IE64478B1 - Process and signal sequence for the extracellular production of proteinaceous material - Google Patents

Description

64478 2 The present invention relates to a process for secreting a proteinaceous material extracellularly and to a signal sequence useful in this process.

The following publications are referred to by corresponding 5 number in this application: 1. Lineback, et a!.. Cereal Chemistry, 49:283 (1972). la. Svensson, et al.t Carlsberg Res. Comtnun., 47:55 (1982). lb. Svensson, et al., Abstract IV-27, Xlth International Carbohydrate Symposium, Vancouver, British Columbia, Aug., 1982. 10 lc. Botstein, et al., in The Molecular Biology of the Yeast Saccharo-myces-Metabolism and Gene Expression, ed. by Strathern, et al. (Mew York: Cold Spring Harbor Laboratory* 1982), p.607ff. Id. Struhl, Nature, 305:391 (1983). le. European Pat. Application 81303155.6 (Publication 45573 dated 15 February 10, 1982) to Stanford University. 2. Chirgwin, et al., Biochem., 18:5294 (1979). 3. Sehgal, Methods in Enzymology, 79:111 (1981), at p. 117. 4. Pelham, et al., Eur. J. Biochem., 67:247 (1976). 5. Maniatis, et al., Molecular Cloning: A Laboratory Manual, publ., 20 Cold Spring Harbor, N.Y. (1982), pp. 344-349. 6. Ivarie, et al.. Anal. Biochem., 97:24 (1979). 7. Chang, et al., Nature, 275:617 (1978). 8. Doel, et al.. Nucleic Acids Res., 4:3701 (1977). 9. Southern. J. MoT. Biol., 98:503 (1975). 25 10. Sanger, et al., Proc. Nat. Acad. Sci. USA, 74:5463 (1977). 11. Messing, et al., Nucleic Acid Res., 9:309 (1981). 12. Maxam, et al., Proc. Nat. Acad. Sci. USA, 74:560 (1977). 13. Mount. Nucl. Acids Res., 10:459 (1982). 14. Langford, et al.. Proc. Natl. Acad. Sci. USA. 80:1496 (1983). 30 15. Langford, et al., Cell, 33:519 (1983). 16. Holland, et al., J. Biol. Chem., 256:1385 (1981). 16a. Sutcliffe , Cold Spring Harbor Symposium on Quantitative Biology, 43: 77 (1978). 16b. Broach, et al., Gene, 8: 121 (1979). 16c. Beach, et al., Nature, 290: 140 (1981). 17a. Erlich, et al., J. Biol. Chem., 254:12,240 (1979). 17b. Erlich, et al., Inf. and Imm., 41:683 (1983). 18. Dewald, et al., in.Methods in Enzymology, Vol. XXXII, Biomem-branes, Part B, ed. by Fleischer et al. (New York: Academic Press, 1974), p.87-88.

The techniques of genetic engineering have been successfully applied to the pharmaceutical industry, resulting in a number of novel products. Increasingly, it has become apparent that the same technologies can be applied on a larger scale to the production of enzymes of value to other industries. The benefits of achieving commercially useful processes through genetic engineering are expected to include: (1) cost savings in enzyme production, (2) production of enzymes in organisms generally recognized as safe which are more suitable for food products, and (3) specific genetic modifications at the DNA level to improve enzyme properties such as thermal stability and other performance characteristics. 0/r6 important Industrial application of genetic engineering is the production of proteinaceous material using transformed microorganisms. In many or even in most cases this proteinaceous material is intracellular^ accumulated in the host. Consequently, the biotechnological production of such a proteinaceous material usually involves special disrupting and purification steps. These, however, are time-consuming and expensive. Therefore, there exists a need for genetic engineering processes for the extracellular production of such a proteinaceous material. 4 10 On the other hand, there is great interest in Improving the ability of industrial yeast strains to degrade complex carbohydrate substrates such as starch. Yeasts such as Saccharomyces cerevisiae which are suitable for alcoholic fermentation do not produce an enzyme capable of hydrolyzing starch to utilizable substrates. Currently, starch used as a food source in alcoholic fermentation must be saccharified, either chemically or enzymatically, In a separate process to produce utilizable substrates for the fermenting yeast.

It would thus be.desirable to construct, by genetic recombination methods, a fermentation yeast such as _S. cerevisiae which itself has the capacity to extracellularly synthesize one or more enzymes capable of breaking down starch to utilizable substrates. European Pat. Appln. 15 0,034,470 discloses preparing recombinant DNA containing an amylase encoding gene by cleaving a bacterial donor microorganism to obtain DNA and inserting those fragments in a vector. The amylase enzymes produced from the DMA which are used to hydrolyze starch are preferably alpha-amylase, beta-amylase or a pullulanase. 20 Accordingly, the invention provides a process for secreting a proteinaceous material extracellularly vrtiich comprises growing a host organism which host is transformed by a DNA expression vector comprising a promoter fragment viiich functions in the host organism, a signal sequence having substantially the following amino acid sequence: 25 met ser fhe arg ser leu leu ala leu ser gly leu val cys thr gly leu ALA asn val he ser lys ARG and a DMA segment Which codes for the proteinaceous material, said proteinaceous material being characterized as a cell protein vjhich is synthesized in its normal host organism with a hydrophobic signal 30 sequence Which is normally cleaved during secretion from its normal host organism.

Preferably, the proteinaceous material is a protein which is normally secreted and most preferably it is glucoamylase. The vector may or may not contain a DNA segment which functions as an origin of replication, a selectable marker or a transcription terminator segment.

In a preferred embodiment, the host organism used is a yeast such as a species of the genus Saccharomyces.

Accordingly, in one aspect, the present invention is also concerned with constructing a fermentation host organism which contains, in recombinant form, a gene coding for a glucoamylase which is active in hydrolyzing starch at both alpha 1-4 and alpha 1-6 linkages to generate glucose.

The present invention generally concerns the construction of a glucoamylase gene which can be introduced in recombinant form into a foreign host including but not limited to yeast or bacteria. Such host may also include virus, plant or animal cells.

According to one aspect of the invention, there is provided a modified DNA sequence coding for fungal glucoamylase protein or its single or multiple base substitutions, deletions, insertions or inversions, wherein said DNA sequence is derived from natural, synthetic or semi-synthetic sources and is capable, when correctly combined with a cleaved expression vector, of expressing a non-native protein having glucoamylase enzyme activity upon transformation by the vector of a host organism. Most preferably the expression vector is the plasmid pACl described further hereinbelow which has been cleaved at its HindiII site so that the sequence can be Inserted at that site.

According to another aspect of the invention, it has been discovered that Aspergillus awairori cells, when grown under conditions which induce glucoamylase, contain a relatively high concentration of ap- 6 proximately 2.2 kilobase poly A RNA which is not detected in cells grown under noninducing conditions. The induced poly A RNA (mRNA) is capable of directing the synthesis, in a cell-free protein synthesizing system, of an unglycosylated polypeptide which has a molecular 5 weight of between about 70,000 and 74,000 daltons. The polypeptide « produced is immunologically reactive with antibodies prepared against A. awamori glucoamylase.

A radioactively labeled cONA copy of the induced poly A RNA is produced which is used in hybridization studies to identify A. 10 awamori genomic DNA fragments containing portions of the glucoamylase gene. The hybridization studies suggest that A. awamori contains a single glucoamylase gene.

Similarly, the cDNA is used to identify phage or plasmid vectors containing such genomic DNA fragments in recombinant form. 15 The identified cloning vectors may be used in determining gene polynucleotide sequences and sequence homology with the cDNA.

When a HindiII fragment containing the A. awamori gluco-amylase gene is Inserted intofyeast, neither transcription nor translation in these heterologous hosts is detected. 20 The invention also provides for recombinant DNA expression vectors containing the DNA sequence. The vector is preferably one which is compatible with a selected foreign host organism, and permits expression of the gene in the host. The exogenous gene which is expressed may be genomic DNA, synthetic DNA or a cDNA obtained from 25 a mRNA by use of reverse transcriptase.

A novel method for producing a glucoamylase gene containing the appropriate DNA sequence generally includes producing genomic digest fragments, providing a glucoamylase probe, using the-probe to Identify genomic digest fragments containing glucoamylase gene 30 regions, molecularly cloning the identified genomic digest fragments, molecularly cloning partial cDNA, sequencing the genomic and cDNA Ci; clones, comparing the sequenced glucoamylase gene regions with all or a portion of the amino acid sequence of the mature glucoainylase enzyme to determine the existence and location of all the Introns and exons 7 in the genomic clones, and constructing a gene whose codon sequence is substantially identical to that of the genomic glucoamylase gene when the sequences comprising the introns are deleted.

In a preferred embodiment of the method, the glucoamylase 5 probe is provided by selecting a fungal source capable of producing a level of glucoamylase, when grown on starch, which is at least about ten times that produced by the fungal species when grown on xylose or glycerol in the absence of starch, culturing cells of the selected fungus under conditions which induce secretion of glucoamylase into 10 the culture medium, obtaining mRNA from the cultured cells, fractionating the mRNA obtained according to size, selecting an mRNA which is detectable as having a relatively high concentration with respect to the equivalent-sized mRNA produced by cells of the selected fungal species cultured under conditions which do not induce secretion of 15 glucoamylase into the culture medium, and copying the selected mRNA to produce the glucoamylase probe.

In yet another embodiment of the invention is.provided a host organism transformed with a DNA expression vector comprising a promoter fragment that functions in that host and a DNA segment having 20 a modified DNA sequence coding for fungal glucoamylase protein, the DNA segment being in an orientation with the promoter fragment such that in the host it is expressed to produce a non-native glucoamylase protein.

The gene herein, when expressed in a host organism trans-25 formed by an expression vector comprising the gene, produces an enzyme having glucoamylase activity. Preferably the glucoamylase enzyme is produced as a preprotein with a signal sequence at its N^-terminus which 1s processed by the host organism during secretion.

In another embodiment, the Invention relates to a process 30 for producing glucose by saccharlfication of starch using a recombinant glucoamylase gene.

In another embodiment, the invention relates to a process for producing ethanol by simultaneous saccharification and fermentation which comprises growing, on a nonfermentable carbon source 8 which is a substrate for glucoamylase enzyme, a host organism transformed by the DNA expression vector described above. The carbon source is preferably starch, soluble starch, maltose or isomaltose.

In the invention herein, the glucoamylase enzyme obtained 5 when the heterologous gene is expressed in^yeast is found to be glycosylated. In addition, a significant portion (e.g., greater than 90%) of the glucoamylase is secreted in the media. Also, when the N-terminus of the non-native glucoamylase protein secreted in the media (having a purity of greater than 85%) was sequenced, the first 29 10 amino acids were found to be identical to the mature glucoamylase protein secreted by Aspergillus. . The apparent molecular weight as determined by SDS polyaery1 amide gel electrophoresis of the glucoamylase protein obtained herein is similar to that observed for the mature processed and glycosylated form of the native glucoamylase 15 secreted by.Aspergillus. Further, the carboxy terminal amino acid is identical .to that of the large molecular weight form of glucoamylase produced by Aspergillus.

