were tested for transformation efficiency. Only the linear form produced transformants; no transformants were recovered when the circular form of pKBE2001 was used in electroporation experiments with T. reesei (data not shown).

T. reesei RUT-C30 was grown on different media and the conidia that were produced were tested for their ability to be transformed by electroporation with plasmid pKBE2001 (Table 9). While growth on V8 agar produced the most conidia, those conidia were the most recalcitrant to transformation with an efficiency of only one transformant per microgram of DNA. Conversely, ISP2 agar and PDA media produced far fewer conidia but they were transformed with greater efficiency up to 12 transformants per microgram of DNA.

TABLE 9: EFFECT OF GROWTH MEDIUM ON YIELD OF CONIDIA AND TRANSFORMATION EFFICIENCY FOR T. reesei RUT-C30

Total conidia harvested from one slant culture in 5 ml of sterile water. ² Transformants per microgram of transforming DNA.

Cell wall weakening agents are frequently added to liquid growth media to aid in fungal protoplast formation from vegetative cells or to enhance electroporation of some bacteria (Chakraborty, et al., supra; Fiedler, et al., supra). We tested the effect of adding 2-deoxy-glucose to solid growth media on conidia transformation efficiency. Two-deoxy- glucose was added to PDA medium at concentrations of 10, 25, 50 and 100 μg/ml medium and, after growth, conidia were electroporated using optimum conditions. Transformation efficiencies ranged from 3 to 7 transformants per microgram DNA, but no difference was observed between PDA with or without 2-deoxy-glucose (data not shown).

Electroporated conidia were plated in top agar medium that usually contained 1.2 M sorbitol as an osmotic stabilizer. One molar mannitol and 1 M sucrose were also tested as osmotic stabilizers but transformation yield was not enhanced (data not shown). G. Optimization of Electroporation Procedure for Aspergillus Changes in the electroporation procedure and in the electrical parameters were examined for effects on transformation of Aspergillus niger KASNl conidia. A. awamori ATCC 11358 conidia were also used in electroporation experiments but the conditions were not optimized. Similar to T. reesei conidia electroporation, the applied voltage and its duration were critical for recovery of transformants, as was, additionally, enzymatic treatment of the conidia prior to electroporation.

Suspensions of partially germinated, enzyme-treated KASNl conidia mixed with 2-4 μg circular-form plasmid pKBE2002 (α-galactosidase expression cassette) were pulsed using field strengths from 10.0 kV/cm to 22.5 kV/cm. The pulse duration was determined by using a capacitance of 25 μF and resistance of either 200, 400 or 600 Ohms. Table 10 shows the yield of transformants for different electrical parameters. Optimum transformation resulted when a field strength of 12.5 kV/cm was used with 400 Ohms and 25 μF (time constant = 9.1 ms). That setting also resulted in a 61% reduction of viable conidia.

TABLE 10: EFFECTS OF ELECTRICAL FIELD STRENGTH AND PULSE DURATION ON ELECTROPORATION OF Aspergillus niger STRAIN KASNl

CONIDIA

Electrical field strength determined by instrument setting and cuvette gap.

Total number of colonies growing on selection plates for each set of conditions for one experiment. Two experiments were done for each set of conditions.

Capacitance was 25 μF for all experiments.

Not determined.

The effect on transformation efficiency of pretreatment of the conidia with β- glucuronidase and Driselase was tested using the optimum electrical parameters. After 2 hours and 20 minutes incubation of the conidia in PD broth with 2-deoxy-glucose, β- glucuronidase and Driselase were added to three separate samples to give concentrations, respectively, of 347 U/ml and 860 μg/ml, 248 U/ml and 610 μg/ml, and 184 U/ml and 460 μg/ml. Incubation continued for two more hours and the conidia were washed by centrifugation to prepare them for electroporation. The best result was obtained using 248 U β-glucuronidase/ml with 610 μg Driselase/ml (Table 11).

TABLE 11: EFFECT OF ENZYME PRETREATMENT ON TRANSFORMATION OF

Aspergillus niger CONIDIA

¹ High = 347 U β-glucuronidase/ml and 860 μg Driselase/ml; medium = 248 U β- glucuronidase/ml and 610 μg Driselase/ml; low = 184 U β-glucuronidase/ml and 460 μg Driselase/ml.

² Total number of colonies growing on selection plates.

³ Capacitance of 25 μF and 400 Ohms for each experiment.