In the accompanying drawings: fig. 1 represents gel electrophoretic patterns showing in vitro translation of b. awamori mRNA from cells grown in medium containing xylose or starch as carbon source. Translation products were immunoprecipitated using rabbit anti-glucoamylase antibody (lane 1, xylose-grown cells; lane 3, starch-grown cells) or normal rabbit antibody (lane 2, xylose-grown cells; lane 4, starch-grown cells); figs. 2A and 2B represent gel electrophoretic patterns identifying glucoamylase mRNA. In FIG. 2A, poly A-containing mRNA from cells grown in medium containing starch (lane 1) or xylose (lane 2) was analyzed by MeHgOH-agarose gel electrophoresis. Human and _E. coli ribosomal RNAs provide molecular weight markers. The A_. awamori ribosomal RNAs are indicated as '28S' and 118S1. The major 'induced' mRNA (arrow) was isolated from the gel and used to direct in vitro translation. In FIG. 2B, total translation products of reactions containing no exogenous mRNA (lane 1) or the isolated major 'induced' mRNA (lane 2) are shown. Immunoprecipitation of protein products in lane 2, using rabbit anti-glucoamylase antibody, is shown in lane 3 ; fig. 3 shows a restriction endonuclease map of jv. awamori genome surrounding the glucoamylase gene*. The entire structural gene is contained within the 3.4 kilobase EcoRI fragment isolated from the Charon 4A library. The protein-encoding regions of the glucoamylase gene are indicated as solid boxes and the arrow Indicates the direction and extent of transcription; fig. 4 shows gel electrophoretic patterns where pGARl is used to hybridize to, and select, glucoamylase mRNA. Total A_. awamori mRNA (lane 1) and mRNA isolated by virtue of hybridization to pGARl DNA (lane 2) was translated in vitro and the protein products are displayed. Protein products of lane 2 are immunoprecipitated using rabbit anti-glucoamylase antibody (lane 3) or normal rabbit antibody (lane 4); 1 o FIG. 5 illustrates primer extension to determine 5' termini of glucoamylase mRNA and the sequence which was determined. The products of primer extension at 42°C (lane 1) and 50°C (lane 2) are displayed on a sequencing gel in parallel with ml3/dideoxynucleq-5 tide sequencing reactions of this region, utilizing the identical 15-mer primer. The sequence presented represents the glucoamylase mRNA sequence and is complementary to that read from the sequencing reactions shown; fig. 6 illustrates a restriction map of the EcoRI fragment 10 containing the genomic glucoamylase gene, where the shaded boxes under the sequence represent the exons or coding regions of the glucoamylase gene and the arrow represents the direction of mRNA transcription; fig. 7 illustrates a plasmid map for pGAC9; fig. 8 illustrates a plasmid map for pGC2l; 15 fig. g illustrates plate assays for degradation of Baker's starch by various transformed yeast strains. The strains given below were streaked on minimal media containing histidine at 40 mg/1 and 2% w/v Baker's starch. After 12 days incubation at 30°C the plates were stained with iodine vapors. The starch was stained purple, and the 20 clear zones represent regions in which the starch has been hydrolyzed.

Area Plate of Plate Yeast Plasmid 1 a C468 pACl b C468 pGAC9 c C468 pGC21 d C468 pGC21 2 , a C468 pACl b C468 pGAC9 c C468 pGAC9 d C468 pGAC9 3 a H18 pACl b H18 pGAC9 c C303 d H18 pGAC9 *C303 strain is S. diastaticus. 11 fig. 10 shows DEAE-Sepharose chromatography of glucoamylase produced by the recombinant yeast in a 10-liter fermentor ; and fig. 11 shows gel electrophoretic patterns of: BioRad High Molecular Weight Protein Standards (lane 1), 25 ng awamori 5 glucoamylase-I (lane 2 and 5)f 25 p.g A. awamori glucoamylase-II (lane 3 and 6), and 25 |ig recombinant glucoamylase (lane 4 and 7). Lanes 1- 4 were stained with Coomassie Blue stain and lanes 5-7 with Periodic Acid Schiff's stain.

The following terms used in the description are defined 10 below: "DNA sequence" refers to a linear array of nucleotides connected one to the other by phosphodiester bonds between the 3' and 5* carbons of adjacent pentoses.

"Modified DNA sequence" refers to a DNA sequence which is 15 altered from the native glucoamylase DNA sequence such as by removing the introns from or modifying the introns of the native sequence. The examples illustrate sequences which are free of introns. Sequences substantially free of Introns means greater than about 80% free.

"Glucoamylase enzyme activity" refers to the amount by which 20 the enzyme in contact with an aqueous slurry of starch or starch hydrolysate degrades starch to glucose molecules.

"Single or multiple base substitutions and deletions, insertions and Inversions" of the basic modified DNA sequence refer to degeneracy in the DNA sequence where the codons may be mutated or the 25 deoxyribonucleotldes may be derlvatlzed to contain different bases or other elements, but the DNA sequence thus altered is still capable, on transformation In a host, of expressing glucoamylase protein.

"Fungal glucoamylase protein" refers to protein which Is not derived from a bacterial source, but rather from a fungal source such 30 as a strain from the genus Aspergillus. Thus, a modified DNA sequence 1 2 coding for fungal glucoamylase protein signifies that the DNA is not derived from a bacterial donor microorganism.

"Non-native glucoamylase protein" refers to glucoamylase protein not produced naturally or natively by the microorganism used 5 as the host.

"Nonfermentable carbon source which is a substrate for glucoamylase" refers to substrates for the glucoamylase enzyme which the host cannot ferment, such as starch, maltose, isomaltose and other starch derived oligosaccharides. Cellulose is not a substrate for 10 glucoamylase and thus is not contemplated in this definition.

The present invention relates to a modified DNA sequence and an expression vector into which the gene has been introduced by recombinant DNA techniques, which, when transformed in a host organism, expresses glucoamylase. The modified DNA sequence may be derived from 15 a natural, synthetic or semi"synthetic source. Preferably it is derived from a selected native fungal source which produces an induced level of glucoamylase which is at least about ten times its uninduced level. The induced level is that which is produced by the fungal species when grown on starch as a sole or primary carbon source, and 20 the uninduced level, that observed when the fungal species is grown on glycerol or xylose.

. The selected fungus for producing glucoamylase is suitably cultured under glucoainylase-induction conditions and a poly A RNA fraction from the cultured cells is isolated and size fractionated to 25 reveal a glucoamylase mRNA present in a detectably higher concentration than in mRNA from uninduced cells. A glucoamylase cDNA is produced by copying the mRNA, using a reverse transcriptase.

A preferred DNA sequence contemplated in the present Invention is the sequence coding for the fungal glucoamylase (amylo-30 glucosidase) from filamentous fungi, preferably a species of the class Ascomycetes, preferably the filamentous Ascomycetes, more preferably from an Aspergillus species, and most preferably Aspergillus awamori. The native enzyme obtained from these sources Is active In breaking down high molecular weight starch, and is able to hydrolyze 13 alpha 1-6 branch linkages as well as alpha 1-4 chain linkages. Relatively high levels of the enzyme are produced and secreted inj\. awamori cultures grown on starch and a variety of 6-carbon sugars, such as glucose. 5 Although the invention will be described with particular reference to A. awamori as a source of the DNA sequence, it is recognized that the invention applies to other fungal species which have an inducible glucoamylase, preferably species of the Aspergillus genus. In particular, A. awamori glucoamylase appears to be similar, if not 10 identical, to Aspergillus niger glucoamylase, as will be seen below.

The fungal species A. awamori was selected for detailed study. This fungal species, when grown on starch as a sole or primary carbon source, produces an amount of glucoamylase in the culture medium, based on measurable enzyme activity per cell dry weight, which 15 is about 200 times that of cells grown on xylose or glycerol.

A. awamori, when grown on starch, produces and secretes at least two physically distinguishable glucoamylase enzymes. One of these enzymes, referred to as glucoamylase-I, has a molecular weight of about 74,900 daltons, as reported in Reference 1, and is glycosyl-20 ated at some or all of the peptide serine and threonine residues. A second enzyme, glucoamylase-II, has a molecular weight of about 54,300 daltons, as reported in Reference 1, and is also glycosylated. It is noted that the sizes of the glycosylated glucoamylase protein given herein are only approximate, because glycoproteins are difficult to 25 characterize precisely.

Several 11nes of evi dence suggest that the two A. awamori glucoamylase enzymes are derived from a common polypeptide. Antibodies prepared against each enzyme form react immunospecif/ically with the other form, as will be seen below. The two enzymes have Identical V 30 amino acid sequences in N-terminal fragments containing about 30 amino acids each. Further, these N-terminal sequences are Identical to those In glucoamylase I and II forms from Aspergillus niger, and the two a. niger glucoamylase forms appear to be derived from a common polypeptide, as reported in Reference la. Experiments performed in support of the present application, discussed below, indicate that a single A. awamori glucoamylase gene codes for a single glucoamylase polypeptide precursor, which is very similar, if not identical, to that produced by A. niger. 5 According to one aspect of the invention, it has been dis covered that cells of a selected fungal species, when grown under conditions which induce the secretion of glucoamylase into the culture medium, contain poly A RNA which is essentially undetectable in cells grown under noninducing conditions. The poly A RNA is capable of 10 directing the synthesis, in a cell-free protein synthesizing system, of a polypeptide which is immunologically reactive With antibodies prepared against the glucoamylase from that fungal species.

Because the gene is not expressed in yeast hosts with its intact regulatory elements, it is necessary to delete or modify the 15 introns and to exchange promoters .so that the yeast will transcribe the gene, translate the mRNA, and produce an active glucoamylase.

The introns may be removed from the glucoamylase gene either by methods known in the literature for removing introns or by the simpler method described in section B of Example 2 below using 20 specific restriction enzymes in various steps to create fragments which are then 11gated together and using site-directed mutagenesis. In the mutagenesis technique the 5'-most intron of the glucoamylase gene is removed using a primer which is homologous to sequences on both sides of the intron and annealing this primer to a s.ingle-25 stranded DNA template of the glucoamylase genomic clone. The primer is then used to prime DNA synthesis of the complementary strand by extension of the primer on an M13 single-stranded phage DNA template. The resulting molecules were double-stranded circular molecules with single-stranded loops containing the intron sequence. When -30 the molecules are transformed into cells, these loops may be excised, thereby removing the intron, but even without excision DNA replication will generate the correct progeny. If the Introns are present In the gene, little or no glucoamylase enzyme 1s produced 1n a yeast 1n which the gene is expressed.

After the introns have been removed therefrom,-the glucoamylase gene may be inserted by genetic recombination into a DNA expression vector, preferably a plasmid, which may then be used to transform a host organism„ Suitable ,organisms for this purpose include bacteria such as _E. coli, viruses and yeasts. The host organism useful in this present invention must contain the appropriate genetic background for transformation thereof, i.e., the expression vector is compatible with the genetic background of the host strain. For example, the host recipient yeast strains C468 and H18, which are haploid S. cerevisiae laboratory strains employed in the following examples illustrating yeast hosts, are deficient in p-isopropylmalate dehydrogenase activity and therefore are complemented to leucine prototrophy by inserting into the expression vector the selectable marker p-isopropylmalate dehydrogenase (LEU 2). While the expression vector may by itself be capable of phenotypic selection by containing a selectable marker, it need not be so capable because the host can be screened or selected for the glucoamylase gene.

The preferred bacterial host herein is Z. coli. The preferred yeast host strain herein is from a species of the genus Saccharomyces, preferably cerevisiae, j>. uvarum, j>. carlsbergensis, or mixtures or mutants thereof, more preferably a S_. cerevisiae strain, and most preferably yeast strain C468 described further here-inbelow.

DNA expression or DNA transfer vectors suitable for transfer and replication have been described, e.g., in References lc and Id. Many of the yeast vectors i n present use are deri ved from E. coli vectors such as pBR322. These references, lc and Id in particular, describe integrative transformation where the microorganism host Is transformed with vectors with no origin of replication that Integrate Into the host chromosome and are maintained and replicated as part of that chromosome. In another embodiment of this invention the host may be transformed by autonomous replication where the vectors contain DNA segments which serve as origins of DNA replication in the host cell. Vectors containing autonomously replicating segments are also described in Reference le. Preferably the DNA segment capable of 16 functioning as an origin of replication is from yeast. Two types of such origins of replication from yeast are: one derived from a naturally occurring yeast plasmid, commonly referred to as the 2 micron circle, which confers the ability to replicate independently of 5 yeast chromosomal DNA, and one derived from the yeast chromosomal replication origin containing a replication origin sequence termed ars (autonomous replication sequence), which also provides autonomous replication capability.

The expression vector of this invention necessarily contains 10 a promoter fragment which functions in microorganisms, i.e., the host being employed, as well as the modified DNA sequence coding for the fungal glucoamylase protein. The protein-encoding segment must be so oriented with the promoter fragment that in a microorganism host it is expressed to produce non-native glucoamylase. For bacteria such as E_. 15 coll a trp promoter is preferred. For yeast, a yeast promoter fragment is preferred. Among possible yeast promoter fragments for purposes herein are included, e.g., alcohol dehydrogenase (ADH-I), 3-phosphoglycerokinase (PGK), pyruvate kinase (PYK), triose phosphate isomerase (TPI), beta-isopropylmalate dehydrogenase (LEU2), glycer-20 aldehyde 3-phosphate dehydrogenase (TDH), enolase I (EN01), and the like. A preferred promoter fragment for purposes* herein Is from the enolase I gene.

The expression vector herein also preferably contains a microorganism transcription terminator segment following the segment 25 coding for the protein, 1n a direction of transcription of the coding segment. Examples of possible transcription segments include the 3' segments of the above-listed genes. A preferred transcription terminator segment Is from the enolase I gene.

A preferred host system consists of the S. cerevisiae yeast 30 host strain C468 transformed by the plasmid pGAC9. This preferred transformed yeast strain was deposited with the American Type Culture Collection (ATCC), 12301 Parklawn Drive, Rockvllle, MD 20852 on November 17, 1983 and assigned ATCC Deposit Number 20,690. Another preferred host system consists of the JL. coli host strain MH70 trans 17 formed by the plasmid pGC24, which transformant was deposited with the ATCC on December 16, 1983, and assigned ATCC Deposit Number 39,537.

The A. awamori glucoamylase signal sequence described below is shown to function in yeast for the efficient processing and secre-5 tion of glucoamylase from yeast. This sequence could also be used for the secretion of other proteins from yeast and preferably for the secretion of proteins that are normally secreted by their native host. Examples of such proteins include amylases, cellulases, proteases, interferons, lymphokines, insulin, and hormones. 10 The following examples serve'to exemplify the practice of the invention. They are presented for illustrative purposes only, and should not be construed as limiting the invention in any way. Percentages are by weight unless specified otherwise. All experiments were performed following the NIH (U.S.A.) guidelines for containment. 15 EXAMPLES All of the strains employed in the examples which have been deposited in depositories were deposited either with the U.S. Department of Agriculture Agricultural Research Service, National Regional Research Laboratories (NRRL) of Peoria, IL 61604 or with the American 20 Type Culture Collection (ATCC) of Rockville, MD 20852. Each strain deposited with ATCC has the individual ATCC designations indicated in the examples pursuant to a contract between the ATCC and the assignee of this patent application, Cetus Corporation. The contract with ATCC provides for permanent availability of the progeny of these strains to 25 the public on the issuance of the U.S. patent describing and identifying the deposits or the publications or upon the laying open to the public of any U.S. or foreign patent application, whichever comes first, and for availability of the progeny of these strains to one determined by the U.S. Commissioner of.Patents and Trademarks to be 30 entitled thereto according to 35 U.S.C. 122 and the Commissioner's rules pursuant thereto (including 37 CFR 1.14 with particular reference to 886 OG 638). The assignee of the present application has agreed that if any of these strains on deposit should die or be lost 1 8 or destroyed when cultivated under suitable conditions, it will be promptly replaced on notification with a viable culture of the same strain. The NRRL deposits mentioned in the examples and not designated patent deposits have been freely available to the public prior 5 to the filing date of this application. In the examples all parts and percentages are given by weight and all temperatures in degrees Celsius unless otherwise noted.

EXAMPLE 1 Determination of Nucleotide Sequence of Glucoamylase Gene 10 Experimentally, A. awamori cells were grown on either starch or xylose, as a primary source of carbon. The b. awamori cells were obtained from NRRL, Deposit Number 3112, and have been recently re-deposited and assigned NRRL Deposit Number 15271. Fungal growth was initiated from a suspension of spores in water. The fungal cells were 15 grown in an agitated culture at 30°C for 2-5 days in a standard growth medium (1% w/v yeast extract, 0.01 M ammonium sulfate, 0.025 M potassium phosphate buffer, pH 7.0) together with 5% w/v of either starch or xylose. As noted above, cells grown on starch produced an amount of glucoamylase in the culture medium, based on measurable enzyme 20 activity per cell dry weight, that was about 200 times that of cells grown on xylose.

Total cellular RNA was isolated from the fungal cultures by a guanidium thiocyanate/CsCl procedure essentially as described in Reference 2. Briefly, mycelia were wrung dry in cheese-cloth, frozen 25 in liquid nitrogen, and ground to a powder in a mortar and pestle in liquid nitrogen. The cell powder was homogenized in a guanidium thlo-cyanate solution containing 10 mM adenosine: voso4 complex. Following centrifugation to pellet cellular debris, CsCl was added to the homog-enate and the RNA was pelleted through a pad of CsCl by a-high speed 30 centrifugation.

Poly A containing RNA (poly A RNA) was Isolated from total RNA by two passages over ollgo-dT cellulose, conventionally, and the poly A RNA was size-fractionated by agarose gel electrophoresis, according to standard procedures.

The induced poly A RNA was extracted from the agarose gel essentially as described in Reference 3. Briefly, the gel was melted and then frozen to release the RNA into solution. The solidified agarose was removed by centrifugation. The extracted poly A RNA was extracted with phenol and precipitated with ethanol.

To examine the translation products of the induced poly A RNA in a cell-free protein synthesizing system, antibodies against A. awamori glucoamylase were prepared. Glucoamylase-I and II from k. awamori were obtained from the filtrate of a culture of k. awamori cells grown under glucoamylase induction conditions. The filtrate was fractionated by ion exchange chromatography using a diethyl aminoethyl-cellulose column. Elution with a pH gradient ranging from pH 8.0 to pH 3.0 yielded two protein peaks that showed glucoamylase activity. The enzyme tha-t'eluted at the lower pH included the larger gluco-amylase-I, and the other peak, glucoamylase-II. Gel electrophoresis indicated that glucoamylase-II was pure, but that glucoamylase-I was not. Glucoamylase-I was purified further by molecular sieve chromatography on a cross-linked dextran, Sepharcryl S-200 column. Two peaks were observed, one of them containing glucoamylase-I, which'was shown to be pure. For both enzyme forms, enzyme purity was established by polyacrylamide gel electrophoresis under non-detergent conditions, and by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE).

The two purified glucoamylase forms were used separately to raise anti-glucoamylase antibodies 1n rabbits. Each of the two Immunoglobulin G(IgG) antibody fractions produced were able to neutralize the glucoamylase activity of both glucoamylase forms. Further, Ouchterlony analysis of the two antibody fractions with the two enzyme forms indicated that each antibody reacts immunospecifically with both enzyme forms.

Poly A RNA from Induced and nonlnduced jA. awamori was used to direct the synthesis of radioactlve-methionine-labeled polypeptides 20 in a rabbit reticulocyte lysate kit obtained from New England Nuclear Co., Boston, Mass., and sold as Reticulocyte Lysate/Methionine L-[3®S]-Translation System. References 4 and 5 describe typical reticulocyte lysate systems. After a defined reaction period, aliguots of 5 the lysate were removed and analyzed, either before or after reaction with anti-glucoamylase antibody or normal rabbit immunoglobulin G (IgG), by SDS-PAGE. The immunoreactive products were precipitated essentially according to the method described in Reference 6.

To determine the molecular basis for the accumulation of 10 glucoamylase protein in starch-grown, but not xylose-grown, cultures of A. awamori, glucoamylase mRNA levels were examined. Total cellular mRNA was isolated and used to direct the synthesis of k. awamori pro-tein in a rabbit reticulocyte lysate system. The translation products were immunoprecipitated using rabbit anti-glucoamylase antibody 15 (lane 1, xylose-grown cells; lane 3, starch-grown cells) or normal rabbit antibody (lane 2, xylose-grown cells; lane 4, starch-grown cells). The results are shown in FIG.l and demonstrate the presence of translatable glucoamylase mRNA in RNA from starch-grown cells. In contrast, no glucoamylase mRNA was detected in xylose-grown cells. 20 This correlates with the 200-fold difference in glucoamylase protein observed in culture supernatants of these cells. Thus, the accumulation of glucoamylase protein in starch-grown cultures appears to result from a comparable increase in translatable glucoamylase mRNA.

MeHgOH-agarose gel electrophoresis of mRNA from starch-grown 25 cells revealed a major approximately 2.2 kllobase mRNA (indicated by an arrow), which was absent 1n mRNA from xylose-grown cells (FIG. 2A). It appeared likely that this predominant 'Induced1 mRNA represented the mRNA of the highly expressed, 'Induced' glucoamylase. To Identify the 'induced1 mRNA, the approximately'2.2-30 kilobase mRNA band was eluted from a gel and translated 1n the rabbit reticulocyte lysate system. Immunoprecip1tat1on of the protein product with rabbit antl-glucoanylase antibody demonstrated the presence of mRNA encoding glucoamylase within the approximately 2.2-kilobase 'induced' mRNA band (FIG. 2B). 21 According to one aspect of the invention, isolated glileoanal ase mRNA from the selected fungal species was used to produce a glucoamylase cDNA by reverse transcription of the mRNA. Experimentally, induced poly A RNA from A_. awamori was pretreated with 10 mM 5 MeHgOH to denature the RNA, and then introduced into a reaction containing oligo-dT as a primer and 2 nM adenosine: voso4 as an RNAse inhibitor. The reader is referred to Reference 7 for a discussion of this general technique. Following cONA synthesis, the poly A RNA was destroyed by treatment with NaOH. The synthesized cDNA was size frac-10 tionated by gel electrophoresis to separate the full-length cDNA from incompletely formed fragments. A typical gel electrophoretic pattern of the cONA fraction showed a single detectable band in the approximately 2.2 kilobase size region.

The induced glucoamylase mRNA and the cDNA produced there-15 from were radiolabeled to provide probes for identifying genomic DNA fragments containing all or portions .of the homologous glucoamylase gene. The cDNA may be labeled readily by performing its synthesis in the presence of radiolabeled nucleotides.

The basic method used for radiolabeling mRNA is discussed in 20 Reference 8. In one example, induced poly A RNA from A_. awamori was partially degraded, using sodium hydroxide to generate fragments containing 5*-OH groups. These fragments were subsequently phosphorylated with radioactive-phosphate (^P)-ATP using a poly-nucleotide kinase. The P-labeled RNA fragments span the entire 25 length of the isolated RNA, and are thus advantageous for use as probes for genomic DNA fragments containing end portions of the glucoamylase gene.

Total genomic DNA isolated from A. awamori was digested to completion with each of a number of restriction endonucleases. The. 30 fragments were size-fractionated by gel electrophoresis and hybridized to one of the above RNA or cDNA probes by the Southern blot method (Reference 9). Details of this method are found generally by Reference 5, at page 387. Briefly, a prehybridization step was performed at 42#C for 24 hours, using a five-times concentrate of standard 2 ^ saline citrate (0.15M sodium chloride, 0.015M trisodium citrate).