H. Detection of genetic transformation of Trichoderma and Aspergillus Phenotypic hygromycin-resistant transformants of Trichoderma, following electroporation with linearized plasmid pKBE2001 , were tested for genetic transformation by PCR amplification of genomic DNA extracted from conidia or mycelium (Figure 6). The primers used for DNA amplification were specific for the hph gene, the agll expression construct and for the T. reesei actin gene as a control. The hph gene could be detected in about 20% of the putative transformants (Figure 6, 550 bp band) and the agll expression construct was usually detected in the ApA-positive strains (Figure 6, 476 bp band). In a few A/?λ-positive strains the agll expression construct was not detected (data not shown) which suggested that the components of the expression vector had become unlinked during recombination or that the agll expression construct had been excised following incorporation into the genome. Phenotypic hygromycin-resistant transformants of Aspergillus were also tested for genetic transformation by PCR amplification of genomic DNA extracted from conidia or mycelium (Figure 7). The primers used for hph gene detection were the same as those used to amplify Trichoderma DNA. Both of the A. awamori ATCC 11358 putative transformants tested were positive for hph (Figure 7, lanes 3 and 4, 550 bp band) as were all three hygromycin-resistant A. niger KASNl strains (Figure 7, lanes 6, 7 and 8, 550 bp band). Fig. 7 represents a gel electrophoresis analysis of PCR products amplified from genomic DNA from putative A. awamori ATCC 11358 and A. niger KASNl transgenic strains following electroporation with pKBE2002 and selection for resistance to hygromycin B. After PCR, 10 μl samples were loaded into wells of a 1% agarose gel and electrophoresed for 25 min at 20 Volts/cm in 0.5% TAE buffer. The gel was stained with 1.5 μg ethidium bromide /ml and nucleic acid bands visualized by UV transillumination. The hph gene product was 550 bp. Lane M, DNA fragment size markers; lane 1, pKBE2002; lane 2, A. awamori ATCC 11358; lane 3, A. awamori #1; lane 4, A. awamori #2; lane 5, A. niger KASNl ; lane 6, A. niger #1 ; lane 7, A. niger #2; lane 8, A. niger #3.

I. Discussion Described in this specification are genetic transformation systems for Trichoderma and Aspergillus based on electroporation of conidia. The transforming DNA contained a recombinant gene expression cassette designed to over-express an α- galactosidase enzyme, which has potential commercial value. A gene that encodes resistance to the antibiotic hygromycin B was included in the expression cassette as a dominant selectable marker.

Optimum conditions for electrotransformation of Trichoderma conidia were an electrical field of 15 kV/cm with 15 ms time constant set by 50 μF capacitance and 300 Ohms resistance. In contrast, conditions for optimum transformation of T. longibrachiatum and T. harzianum protoplasts are 2.8 kV/cm, 25 μF and 800 Ohms (Goldman, G. H., M. van Montagu and A. Herrera-Estrella. 1990. Transformation of Trichoderma harzianum by high-voltage electric pulse. Curr. Genet. 17: 169-174; Sanchez-Torres, P., R. Gonzalez, J. A. Perez-Gonzalez, L. Gonzalez-Candelas and D. Ramon. 1994. Development of a transformation system for Trichoderma longibrachiatum and its use for constructing multicopy transformants for the egll gene. Appl. Microbiol. Biotechnol. 41 : 440-446). In our experiments, the outer protein coatings of conidia clearly presented a greater barrier to DNA entry than the lipid bilayer membrane of protoplasts, therefore, the added 'push' of increased electrical field was necessary. The yield of transformants from electroporation of conidia, however, was similar to the 0.5 to 5 transformants/μg DNA obtained by protoplast electroporation. Subsequent experiments in our lab (data not shown) have produced transformation efficiencies as high as 150 transformants/μg DNA, which is on the same order of magnitude as polyethylene glycol- mediated Trichoderma protoplast transformation (Gruber, F., J. Visser, C. P. Kubicek and L. H. de Graaff. 1990. The development of a heterologous transformation system for the cellulolytic fungus Trichoderma reesei based in a/> rG-negative mutant strain. Curr. Genet. 18: 71-76; Penttila, et al., supra). Electroporation of conidia, rather than protoplasts, is advantageous since preparation of the conidia is relatively simple compared to the complexity and attendant viability losses of protoplast preparation. Transformation of A. niger and A. awamori conidia was successful only after partial germination and treatment with enzymes to weaken the cell wall. We used treatment with 248 U β-glucuronidase/ml and 610 μg Driselase/ml (which contains laminarinase, xylanase and cellulase) followed by electroporation with 12.5 kV/cm, 25 μF and 400 Ohms to obtain an efficiency of 5-6 transformants/μg DNA. Our procedure was adapted from those used to electrotransform several different fungal genera which had yielded transformation efficiencies up to 21 transformants/μg DNA when swollen