This was followed by a hybridization step carried out at 42°C for 24 hours, using a two-times concentrate of the standard saline citrate. In the studies involving A. awamori genomic DNA, several of 5 the endonucleases used—including. Hindi11, Xhol, Bell, and Pvul— generated only one fragment which hybridized to the above A. awamori labeled RNA or cDNA probes. Some of the single gene fragments are in the same size range as the RNA transcript, strongly indicating that A. awamori contains only one gene which codes for the glucoamylase poly-10 peptide. EcoRI generated a 3.4 kilobase fragment which hybridized to the labeled cDNA.

The A. awamori genomic DNA fragments produced by digestion with EcoRI were spliced, by conventional techniques, into a lambda Charon 4A phage vector. The library of EcoRI fragments were screened 15 for recombinants which hybridized to the A. awamori glucoamylase cDNA. Hybridizing plaques were purified, and all contained a common 3.4 kilobase EcoRI fragment which hybridized to the glucoamylase cDNA probe. This 3.4 kilobase EcoRI fragment was then subcloned into the EcoRI site of a pACYC184 plasmid (ATCC Deposit No. 37,033), producing 20 a recombinant plasmid which is designated herein as pGARl. A sample of JE. coli K12 strain MM294 transformed with pGARl was deposited in the American Type Culture Collection, 12301 Parklawn Drive, Rockville, MD 20852, USA on December 2, 1983, and has been assigned ATCC Number 39,527. Subsequent libraries were screened using pGARl as probe. 25 Approximately 20 kilobases of A. awamori genomic DNA surrounding the glucoamylase gene was isolated from EcoRI, Hindi 11 and BglH libraries. A composite restriction map of this 20 kilobase region 1s shown in FIG. 3; the EcoRI fragment insert is expanded. The locations of the cleavage sites of the designated restriction endonucleases were 30 determined by digesting the plasmids with selected combinations of'the endonucleases, and size-fractionating the fragments obtained, according to known methods. The five solid rectangles represent sequenced protein-encoding regions of the glucoamylase gene. The direction of transcription of the mRNA 1s Indicated by the 5' to 3* line. 23 The plasmid pGARl was confirmed to contain glucoamylase gene sequences by virtue of its ability to hybridize to and select awamori glucoamylase mRNA sequences. pGARl DNA was immobilized onto nitrocellulose and hybridized to total A. awamori mRNA. The selected 5 mRNA was translated in vitro, and the products were identified by immunoprecipitation with rabbit anti-glucoamylase antibody. The results, shown in FIGURE 4, confirm the identification of pGARl, and thus of the approximately 2.2 kilobase "induced" mRNA, as encoding glucoamylase. In FIGURE 4 total A. awamori mRNA (lane 1) and mRNA 10 isolated by virtue of hybridization to pGARl DNA (lane 2) was translated jn^vitro^ and the protein products are displayed. Protein products of lane 2 were immunoprecipitated using rabbit antigluco-amylase antibody (lane 3) or normal rabbit antibody (lane 4).

Subclone pGARl containing the A. awamori glucoamylase gene 15 was digested substantially to completion with various restriction enzymes whose sequences are included within the EcoRI fragment (i.e., those in FIGURE 6), and several of the fragments were subcloned into M13 vectors M13mp8 and M13mp9. These bacteriophage vectors are available from Bethesda Research Laboratories, P.O. Box 6009, Gaithersburg, 20 MD 20877.

The fragments of the glucoamylase genomic region subcloned into the vectors M13mp8 and M13mp9 were sequenced by the dideoxy-nucleotide chain termination method described in References 10 and 11. Portions of the sequence were confirmed by the Maxam-Gi1bert 25 sequencing technique (Reference 12). The entire sequence of the 3.4 kilobase EcoRI fragment Is shown in Table I below.

TABLE I GAATTCAAGC TAGATGCTAA GCGATATTGC ATGGCAATAT GTGTTGATGC 50 ATGTGCTTCT TCCTTCAGCT TCCCCTCGTG CAGATGAAGG TTTGGCTATA 100 30 AATTGAAGTG GTTGGTCGGG GTTCCGTGAG GGGCTGAAGT GCTTCCTCCC 150 TTTTAGACGC AACTGAGAGC CTGAGCTTCA TCCCCAGCAT CATTACACCT 200 CAGCA ATG TCG TTC CGA TCT CTA CTC GCC CTG AGC GGC CTC GTC 244 24 MET SER PHE ARG SER LEU LEU ALA LEU SER GLY LEU VAL -12 TGC ACA GGG TTG GCA AAT GTG ATT TCC AAG CGC GCG ACC TTG GAT 289 CYS THR GLY LEU ALA ASN VAL ILE SER LYS ARG Ala Thr Leu Asp 4 TCA TGG TTG AGC AAC GAA GCG ACC GTG GCT CGT ACT GCC ATC CTG 334 5 Ser Trp Leu Ser Asn Glu Ala Thr Val Ala Arg Thr Ala lie Leu 19 AAT AAC ATC GGG GCG GAC GGT GCT TGG GTG TCG GGC GCG GAC TCT 379 Asn Asn lie Gly Ala Asp Gly Ala Trp Val Ser Gly Ala Asp Ser 34 GGC ATT GTC GTT GCT AGT CCC AGC ACG GAT AAC CCG GAC T 419 Gly lie Val Val Al,a Ser Pro Ser Thr Asp Asn Pro Asp 47 10 gtatgtttcg agctcagatt tagtatgagt gtgtcattga ttgattgatg 469 ctgactggcg tgtcgtttgt tgtag AC TTC TAC ACC TGG ACT CGC GAC 517 Tyr Phe Tyr Thr Trp Thr Arg Asp 55 TCT GGT CTC GTC CTC AAG ACC CTC GTC GAT CTC TTC CGA AAT GGA 562 15 Ser Gly Leu Val Leu Lys Thr Leu Val Asp Leu Phe Arg Asn Gly 70 GAT ACC AGT CTC CTC TCC ACC AH GAG AAC TAC ATC TCC GCC CAG 607 Asp Thr Ser Leu Leu Ser Thr He Glu Asn Tyr He Ser Ala Gin 85 GCA ATT GTC CAG GGT ATC AGT AAC CCC TCT GGT GAT CTG TCC AGC 652 Ala He Val Gin Gly He Ser Asn Pro Ser Gly Asp Leu Ser Ser 100 20 GGC GGT CTC GGT GAA CCC AAG TTC AAT GTC GAT GAG ACT GCC TAC 697 Gly Gly Leu Gly Glu Pro Lys Phe Asn Val Asp Glu Thr Ala Tyr 115 ACT GGT TCT TGG GGA CGG CCG CAG CGA GAT GGT CCG GCC TCT GAG 742 Thr Gly Ser Trp Gly Arg Pro Gin Arg Asp Gly Pro Ala Ser Glu 130 AGA ACT GCT ATG ATC GGC TTC GGG CAA TGG CTG CTT gtatgttctc 788 25 Arg Thr Ala Met lie Gly Phe Gly Gin Trp Leu Leu 142 cacccccttg cgtctgatct gtgacatatg tagctgactg gtcag GAC AAT 839 Asp Asn 144 GGC TAC ACC AGC ACC GCA ACG GAC ATT GTT TGG CCC CTC GTT AGG 884 Gly Tyr Thr Ser Thr Ala Thr Asp He Val Trp Pro Leu Val Arg 159 AAC GAC CTG TCG TAT GTG GCT CAA TAC TGG AAC CAG ACA GGA TAT 929 30 Asn Asp Leu Ser Tyr Val Ala Gin Tyr Trp Asn Gin Thr Gly Tyr 174 G gtgtgtttgt tttattttaa atttccaaag atgcgccagc agagctaacc 980 cgcgatcgca g AT CTC TGG GAA GAA GTC AAT GGC TCG TCT TTC Asp Leu Trp Glu Glu Val Asn Gly Ser Ser Phe 1023 185 25 TTT Phe ACG Thr ATT GCT lie Ala GTG Val CAA CAC Gin His CGC Arg GCC Ala cn GTC Leu Val GAA GGT Glu Gly AGT Ser GCC Ala 1068 200 TTC Phe GCG Ala ACG GCC Thr Ala GTC Val GGC TCG Gly Ser TCC Ser TGC Cys TCC TGG Ser Trp TGT GAT Cys Asp TCT Ser CAG Gin 1113 215 5 GCA Ala CCC Pro GAA ATT Glu lie CTC Leu TGC TAC Cys Tyr CTG Leu CAG Gin TCC TTC Ser Phe TGG ACC Trp Thr GGC Gly AGC Ser 1158 230 TTC Phe ATT lie CTG GCC Leu Ala AAC Asn TTC GAT Phe Asp AGC Ser AGC Ser CGT TCC Arg Ser CGG AAG Arg Lys GAC Asp GCA Ala 1203 245 10 AAC Asn ACC Thr CTC CTG Leu Leu GGA Gly AGC ATC Ser He CAC His ACC Thr TTT GAT Phe Asp CCT GAG Pro Glu GCC Ala GCA Ala 1248 260 TGC Cys GAC Asp GAC TCC Asp Ser ACC Thr TTC CAG Phe Gin CCC Pro TGC Cys TCC CCG Ser Pro CGC GCG Arg Ala CTC Leu GCC Ala 1293 275 AAC Asn CAC His AAG GAG Lys Glu GTT Val GTA GAC Val Asp TCT Ser TTC Phe CGC TCA Arg Ser ATC TAT He Tyr ACC Thr CTC Leu 1338 290 15 AAC Asn GAT Asp GGT CTC Gly Leu AGT Ser GAC AGC Asp Ser GAG Glu GCT Ala GTT GCG Val Ala GTG GGT Val Gly CGG Arg TAC Tyr 1383 305 CCT Pro GAG Glu GAC ACG Asp Thr TAC Tyr TAC AAC Tyr Asn GGC Gly AAC Asn CCG TGG Pro Trp TTC CTG Phe Leu TGC Cys ACC Thr 1428 320 20 TTG Leu GCT Ala GCC GCA Ala Ala GAG Glu CAG TTG Gin Leu TAC Tyr GAT Asp GCT CTA Ala Leu TAC CAG Tyr Gin TGG Trp GAC Asp 1473 335 AAG Lys CAG Gin GGG TCG Gly Ser TTG Leu GAG GTC Glu Val ACA Thr GAT Asp GTG TCG Val Ser CTG GAC Leu Asp TTC Phe TTC Phe 1518 350 AAG lys GCA Ala CTG TAC Leu Tyr AGC GAT GCT Ser Asp Ala GCT Ala ACT Thr GGC ACC Gly Thr TAC TCT Tyr Ser TCG Ser TCC Ser 1563 365 25 AGT Ser TCG Ser ACT TAT Thr.Tyr AGT AGC ATT Ser Ser He GTA Val GAT Asp GCC GTG Ala Val AAG ACT Lys Thr TTC Phe GCC Ala 1608 380 GAT Asp GGC Gly TTC GTC Phe Val TCT ATT GTG Ser He Val gtaagtctac gctagacaag cgctcatgtt 1659 387 30 gacagagggt gcgtactaac agaagtag GAA ACT CAC GCC GCA AGC AAC Glu Thr His Ala Ala Ser Asn 1708 394 GGC Gly TCC Ser ATG TCC Met Ser GAG Glu CAA TAC Gin Tyr GAC Asp AAG TCT GAT GGC GAG CAG Lys Ser Asp Gly Glu Gin CTT Leu 1753 409 TCC Ser GCT Ala CGC GAC Arg Asp CTG Leu ACC TGG Thr Trp TCT Ser TAT GCT GCT CTG CTG ACC Tyr Ala Ala Leu Leu Thr GCC Ala 1798 424 35 AAC AAC CGT CGT AAC GTC GTG CCT TCC GCT TCT TGG GGC GAG ACC Asn Asn Arg Arg Asn Ser Val Val Pro Ala Ser Trp Gly Glu Thr 1843 439 1888 454 1933 469 1978 484 2023 499 2068 514 2113 529 2158 544 2203 559 2248 574 2293 589 2338 604 2381 615 2431 2481 2531 2581 2631 2681 2731 2781 26 TCT Ser GCC Ala AGC Ser AGC Ser GTG Val CCC Pro GGC ACC Gly Thr TGT Cys GCG Ala GCC Ala ACA Thr TCT Ser GCC ATT Ala He GGT Gly ACC Thr TAC Tyr AGC Ser AGT Ser GTG Val ACT GTC Thr Val ACC Thr TCG Ser TGG Trp CCG Pro AGT Ser ATC GTG lie Val GCT Ala ACT Thr GGC Gly GGC Gly ACC Thr ACT Thr ACG ACG Thr Thr GCT Ala ACC Thr CCC Pro ACT Thr GGA Gly TCC GGC Ser Gly AGC Ser GTG Val ACC Thr TCG Ser ACC Thr AGC Ser AAG ACC Lys Thr ACC Thr GCG Ala ACT Thr GCT Ala AGC Ser AAG ACC Lys Thr AGC Ser ACC Thr AGT Ser ACG Thr TCA Ser TCA Ser AGC TCC Ser Ser TGT Cys ACC Thr ACT Thr CCC Pro ACC Thr GCC GTG Ala Val GCT Ala GTG Val ACT Thr TTC Phe GAT Asp CTG Leu ACA GCT Thr Ala ACC Thr ACC Thr ACC Thr TAC Tyr GGC Gly GAG AAC Glu Asn ATC lie TAC Tyr CTG Leu GTC Val GGA Gly TCG Ser ATC TCT He Ser CAG Gin CTG Leu GGT Gly GAC Asp TGG Trp GAA ACC Glu Thr AGC Ser GAC Asp GGC Gly ATA lie GCT Ala CTG Leu AGT GCT Ser Ala GAC Asp AAG Lys TAC Tyr ACT Thr TCC Ser AGC GAC Ser Asp CCG Pro CTC Leu TGG Trp TAT Tyr GTC Val ACT Thr GTG ACT Val Thr CTG Leu CCG Pro GCT Ala GGT Gly GAG Glu TCG TTT Ser Phe GAG Glu TAC Tyr AAG Lys TTT Phe ATC He CGC Arg ATT GAG He Glu AGC Ser GAT Asp GAC Asp TCC Ser GTG Val GAG TGG Glu Trp GAG Glu AGT Ser GAT Asp CCC Pro AAC Asn CGA Arg GAA TAC Glu Tyr ACC Thr GTT Val CCT Pro CAG Gin GCG TGC GGA Ala Cys Gly ACG Thr TCG Ser ACC Thr GCG Ala ACG Thr GTG Val ACT GAC Thr Asp ACC Thr TGG Trp CGG Arg TAGACAATCA ATCCATTTCG CTATAGTTAA AGGATGGGGA TGAGGGCAAT TGGTTATATG ATCATGTATG TAGTGGGTGT GCATAATAGT AGTGAAATGG AAGCCAAGTC ATGTGATTGT AATCGACCGA CGGAATTGAG GATATCCGGA AATACAGACA CCGTGAAAGC CATGGTCTTT CCTTCGTGTA GAAGACCAGA CAGACAGTCC CTGATTTACC CTGCACAAAG CACTAGAAAA TTAGCATTCC ATCCTTCTCT GCTTGCTCTG CTGATATCAC TGTCATTCAA TGCATAGCCA TGAGCTCATC TTAGATCCAA GCACGTAATT CCATAGCCGA GGTCCACAGT GGAGCAGCAA CATTCCCCAT CAnGCTTTC CCCAGGGGCC TCCCAACGAC TAAATCAAGA GTATATCTCT ACCGTCCAAT AGATCGTCTT CGCTTCAAAA TCTTTGACAA 2831 TTCCAAGAGG GTCCCCATCC ATCAAACCCA GTTCAATAAT AGCCGAGATG 2881 CATGGTGGAG TCAATTAGGC AGTATTGCTG GAATGTCGGG GCCAGTTCCG 2931 GGTGGTCATT GGCCGCCTGT GATGCCATCT GCCACTAAAT CCGATCATTG 2981 ATCCACCGCC CACGAGGGCG TCTTTGCTTT TTGCGCGGCG TCCAGGTTCA 3031 ACTCTCTCTG CAGCTCCAGT CCAACGCTGA CTGACTAGTT TACCTACTGG 3081 TCTGATCGGC TCCATCAGAG CTATGGCGTT ATCCCGTGCC GTTGCTGCGC 3131 AATCGCTATC TTGATCGCAA CCTTGAACTC ACTCTTGTTT TAATAGTGAT 3181 CTTGGTGACG GAGTGTCGGT GAGTGACAAC CAACATCGTG CAAGGGAGAT 3231 TGATACGGAA TTGTCGCTCC CATCATGATG TTCTTGCCGG CTTTGTTGGC 3281 CCTATTCGTG GGATCGATGC CCTCCTGTGC AGCAGCAGGT ACTGCTGGAT 3331 GAGGAGCCAT CGGTCTCTGC ACGCAAACCC AACTTCCTCT TCATTCTCAC 3381 GGATGATCAG GATCTCCGGA TGAATTC 3408 5 The nucleotide sequence obtained was compared, in a com- puter-programmed matching operation, with the regions of known amino acid sequence of A. niger (References 1a and 1b) and* A. awairori glucoamylase. The matching operation examined the nucleotide sequence in each of the six possible reading frames for codon correspondence with the given amino 0 acid sequence. The matching operation produced nearly-complete correspondence between coding regions of the glucoamylase gene and the regions of known amino acid sequence of glucoamylase from A. awamori. The amino acid sequence of one of the internal peptides of A. niger (Fig. 7 of Reference la) was found not to be contiguously 5 encoded by the nucleic acid sequence (nucleotides 753-895 of Table I). An intervening sequence of 55 nucleotides was presumed to Interrupt this protein coding region. The Introns 1n the glucoamylase gene are in lower case. The amino acid sequence of the glucoamylase gene Is indicated below the appropriate nucleotides and Is numbered ) below the nucleotide sequence numbers at the right. Amino acids -24 to -1 (all capital letters) represent the signal sequence of the pre-glucoamylase protein. 28 To confirm the Identification of this interrupting sequence as the intervening sequence, and to identify other intervening sequences within the glucoamylase gene, the cDNA sequences derived from glucoamylase mRNA were molecularly cloned. Double-stranded cDNA was 5 prepared from mRNA of starch-grown A. awamori and a cDNA library was prepared in pBR322, also available from Bethesda Research Laboratories, as described above. Sixteen glucoamylase cDNA-containing plasmids were identified using pGARl probe; the largest plasmid, p24A2, which was deposited with the National Regional Research Labora-10 tory in Peoria, Illinois, USA on December 7, 1983 and assigned NRRL No. B-14217, contained 1.8 kilobases of sequence derived from the 3'-end of the approximately 2.2 kilobase glucoamylase mRNA. The nucleotide sequence of the glucoamylase cDNA in p24A2 was determined and found to span the genomic sequence, shown in Table I, from nucleotide 15 501 through the polyadenylation site at position 2489-2491. (The precise polyadenylation site cannot be determined unambiguously due to the presence of'two A residues at nucleotides 2490-2491.) Comparison of the nucleotide sequence of the molecularly cloned glucoamylase gene with that of the glucoamylase mRNA, as determined from molecularly 20 cloned glucoamylase cDNA, and with glucoamylase amino acid sequence, has revealed the presence of four intervening sequences (introns) within the A. awamori glucoamylase gene. (The junctions of the first intervening sequence were deduced from incomplete amino acid sequence data at residues 43-49 of A. awamori glucoamylase-I.) The intervening 25 sequences were short (ranging from 55 to 75 base pairs) and were all located within protein-encoding sequences. These sequences adjoining the Intervening sequence junctions of the glucoamylase gene were compared to consensus splice junction sequences from eucaryotes in general (Reference 13) and from S_. cerevisiae in particular (Ref-30 erence 14). Splice junctions within the glucoamylase gene conform-closely to the consensus sequences at the 5* and 3* Intervening sequence termini. Sequences related to the consensus sequence TACTAACA postulated by Langford, et al. In Reference 15 to be required for splicing in S. cerevisiae are found near the 3* terminus of all gluco-35 amylase intervening sequences. 29 The 5' end of the glucoamylase mRNA was determined using a synthetic oligonucleotide to prime reverse transcriptase synthesis from the mRNA template. Four major primer extension products were synthesized using the pentadecamer 5'GCGAGTAGAGATCGG3' which is com-5 plementary to sequences within the signal peptide-encoding region near the 5* end of the glucoamylase mRNA, as indicated in FIGURE 5.