conidia were treated with 1 mg β-glucuronidase/ml (about 100-1000 U β- glucuronidase/ml) prior to electroporation (Chakraborty, et al., supra; Sanchez, et al., supra). Our optimal electrical parameters were similar to those used with germinated conidia from other fungi except for A. nidulans which required a much lower electrical field of 5 kV/cm (Sanchez, et al., supra). Polyethylene glycol-mediated transformation and electrotransformation of A. niger protoplasts yield similar transformation efficiencies (Ward, M., K. H. Kodama and L. J. Wilson. 1989. Transformation of Aspergillus awamori and niger by electroporation. Exp. Mycol. 13: 289-293). However, electroporation of Aspergillus conidia, like Trichoderma conidia, was simpler than protoplast transformation and, therefore, advantageous.

Mitotic stability of genetically transformed filamentous fungi varies from species to species. While 100% of 7. harzianum transformants retain their transgenes through multiple generations without selection pressure, only 20% to 80% of T. longibrachiatum and T. reesei transformants are stably transformed (Gruber, et al., supra; Sanchez-Torres, et al., supra). Similar to those previous results, we found that 20% of phenotypic T. reesei transformants contained the hph selectable marker gene after growth without hygromycin B, when analyzed by PCR amplification of genomic DNA. Aspergillus transgenic strains, on the other hand, are more stable after multiple generations of growth without selection pressure (Sanchez-Torres, et al., supra). In accord with that observation, although our sample size is small, all five of our Aspergillus phenotypic transformants contained the hph gene, as shown by PCR of genomic DNA.

The genetic transformation procedures and the gene expression cassette construction techniques described in this application, are the cornerstones of recombinant DNA technology that will yield transgenic microorganisms that over-express enzymes and proteins for use in commercial products. Efficient genetic transformation procedures are necessary to provide the large numbers of transgenic microorganisms that, after screening for enzyme/protein overproduction, will yield potential production strains.

Experiment 3 - Fermentative Production A. Seed Cultures

The strain of Trichoderma reesei, Strain KBT2147, genetically transformed to over-produce as described above was used as a seed strain in a series of experiments directed to using fermentation to produce large quantities of α-galactosidase. The media used for producing fungal seed cultures is listed in the following Table 12. The ingredients are presented as the maximum quantity per liter.

TABLE 12 - SOLUBLE SEED MEDIA

Cultures grown in fermentors may require the addition of up to 0.01% of an antifoam agent. Ammonium hydroxide, ammonia gas, O-phosphoric acid, sodium hydroxide, CO₂, oxygen and/or compressed air may be added as required to maintain desired pH and dissolved oxygen levels. The media used for producing fungal seed stock cultures is listed in the following Table 13. The ingredients are presented as the maximum quantity (g/L).

TABLE 13 - V8 AGAR MEDIA

The media used for the production of α-galactosidase is listed in the following Table 14. The ingredients presented as the maximum quantity per liter.

TABLE 14 - GROWTH MEDIA

C. Methods of Growing Cultures, Inoculation and Harvest The cultures are grown in containers of various sizes, either flasks, bottles, or carboys, as appropriate for the volume of culture being grown. The master seed and production seed stocks are stored at -20°C or colder.

Between about 0.01% and about 10%, and preferably between about 0.05% and about 0.5%, inoculum by volume is used to inoculate cultures. Either frozen or fresh cultures may be pooled and are inoculated at the appropriate levels. Once inoculated, the cultures are incubated for between 18 and 144 hours at 29°C±2°C. Fungal growth is demonstrated by the amount of biomass (i.e. wet weight). If fungal growth is unsatisfactory, the culture is eliminated. Prior to harvest, the cultures are checked macroscopically for abnormalities or signs of contamination. To be acceptable for harvest, the cultures must have an acceptable level of bacterial and/or fungal contamination. The minimum time from inoculation to harvest is about three days and the maximum time is about ten days. Culture fluid may be harvested into sanitized containers.