The shorter band of the doublets is interpreted to represent the incompletely extended form of the longer band. To examine possible effects of RNA secondary structures on this pattern, primer 10 extension was preferred at 42 and 50°C. The products of primer extension at 42°C (lane 1) and 50aC (lane 2) are displayed on a sequencing gel described in Reference 16 in parallel with ml3/di-deoxynucleotide sequencing reactions of this region, using the identical pentadecamer primer. The sequence presented in FIGURE 5 15 represents the glucoamylase mRNA sequence and is complementary to that read from the sequencing reactions shown. The pattern of primer extension was unchanged, supporting the conclusion that four distinct 5' termini exist within the population of glucoamylase mRNA. Primer extension reactions performed in the presence of dideoxynucleotides 20 confirmed the colinearity of genomic and mRNA sequences in this region. The primer extension products map to T residues, at positions -71, -66, -59, and -52 from the site of translation initiation, and are indicated in Table I. To the extent that reverse transcriptase is able to copy Jthe extreme terminal nucleotlde(s) of the mRNA, the 51 25 termini of the glucoamylase mRNAs are localized to these four regions. DNA sequences 5* of the region of transcription initiation were found to contain sequences homologous to consensus sequences previously shown to be Involved In transcription initiation by RNA polymerase II. 30 Table IIA Illustrates the nucleotide sequence encoding the mature glucoamylase polypeptide. 30 table 11a GCG ACC nG GAT TCA TGG nG AGC AAC gaa GCG ACC GTG GCT CGT ACT GCC ATC CTG AAT AAC ATC GGG GCG GAC ggt GCT TGG GTG TCG GGC GCG GAC TCT GGC An GTC Gn GCT AGT CCC AGC ACG GAT AAC 5 CCG GAC tac nc TAC ACC TGG ACT CGC GAC TCT GGT CTC GTC CTC aag ACC CTC GTC GAT CTC nc CGA AAT GGA GAT ACC AGT CTC CTC TCC ACC An GAG AAC TAC ATC TCC GCC CAG GCA An GTC CAG GGT ATC AGT AAC CCC TCT GGT GAT CTG TCC AGC GGC GGT CTC GGT GAA CCC AAG nc AAT GTC GAT GAG ACT GCC TAC ACT GGT TCT TGG GGA 10 CGG CCG CAG CGA GAT GGT CCG GCC TCT GAG AGA ACT GCT ATG ATC GGC nc GGG CAA TGG CTG cn GAC AAT GGC TAC ACC AGC ACC GCA ACG GAC An Gn TGG CCC CTC Gn AGG AAC GAC CTG TCG TAT GTG GCT CAA TAC TGG AAC CAG ACA GGA TAT GAT CTC TGG GAA GAA GTC AAT ggc TCG TCT nc ■ni ACG An GCT GTG CAA CAC CGC gcc cn 15 GTC gaa GGT AGT GCC ne gcg ACG GCC GTC GGC TCG TCC TGC TCC tgg TGT GAT TCT CAG GCA CCC GAA An CTC TGC TAC CTG CAG TCC TTC tgg ACC GGC AGC nc An CTG GCC AAC nc GAT AGC AGC CGT TCC cgg AAG GAC GCA AAC ACC CTC CTG gga AGC ATC CAC acc rrr GAT CCT GAG GCC GCA TGC GAC GAC TCC ACC nc CAG CCC TGC TCC 20 ccg cgc GCG CTC gcc AAC CAC aag gag Gn GTA GAC TCT nc cgc TCA ATC TAT ACC CTC AAC GAT GGT CTC AGT GAC AGC GAG GCT GTT gcg GTG GGT CGG TAC CCT GAG GAC ACG TAC TAC AAC GGC AAC CCG tgg nc CTG TGC ACC nG GCT GCC GCA gag CAG nG TAC GAT GCT CTA TAC CAG TGG GAC AAG CAG GGG.TCG nG GAG GTC ACA GAT GTG 25 TCG CTG gac nc nc aag GCA CTG TAC AGC GAT GCT GCT ACT GGC ACC TAC TCT TCG TCC AGT TCG ACT TAT AGT AGC An GTA GAT GCC gtg aag act nc GCC GAT GGC nc GTC TCT An GTG GAA ACT CAC 3 1 GCC GCA AGC AAC GGC TCC ATG TCC GAG CAA TAC GAC AAG TCT GAT GGC GAG CAG CTT TCC GCT CGC GAC CTG ACC TGG TCT TAT GCT GCT CTG CTG ACC GCC AAC AAC CGT CGT AAC GTC GTG CCT TCC GCT TCT TGG GGC GAG ACC TCT GCC AGC AGC GTG CCC GGC ACC TGT GCG GCC 5 ACA TCT GCC ATT GGT ACC TAC AGC AGT GTG ACT GTC ACC TCG TGG CCG AGT ATC GTG GCT ACT GGC GGC ACC ACT ACG ACG GCT ACC CCC ACT GGA TCC GGC AGC GTG ACC TCG ACC AGC AAG ACC ACC GCG ACT GCT AGC AAG ACC AGC ACC AGT ACG TCA TCA AGC TCC TGT ACC ACT 10 CCC ACC GCC GTG GCT GTG ACT nc GAT CTG ACA GCT ACC ACC ACC TAC GGC GAG AAC ATC TAC CTG GTC GGA TCG ATC TCT CAG CTG GGT GAC TGG GAA ACC AGC GAC GGC ATA GCT CTG AGT GCT GAC AAG TAC ACT TCC AGC GAC CCG CTC TGG TAT GTC ACT GTG ACT CTG CCG GCT 15 GGT GAG TCG TTT GAG TAC AAG TTT ATC CGC ATT GAG AGC GAT GAC TCC GTG GAG TGG GAG AGT GAT CCC AAC CGA GAA TAC ACC GTT CCT CAG GCG TGC GGA ACG TCG ACC GCG ACG GTG ACT GAC ACC TGG CGG 20 Nucleotides 206 to 277 encode the signal sequence for the J\. awamori glucoamylase. As used in the specification and claims, the term "signal sequence" refers generally to a sequence of amino acids which are responsible for initiating export of a protein chain. A signal sequence, once having Initiated export of a growing protein 25 chain, is cleaved from the mature protein at a specific site. The term also includes leader sequences or leader peptides. The preferred signal sequence herein Is the deduced signal sequence from the A. -awamori glucoamylase gene given in Table IIB. 3 2 TABLE IIB MET SER PHE ARG SER LEU LEU ALA LEU SER GLY LEU VAL CYS THR GLY LEU ALA ASN VAL ILE SER LYS ARG EXAMPLE 2 Expression of Glucoamylase Gene in Yeast A. Construction of HindiII Cassette of Genomic Glucoamylase Gene ■ A method for expressing genes at high levels in yeast involves constructing vectors which contain the yeast enolase I pro-10 moter and terminator regions (Reference 16). The enolase segments were previously engineered so that the promoter and terminator were separated by a unique HindiII site.

Plasmid pACl (10.67 kilobase) is an £. coli/yeast shuttle vector, capable of autonomous replication in both E. coli and yeast 15 strains. The plasmid confers resistance in E_. coli and related species to the p-lactam antibiotic ampicillin and related compounds as a result of synthesis of the TEM type I p-lactamase. Further, the plasmid carries the yeast LEU2 gene which is expressed in both_E. coli and S. cerevisiae strains. Thus, the presence of the plasmid in 20 either coli or JS. cerevisiae strains reverses a leucine growth requirement resulting from loss of p-isopropylmalate dehydrogenase activity.

Plasmid pACl Is comprised of the following DMA segments. Numbering starts at the EcoRI site of the enolase I promoter fragment 25 and proceeds In a clockwise direction. Coordinates 0 to 725 comprise a 725 base pair EcoRI to HindiII DNA fragment derived from a similar fragment In the plasmid p eno 46 (Reference 16), containing DNA from the 5' untranslated region of the S. cerevisiae Enol gene. This fragment has been modified in the region just prior to the Initiation 30 codon (ATG) of the enolase gene in order to create a HindiII site. 33 Specifically, the sequence was changed from CACTAAATCAAAATG to CACGGTCGAGCAAGCTT(ATG). Coordinates 726 to 2281 comprise the 1.55 kilobase Hindi 11 to BglH DNA fragment from the 3' untranslated region of the cerevisiae Enol gene and was originally obtained from the 5 plasmid peno 46 (Reference 16). Coordinates 2282 to 2557 comprise a 275 basepair DNA fragment from the plasmid pBR322 (Reference 16a) between the BamHI and Sail recognition sites (pBR322 coordinates 375 to 650). Coordinates 2558 to 4773 comprise the 2.22 kilobase Xhol to Sail DNA fragment from j>. cerevisiae that encodes the LEU2 gene 10 product, p-isopropylmalate dehydrogenase. The plasmid YEpl3 (Reference 16b) provided a convenient source for the desired 2215 basepair DNA fragment. Coordinates 4474 to 8528 comprise a 3.75 kilobase DNA fragment which permits autonomous replication of the plasmid AC1 in yeast strains. This region encodes a portion of the 15 yeast 2(i plasmid and was derived from the plasmid pDB248 (Reference 16c). Digestion of plasmid pDB248 with the enzymes EcoRI and Sail liberated the desired 3.75 kilobase DNA fragment incorporated in plasmid AC1. Coordinates 8529 to 10672 comprise DNA sequences which permit autonomous replication in E_. coli host strains and confer 20 ampicillin resistance. The desired 2143 basepair DNA fragment was obtained from_E. coli plasmid pBR322 as a Tthllll to EcoRI DNA fragment (pBR322 coordinates 2218 and 4360, respectively). A sample of E. coli K12 strain MM294 transformed with pACl was deposited in the American Type Culture Collection on December 2, 1983 and has been 25 assigned ATCC No. 39,532.

The glucoanylase gene, while not having a convenient restriction site closely preceding Its Initiation codon (ATG) useful for cloning Into vectors, can have a single base pair change 32 base pairs upstream from the ATG so as to create a unique HindiII site, allowing 30 use of the enolase promoter for Initiation of transcription. Site--specific mutagenesis was used to obtain the desired mutation. A hexa-decamer oligonucleotide which 1s complementary to the region surrounding the desired HindiII site and which contains the appropriate mismatch was used to prime DNA synthesis on a single-stranded 35 M13 template of the glucoamylase gene. The sequence of the primer 34 employed was : GAGCCGAAGCTTCATC, with the mismatches underlined. A second mismatch was incorporated into the primer to aid in the screening for correct clones by hybridizing candidate plaques with the same oligonucleotide used for the primer extension, after the latter 5 had been radioactively labeled.

One picomole of a single stranded DNA phage, M13mp9 containing a 2.3 kilobase glucoamylase gene fragment (from EcoRI to Sail), was annealed to 10 picomoles of the primer in a 15 ^1 reaction mix which also contained 20 mM Tris pH 7.9, 20 mM MgC^, 100 nM NaCl, and 10 20 mM p-mercaptoethanol. The mixture was heated to 67°C, incubated at 37°C for 30 minutes, then placed on ice.

To the above annealing mixture 1 til of each deoxynucleotide triphosphate at 10 mM was added, to a final concentration of 500 |iM. Five units of E. coli Klenow fragment of DNA polymerase I (0.5 |il) was 15 then added and the extension reaction was left on ice for 30 minutes. Starting on ice minimizes 3'-5' exonuclease digestion of the primer and subsequent mismatch correction. After 30 minutes on ice,, the reaction was continued at 37°C for 2 hours, then inactivated by heating at 679C for 10 minutes. 20 Note that the primer was not kinased and no ligase was used in contrast to other published methods. JM103 competent cells were transformed with 1 jil of the reaction and either 5 ^1 or 50 ^1 were plated.

The hexadecamer used for priming was kinased with 25 labeled 32P-ATP to a specific activity of 3 x 107 cpm/ng. Nitrocellulose filters were used to bind phage DNA from the plaques by direct lifting, and these filters were denatured, neutralized and washed. After baking for 2 hours at 80°C, the filters were prehybridized for 3 hours at 45°C in 25 ml of a solution of 9 M NaCl 30 and 0.9 M sodium citrate, sodium dodecylsulfate, 50 ml of a solution of 0.5 g bovine serum albumin, 0.5 g Ficoll 400 (a carbohydrate polymer) and 0.5 g poly vinyl pyrrolidine, and 50 iig/ml yeast RNA.

After prehybridizatlon, 1.5 x 105 cpm/ml of kinased primer was added, and hybridization continued overnight at 45°C. The next day, filters 35 were washed 2 times, 5 minutes each in a solution of 9 M NaCl and 0.9 M sodium citrate at roughly 5°C (to remove non-specifically bound counts), then once at 45°C for 5 minutes (to remove probe hybridized to non-mutant phage DNA). Filters were air dried, put on Kodak XAR 5 (high speed) film with an intensifying screen and exposed overnight at -70°C.

One mutant clone among several thousand plaques was discovered in the first round of screening. Subsequent restriction enzyme digests of this clone confirmed the introduction of the HindiII 10 site in front of the glucoamylase gene.

In the next step a HindiII site was created at the 3' end of the glucoamylase gene. A clone with the engineered HindiII site near the 5" end of the gene was cut with Ncol, its sticky ends were converted to blunt ends by enzymatic repair using Klenow fragment of E_. 15 coli DNA polymerase-I, and it was cut with EcoRI. FIGURE 7 illustrates a restriction map of this region. This method produced a fragment containing the glucoamylase gene and having an EcoRI sticky end before the 5* end of the gene and a blunt end after the 3* end of the gene. This fragment was cloned into a polylinker region of plas-20 mid pUC8, available from Bethesda Research Laboratories, to place a Hindl11 site within 20 nucleotides of the 3' end of the fragment so as to produce a HindlII cassette.

B. Construction of Full-Length cDNA Clone of Glucoamylase Gene Lacking Introns 25 The longest cDNA clone produced and isolated which had regions homologous to the genomic clone of the glucoamylase gene, p24A2, corresponds in sequence to the genomic clone from nucleotides 501 to 2490, minus the nucleotides corresponding to Introns Indicated In lower case In Table I. This clone 1s still several hundred nucleo-30 tides shorter than necessary for a full-length cDNA clone. The construction of a full-length cDNA copy of the gene was accomplished In several steps. The genomic clone with the HindlII site near the 5* end of the gene was cut with EcoRI and Avail and this fragment was 3 6 purified. The longest cDNA clone described above was digested with Avail and PstI, and the small Avail to PstI fragment was purified. The phage vector M13mpll, available from P-L Biochemicals, 1037 W. McKinley Ave., Milwaukee, MI 53205, was digested with EcoRI and PstI, 5 and the large vector fragment was purified from the small polylinker fragment. These three fragments were ligated together to generate a M13mpll vector containing the EcoRI and PstI region of the genomic clone, but now missing the second intron.

The longest cDNA clone was then cut with PstI using con-10 ditions supplied by the manufacturer of the restriction enzyme and the large PstI fragment was isolated. The M13mpll vector described above was cut with PstI, and the large PstI fragment from the cDNA clone was ligated into this site. The clones generated from this ligation were screened to identify the clone with the PstI fragment inserted in the 15 correct orientation. The clone isolated from this step had the genomic sequence from EcoRI to Avail (containing the first intron and the new 5* Hi ndl11 site) and the cONA sequence from Avail to the PstI site beyond the poly-A tail region. The remaining intron at the 5' end of the gene was removed by site-directed mutagenesis using a 20 nonacosamer oligonucleotide to span the intron region. The nona-cosamer, which had homology to 15 base pairs on the 5' side of the intron and 14 base pairs on the 3' side, had the sequence: 5' CGGATAACCCGGACTACTTCTACACCTGG 3' In the procedure for conducting site-directed mutagenesis, 25 one picomole of a single-stranded DNA phage derivative designated as M13mp9 (which 1s commercially available), containing a 2.3 kilobase glucoamylase gene fragment (from EcoRI to Sail), was annealed.to 10 plcomoles of primer in 15 pi containing 6 mm of tris(hydroxy-methyl)aminomethane (hereinafter Tr1s) at pH-7.9, 6 mm MgCl2 and 30 100 mM NaCl. The mixture was heated to 67°C, Incubated at 37°C for 30 minutes, and then placed on ice. At this temperature, either half of the nonacosomer can anneal to Its complement on the template without the other, allowing the proper loop to be formed.

To the above annealing mixture 1 pi of each deoxynucleotlde triphosphate at 10 nM was added, to a final concentration of 500 pM. Five units of E. coli Klenow fragment of DNA polymerase I (0.5 pi) was then added and the extension reaction was left on ice for 30 minutes to minimize 3*-5' exonuclease digestion of the primer. After 30 minutes on ice, the reaction was continued at 37*C for 2 hours, and then inactivated by heating at 67*C for 10 minutes.

In the procedure employed herein the primer was not kinased and no llgase was employed in contrast to other published methods. JM 103 competent cells were transformed with 1 pi of the reaction and either 5 pi or 50 pi were plated. (JM103 is an E, coli strain distributed by Bethesda Research Laboratories, Inc., Gaithersburg, MD 20877.) The nonacosamer used for priming was kinased with labeled 3*P-ATP to a specific activity of 3 x 107 cpm/pg.

Nitrocellulose filters were employed to bind phage DNA from the plaques by direct lifting, and these filters were denatured, neutralized and washed. After baking for 2 hours at 80°C, the filters were prehybridlzed for 3 hours at 55*C in 25 ml of a solution of 9 M NaCl and 0.9 M sodium citrate, 0.1% sodium dodecyl sulfate, 50 ml of a solution containing 0.5 g bovine serum albumin, 0.5 g Ficoll 400 (which is a carbohydrate polymer obtainable from Pharmacia Fine Chemicals) and 0.5 g polyvinylpyrrolidone, and finally 50 pg/ml yeast RNA. After prehybrldlzatlon, 1.5 x 10s cpm/ml of kinased primer was added, and hybridization was continued overnight at 55°C.

The next day, the filters were washed two times for five' minutes each in a solution of 9 N NaCl and 0.9 M sodium citrate at roughly 5*C (to remove non-speclflcally bound counts), and then once at 55°C for five minutes (to remove probe hybridized to non-mutant phage DNA). Filters were air-dried and placed on Kodak XAR (high speed) film with an intensifying screen and exposed overnight at -70#C.

The frequency of positives recovered was about 4%. Positive candidate plaques were further examined by preparing mini-preps and 38 digesting them to see if a size reduction occurred due to removal of the 75 base pair intron. Sequencing of one of the positives revealed that the Intron had been precisely removed.

In the final step this plasmid vector was digested with S EcoRI and BamHI, and the fragment was purified and used to replace the EcoRI to BamHI fragment in the genomic HI ndl 11 cassette vector described under section A above. The result is a cONA HindlII cassette which will have the normal polyadenylation signal at the 3* end of the clone but lacks all four introns. 10 C. Yeast Strains Transformed with Yeast Expression Vector The intron-containing HindiII cassette of the genomic gluco-amylase gene as described in section A above was excised and inserted into a yeast expression vector plasmid pACl to produce a plasmid IS designated as pGC2l, the map of which Is presented in FIG. 8. A sample of JE. coli K12 strain MM294 transformed with pGC21 was deposited 1n the NRRL on December 7, 1983 and has been assigned NRRL. No. B-14215. A sample of Z. coll K12 strain MM294 transformed with pACl was deposited in the American Type Culture Collection on Decem-20 ber 2, 1983 and has been assigned ATCC No. 39,532. The cassette of full-length cDNA clone lacking Introns as described 1n section B above was similarly excised and inserted Into the vector pACl to produce a plasmid designated as pGAC9, the map of which Is presented in FIG. 7.

Plasmid.DNAs pGC21 and pGAC9 were amplified in£. coll, 25 purified on a cesium chloride gradient and used to transform two strains of yeast: yeast strain C468, which is a haploid Saccharomyces cerevisiae with auxotrophic markers for leucine and hlstidlne, and yeast strain H18( which 1s a hap1o1d_S. cerevisiae with auxotrophic markers for leucine and hlstidlne, which lacks the repressor for the 30 glucoamylase gene of Saccharomyces dlastaticus. Leu* transfomants were screened for expression of the Aspergillus awamori glucoamylase gene. H18 was deposited in the National Regional Research Laboratory in Peoria, Illinois, USA on December 7, 1983 and has been assigned - NRRL Number Y-12842. 39 Yeast strains which were transformed with the yeast expression vectors pGC21 and pGAC9 were compared with the same strains transformed with the parent plasmid pACl as a control for growth on various starches in liquid and on solid media. Three types of starch were used: "washed" starch (a soluble starch washed three times with 70% ethanol to remove sugars and short chain carbohydrates), cassava starch, and soluble potato (Baker's) starch. Yeasts transformed with any of the three plasmids grew on the three starches; however, the cDNA clones (pGAC9) always showed better growth than the other clones, both in liquid and on solid media. When Baker's starch, which is the most highly polymerized of the three starches, was used in solid media at a concentration of 2% (w/v), the plates were turbid. These plates were spread with yeast from both strains carrying the parent plasmid, the genomic clone or the cDNA clone, and with yeast strain Saccharomyces diastaticus, having NRRL Deposit No. Y-2044, which expresses a yeast glucoamylase. The plates are shown in FIG. 9. The strains carrying the cDNA clone (pGAC9) were able to clear the starch around the growth zone, indicating that they could degrade the starch completely. In contrast, the S. diastaticus strain and the yeast strains transformed with either the parent plasmid pACl or the genomic clone pGC21 were unable to clear the starch from around the growth area. The clearing of the highly polymerized starch exhibited by pGAC9-containing strains indicates the functional expression of the A. awamori glucoamylase gene that has both alpha 1-4 and alpha 1-6 amylase activity.

In another test for glucoamylase expression, yeast cells carrying the control plasmid, pACl, or the cDNA clone, pGAC9, were grown in a washed starch liquid medium. The cells were harvested and lysed by ten cycles of freeze, thaw, and vortexlng with glass beads. Each cell lysate, containing Intracellular proteins, was electro-phoresed on a 7% acrylamide gel containing 0.1% sodium dodecyl sulfate (SDS) and 7.6 M urea and transferred to cellulose paper activated with cyanogen bromide. After the proteins were transferred, the paper was first probed with antiserum from a rabbit Immunized against awamori glucoamylase and then with radioactively labeled Staph A protein that 40 binds to antibody molecules. After unbound radioactivity was washed off, the paper was dried and exposed to X-ray film. This technique, which is called a "Western" and is described in Reference 17, can be performed with antiserum or purified antibody. Protein that reacts 5 with glucoamylase antisera was detected in the lysates from the pGAC9 cDNA clones but not in the pACl controls.

The expression of the k. awamori glucoamylase gene was also tested directly by the ability of a yeast containing such a gene to grow on an otherwise non-utilizable carbon source. For yeast strains 10 C468 and H18, this growth test was accomplished using maltose as the carbon source, because both of these strains carry a mutation (mal) blocking the utilization of maltose as a carbon source. The ability of strains C468 and H18 containing the control plasmid pACl or the cDNA plasmid pGAC9 to grow on maltose and glucose as a carbon source 15 is indicated in Table III. The glucose plates contained histidine while the maltose plates contained both histidine and leucine supplementation. From this table it can be seen that the presence of the glucoamylase gene on the plasmid allows C468 to grow slowly on maltose and H18 to grow slightly better than the control. 20 These tests indicate that the presence of the glucoamylase gene complements the mal mutation in C468 and facilitates direct selection experiments where the growth of the yeast is solely dependent on proper and adequate functioning of the A. awamori gluco-amylase gene. 25 All of these experiments demonstrate that yeast strain C468 containing the plasmid pGAC9 is most superior in expressing the glucoamylase gene. A sample of yeast strain C468 transformed with pGAC9 was deposited with the American Type Culture Collection on November 17, 1983 and has been assigned the ATCC Deposit No. 20,690.

Table III Growth Response of Strain Carbon Source Glucose Maltose Yeast Strain PIasmi d day2 day4 day6 day2 day4 day6 daylO day 13 C468 pACl* + + + ■ 0 0 0 0 0 C468 pGAC9 ± + + 0 0 m + + H18 pACl* + + + 0 0 0 0 0 H18 pGAC9 + + ' + 0 0 0 0 m ^Control 0 = no visible colonies m = minute colonies < 0.3 mm ± = small colonies < 1 mm + = normal colonies 2-3 mm 42 D. 1. Characterization of Glucoamylase Activity in Yeast Cultures Standing cultures of yeast strain C468 containing pACl or containing pGAC9 prepared as described above were grown in minimal media with glucose or washed Difco soluble starch as the carbon 5 sources. The cultures were harvested, after 5 days for the glucose cultures and after 7 days for the starch cultures, and cell-free supernatants were prepared by centrifugation. These supernatants were concentrated 10-20 fold using an Amicon concentrator with a PMIO membrane. Glucoamylase assays were negative for the supernatants from 10 the glucose- and starch-grown cultures of yeast strain C468 containing pACl plasmid. In contrast, cells containing the control plasmid pGAC9 secreted approximately six units of glucoamylase activity per liter. (For a definition of a unit of glucoamylase activity, see the legend to Table IV). 15 Glucoamylase production in aerobic shake-flask cultures of yeast strain C468 containing pGAC9 plasmid was then assayed. After two days of incubation at 30°C and agitation at 250 rpm, the culture of C468 yeast strain containing pGAC9 had consumed all of the glucose and was in stationary phase. The culture had achieved a cell density 20 of approximately 2 g/liter dry weight. A glucoamylase assay on the unconcentrated supernatant indicated that approximately 47 units of activity per liter of supernatant was produced. 2. Location of Glucoamylase Activity in Cultures of Transformed Veast Cells 25 The experiment given below was used to resolve whether the majority of the glucoamylase activity is found in the culture medium or Inside the cell.