D. Preparation of the Product

The fungal culture fluids are concentrated up to 50X using either centrifugation, filtration or, preferably, ultrafiltration, or combinations of such techniques. Following concentration, potassium sorbate may be added as a preservative or stabilizer at a concentration not to exceed 0.1%. The concentrate may be stored as a liquid and is stored at between about 2°C and about 7°C, or may be preferably spray dried at about 130°C, whereupon the shelf life at usual storage temperatures exceeds two months and is expected to be up to a year or more without substantial loss in efficacy. The T. reesei spray-dried product may be standardized to achieve the desired potency by dilution with calcium carbonate (CaCO₃). By way of example, if the end product is 2500 Kg, with a potency of 12,000 IU/mL, and it is desired to produce a product having a potency of 10,000 IU/mL, 500 Kg of calcium carbonate are added.

E. Sample Fermentation Run

Two Working Seeds were prepared from the Master Seed. The Master Seed (X) was serially propagated on V8 agar slants. The X+2 passage was harvested and used to establish the Working Seed (X+2), which was subsequently serially propagated on V8 agar slants. The X+4 passage was harvested and used to establish the (X+4) Working Seed. The Master Seeds and Working Seeds were used as the frozen seed stocks. In general, a 2,500 ml Erlenmeyer flask containing 1,000 ml of NN seed media was inoculated with 1.0 ml of the harvested culture per flask.

A general description of the 100 L fermentation process is as follows. The culture was incubated at 29°C±2°C, 200rpm for 40-52 hours in an incubator/shaker. A volume of the seed cultures (1,000 ml) was inoculated into 100,000 ml of α-galactosidase fermentation media to achieve a one percent level of inoculum. The culture was incubated for approximately 144 hours in an ABEC 150 L fermentor. The fermentors were fed with a 20% lactose solution. The feed was added at a rate of 0.2 gm/L/hr. Approximately 50 ml of Antifoam was added during the fermentation. Four baffles and two Rhuston impellers were used. The rpm of the vessel was adjusted to maintain a dissolved oxygen concentration of 10%. The maximum speed used was 150 rpm. The temperature of the fermentation was held at 29°C for 40-48 hours and then dropped to

27°C for the duration of the fermentation. The pH of the culture was allowed to drop naturally until the pH leveled off. Then, the pH of the fermentation was controlled with

25% ammonium hydroxide at a pH of 4.0 for the duration of the fermentation.

A general description of the 5,000 L fermentation process is as follows. A volume of the seed cultures (1,000 ml) was inoculated into 100,000 ml of α-galactosidase fermentation media to achieve a one percent level of inoculum. The culture was incubated for approximately 24 hours in ABEC 150 L fermentors. A volume of the seed cultures (50 L) was inoculated into 5,000 L of α-galactosidase fermentation media to achieve a one percent level of inoculum. The culture was incubated for approximately 144 hours in an ABEC 7,500 L fermentor. Thereafter, the process was the same as described above for the 100 L process.

The protein determination was assayed following the Modified Lowry protein assay (Sigma). The α-galactosidase activity was assayed at 37°C using PNPG as a substrate according to a plate reader modification of the α-galactosidase assay taken from pages 794-95 of the Food Chemical Codex, 4^th Edition (1996). Samples from the fermentors were taken throughout the fermentation and assayed for wet weight. The 1.0 ml samples were centrifuged at 13,000 rpm in a micro-centrifuge tube for 5 minutes. The supernatant was decanted and the centrifuge tubes were inverted to dry. Any remaining supernatant was blotted from the tubes prior to weighing. The harvested fermentation culture was cooled to 20°C at 144 hours post inoculation. The culture was clarified by centrifugation through an Alpha-Laval continuous flow centrifuge at approximately 8 L/min. The supernatant was then concentrated 10-20 times by hollow fiber filtration (10,000 molecular weight cut off from AG Technology). The concentrate was harvested, the system was rinsed with water, and the rinse was collected and added to the concentrate. The concentrate was centrifuged again to remove any contaminants or residual biomass. Potassium sorbate was added to the concentrate for a final formulation of 0.1% for both the liquid and dry forms. Sorbitol was added to the liquid formulation at 10%. The final products were formulated to contain not less than 1,000 α-galactosidase IU per gram. A flowchart of a preferred embodiment of the process for producing commercial quantities of α-galactosidase from the transformed T. reesei organism is illustrated in Fig. 8.

Fermentation experiments were conducted to determine the preferred parameters for the production of α-galactosidase. The parameters were adjusted throughout the fermentations. Table 15 lists the results for the fermentation and down-stream processing fluids of three 100 L and one 5,000 L fermentations. Table 16 provides analytical data on the final liquid formulated product from these fermentations.

Batch 1 (100 L): Contamination was detected in the fermentor after 120 hours post-inoculation. The viable count was 3x10 org./ml. The contamination levels during down stream processing continued to decline until the final blend was completed. The maximum airflow added to the vessel was 45 L/min. The average percent dissolved oxygen content for the first 72 hours was approximately 15%. The culture did not present the typical pH profile during the first 48 hours of growth in that there was no secondary increase in pH. The automated pH control brought the culture to pH of 4.0 and then failed. Thereafter, the pH was manually adjusted up to 4.0 four times from 48 to 120 hours post inoculation and continued to slowly drop between cycles.

Batch 2 (100 L): No contamination was detectable in the fermentor. At one stage during the down stream processing, the viable contamination count reached 7.5x10 org./ml. The maximum airflow added to the vessel was 70 L/min. The average percent dissolved oxygen content for the first 72 hours was approximately 15%. The fermentor control of the rpm was improperly set and the vessel reached 150 rpm for approximately

12 hours.

Batch 3 (100 L): No contamination was detectable in the fermentor. The maximum airflow added to the vessel was 73 L/min. The average percent dissolved oxygen content for the first 72 hours was approximately 50%. The automatic pH control failed after the culture was brought up to a pH of 4.0. The pH of the culture was at 2.8 for approximately 30 hours before it was adjusted up to 4.0. Thereafter, the pH controller maintained pH at 3.75 for the duration of the run. This fermentation also used 150 rpm to control the dissolved oxygen.

Batch 4 (5,000 L): No contamination was detectable in the fermentor. The maximum airflow added to the vessel was 3,000 L/min. The average percent dissolved oxygen content for the first 72 hours was approximately 100%.

TABLE 15 - RESULTS OF FERMENTATION AND DOWN STREAM PROCESSING FLUIDS

TABLE 16 - ANALYSIS OF FORMULATED PRODUCT

Although the invention has been described in considerable detail through the preceding examples, such detail is for the purpose of illustration. Many variations and modifications can be made by one skilled in the art without departing from the spirit and scope of the invention as described in the appended claims.

Claims

We claim:

1. An isolated filamentous fungus that expresses at least 5 times more α- galactosidase than the wild type strain Trichoderma reesei RUT-C30.

2. An isolated filamentous fungus that expresses at least 20 IU of α- galactosidase per ml of supernatant when the fungus is grown in a production fermentation.

3. An isolated fungus according to claim 1, further comprising multiple copies of an integrated Agll gene from Trichoderma reesei.

4. An isolated fungus according to claim 3, wherein the Agll gene is operably linked to at least one regulatory sequence which directs over-expression of the Agll gene.

5. A method for over-producing α-galactosidase, comprising the step of cultivating a filamentous fungus according to claim 1 in a suitable growth medium under conditions where the fungus produces α-galactosidase.

6. The method of claim 5, wherein said growth medium includes soy bean hulls.

7. The method of claim 5, further comprising the step of isolating the α- galactosidase in the form of a concentrate.

8. The method of claim 6, wherein said concentrate is dried.

9. An expression cassette for transforming a filamentous fungus, comprising a Trichoderma cellobiohydrolase 1 gene promoter and terminator operably linked to a Trichoderma α-galactosidase 1 gene.

10. An expression cassette as defined in claim 9, wherein overlapping PCR primers having the nucleic acid SEQ ID NO. 2 and 3 comprise the junction between the cellobiohydrolase 1 gene promoter and the α-galactosidase 1 gene.

11. An expression cassette as defined in claim 9, wherein overlapping PCR primers having the nucleic acid SEQ ID NO. 4 and 5 comprise the junction between the cellobiohydrolase 1 gene terminator and the α-galactosidase 1 gene.

12. The fungus according to claim 1, wherein said filamentous fungus is a strain of Trichoderma reesei.

13. The fungus according to claim 2, wherein said filamentous fungus is a strain of Trichoderma reesei.

14. A transformed fungal cell comprising an expression cassette according to claim 9.

15. A process for transforming a filamentous host cell such that said host cell is capable of expressing a level of α-galactosidase that is at least 5 times higher than that expressed by wild type Trichoderma, comprising the steps of:

(a) treating a host cell with an expression cassette according to claim 9 under conditions such that said α-galactosidase 1 gene with said operably linked cellobiohydrolase 1 gene promoter and terminator integrates into the genome of said cell and transformed cells are effectuated; and

(b) isolating said transformed cells which over-express and secrete said α- galactosidase.

16. The process according to claim 15, wherein said host cell is Trichoderma reesei.

17. A method for producing α-galactosidase, comprising the steps of growing

the transformed cell of claim 15 under conditions suitable for production of said α-

galactosidase and isolating said α-galactosidase.