Strains C468-pGAC9 and C468-pACl were grown in 500 ml of medium containing 1.45 g of Difco Yeast nitrogen base (Difco Labora-30 tories, Detroit, MI 48232), 5.2 g of ammonium sulfate and 2% glucose per liter to a cell density of 2-3 x 107 cells per ml. The cultures were centrifuged at 49C and the supernatants and cell pellets were 43 processed separately. The supernatant samples were filtered through a 0.45 n filter and then concentrated 15 to 20X using an Amicon stirred cell with a PM-10 membrane. The cell pellet was washed once in 1 M Sorbitol 0.1 M phosphate buffer pH 7.5 and then the packed cell volume 5 was determined by centrifuging at approximately lOOOxg for 5 minutes in a conical graduated centrifuge tube. Each ml of packed cells was resuspended to 1.5 ml in 1.0 M Sorbitol-0.1 M phosphate buffer at pH 7.5 and and equal volume of Zymolyase 5000 (Miles Laboratory, Elkhart IN 46515) was added. The cells were gently mixed at room temperature 10 for 1 hr and then centrifuged at 500xg to recover the protoplasts.

The supernatant, representing the protein that was present between the cell wall and the inner membrane, was put on ice for later processing. The space between the cell membrane and wall in yeast is referred to as the interstitial space and this protoplast supernatant 15 sample will be referred to as the interstitial sample in the following text. The protoplasts were resuspended in 1 M Sorbitol-0.1 M kpo4 buffer-10 mM NaN3 and washed lx by centrifuging at 500xg. The pellet was resuspended in 5 ml 1 M Sorbitol-0.1 M kpo4 at pH 7.5-10 ntt NaN3 and 1 ml was used to assay the glucoamylase activity present in the 20 intact, azide-treated protoplasts. To the remaining 4 ml of protoplast 4 ml of 50 mM Tris at pH 7.4-10 mM EDTA was added along with 6 g of sterile glass beads (0.45-0.5 mm B. Braun) and the mixture was vortexed vigorously for 20 seconds, cooled on ice and this procedure was repeated until microscopic observation revealed membrane 25 ghosts or particles but few or no intact protoplasts. Sterile 2 M sucrose was added slowly with a pasteur pipette inserted to the bottom of the tube and the lysate was floated out of the glass beads. The lysate was removed to a new tube and centrifuged along with the interstitial sample at approximately 20,000xg for 30 min at 4°C. The 30 supernatant from the broken protoplasts was designated the Intracellular sample and the pellets from the interstitial sample and the broken protoplast sample were comhined to make the membrane sample. Thus the yeast culture has been fractionated Into five samples: the extracellular or supernatant sample, the Interstitial, 35 membrane associated and Intracellular samples, as well as a sample containing intact azide-treated protoplasts. 4 1 The culture samples were analyzed for glucoamylase activity utilizing the peroxidase-glucose oxidase (PGO)/o-dianisidine (ODAD) assay (Sigma Kit #510) which detects glucose released from soluble starch by the glucoamylase. The assay can be affected by other 5 enzymes present which utilize glucose or by glucose present in the samples. Each PG0-00AD Assay mix was tested°with known quantities of glucose (generally a dilution series from 0 to 550 nanomoles) and a standard curve was constructed. One glucoamylase unit is defined as the amount of glucoamylase which releases one pmole of glucose per 10 minute from washed soluble starch at 37°C.

Samples were reacted with washed soluble starch on the day they were prepared, then boiled and frozen at -20°C for later glucose assay. A portion of each fresh sample was precipitated by addition of 3 volumes of cold 95% ethanol, then allowed to stand overnight and the 15 precipitate was collected by centrifugation at 2000xg for 5 min at 49C. The supernatant sample required a second centrifugation to recover small floes which remained suspended in the ethanol supernatant. The pellets were dried and then resuspended in 50 mM Tris at pH 7.4-10 mM EDTA to one half their original volume, except the super-20 natant sample which was resuspended to one twentieth its original volume. These ethanol-precipitated samples were reacted with washed soluble starch and then boiled and frozen -20°C for assay with the fresh samples.

Intact azide-treated protoplasts were assayed in a reaction 25 mix containing 1 M Sorbitol-0.5% washed starch and 200 pi of protoplasts. These mixes were incubated at 37°C for 30 min, then centrifuged at 500xg and the supernatant was filtered, then boiled and assayed or stored at -20°C. These assays revealed that the reaction mix contained some residual glucose and that the protoplasts reduced 30 the amount of glucose in the mix during incubation. When lysed protoplasts were Incubated In the same mix, more glucose was utilized than when the protoplasts were Intact. Values for the glucoamylase plasmid carrying strain were similar to those for the strain carrying the same plasmid without the glucoamylase DNA insert, implying that little, 1f 35 any, glucoamylase activity is associated with the membrane. 45 The fresh fractionated samples were assayed and the intracellular samples were found to have residual glucose levels that were too high for the assay. Membrane-associated and interstitial samples from pGAC9- and pACl-transformed cells both failed to produce detect-5 able levels of glucose from soluble starch. The supernatant sample from pGAC9-transformed yeast demonstrated glucoamylase activity of about 22 units/liter, while the sample from pACl-transformed yeast showed no glucoamylase activity. Ethanol-precipitated samples from the pACl-transformed yeast showed negligible (less than or equal to 10 0.08 units/liter) or no glucoamylase activity. Ethanol-precipitated samples from yeast transformed with pGAC9 all demonstrated glucoamylase activity of 0.15 units per liter or higher. The supernatant sample contained over 90% of the total glucoamylase activity and the intracellular, membrane associated and interstitial samples contained 15 from 1 to 4% of the total activity depending on the sample. Therefore, most of the glucoamylase enzyme is secreted into the extracellular medium.

E. Production of Recombinant Glucoamylase from Yeast in a 10 Liter FermentoT ! : 20 To produce sufficient glucoamylase for characterization, a 10-liter fermentation of C468 yeast strain containing pGAC9 in minimal media with glucose as the sole carbon source was set up. A 100-ml seed culture was grown in minimal media to an optical density at 680 nm (ODggg) of 6 and added to the fermentor. The fermentor was run 25 as an aerobic batch fermentation until It reached an ODggg of 10, and then a glucose feed was begun. The glucose feed was continued to an 00680 of approximately 30 and then stopped, allowing the residual glucose to be consumed. Total fermentation time was approximately 32 hours. The final cell density was approximately 10 g/l1ter dry 30 weight. Diluted samples of the unconcentrated fermentor supernatant were assayed for glucoamylase activity, with the assay data given in Table IV. The supernatant was concentrated 15-fold using an Amicon Hollow Fiber Concentration unit with a 10,000 molecular weight size exclusion.

Sample Table IV Recombinant Glucoamylase Purification Glucoamylase Activity (units) Volume (ml) Protejp (mg) _ Specific Percent Activity Recovery (unlts/mg) Fermentor Supernatant Concentrated Supernatant DEAE-Sepharose Column 3146 1605 2300 10,000 660 160 219 173 100 51 73 7.3 13.3 One unit of glucoamylase activity Is the release of 1 pmole glucose/minute from washed Difco soluble starch In 0.1 M citrate buffer, pH 5.0, at 37°C.

** The protein concentration of the concentrated supernatant was determined using a BloRad protein assay kit. The protein concentration from the DEAE-Sepharose column was estimated by Integration of area under the ODggg peak (1 0D280 units = 1 mg/ml protein).

II 47 The concentrated fermentor supernatant was adjusted to 50 mM phosphate, pH 7.5, by adding concentrated buffer thereto and was loaded on a OEAE Sepharose (CL-6B) column. The column was eluted with a pH gradient (starting pH 75, final pH 3.0). The elution profile is 5 shown in FIG. 10. Various samples from the column were analyzed by SDS-urea polyacrylamide gel electrophoresis. A photograph of the gel stained with BioRad silver stain showed that the concentrated fermentor supernatant contained only a few proteins, demonstrating that the glucoamylase was secreted into the media and not released by cell 10 lysis. A comparison of a sample from this concentrated fermentor supernatant with an equal volume of the peak fraction of glucoamylase activity indicated a considerable increase in the purity of the protein. Estimates indicated that 20-30% of the supernatant protein was glucoamylase and the peak fraction was approximately 80% gluco-15 amylase. The recombinant glucoamylase migrated with a mobility slightly slower than the A. awamori glucoamylase, indicating that the glucoamylase produced in the transformed yeast was also glycosylated.

An assay on the peak column fraction of glucoamylase activity indicated that the recombinant glucoamylase has a specific activity 20 comparable to native A. awamori glucoamylase, namely 25-50 units/mg.

Experiments prove that the recombinant glucoamylase produced by yeast C468/pGAC9 Is glycosylated. Duplicate samples of A. awamori glucoamylase-I and glucoamylase-II and the recombinant glucoamylase were electrophoresed in a 10% polyacrylamide-SDS gel using stand-25 ard procedures. After electrophoresis, the gel was split and lanes 1-4 were stained for protein with a Coomassie Blue stain and lanes 5-7 were stained for carbohydrate with Periodic Acid Schiff's stain. Details of these procedures are found In Reference 18. A comparison of glucoamylase-I (lanes 2 and 5), glucoamylase-II (lanes 3-and 6) and 30 the recombinant glucoamylase (lanes 4 and 7) is shown 1n FIGURE 11. Since the bands corresponding to these proteins also stain with the carbohydrate stain, this demonstrates that the recombinant glucoamylase is glycosylated by the yeast. 4 8 EXAMPLE 3 Production of Alcohol from Transformed Yeast Yeast strain C468 containing pGAC9, and the control C468 yeast strain containing pACl were inoculated into 50 ml of the following medium: succinic acid 11.81 9 h3po4 0.58 9 h2so4 0.31 9 KC1 0.37 9 NaCl 58.4 mg MgCl2*6H20 0.2 9 MnS04*H20 1.7 mg CuS04*5H20 0.25 mg ZnS04*7H20 1.44 mg CoC12"6H20 1.19 mg Na2Mo04*2H20 1.21 mg h3bo3 3.09 mg CaCl2*2H20 . 14.7 mg FeS04*7H20 11.1 mg histidine 40 mg washed soluble starch* 100 9 add water in quantities sufficient to 1 liter Fermentation was carried out in 250 ml flasks which were equipped with air restrictors to restrict the flow of oxygen into the flask. The flasks were incubated at 32°C and shaken at 200 rpm for 7 days.

The starch was washed three times in 70% ethanol to remove low molecular weight carbohydrates. The precipitate was then dried, but some ethanol and water may have remained. 49 The ethanol content of each flask was evaluated using gas chromatography. The C468/pGAC9 culture contained 23.4 g/1 ethanol while the control C468/pACl culture contained 4.5 g/1 ethanol. The results show that the production of glucoamylase by the C468/pGAC9 5 culture enabled the strain to convert the soluble starch into glucose and then to ferment the glucose to ethanol.

EXAMPLE 4 Expression of the Glucoamylase Gene in E. coli In order to express the glucoamylase gene in E. coli, a 10 modification was made to the 5* untranslated region in order to make the DNA sequence more compatible with transcription and translation in E. coli. Specifically, 27 base pairs between the HindlII site which was constructed 32 base pairs upstream from the ATG initiation codon (see Example 2) and the ATG codon were deleted by oligonucleotide 15 mutagenesis using the procedure described in Example 2B for removal of an intron. The oligonucleotide, which had homology to 12 base pairs on the 5' side of the region to be deleted and 11 base pairs on the 3* side, had the sequence: 5' GAGCCGAAGCTTTATGTCGTTCCG 3' 20 Except for this deletion, the final HindlII cassette was identical to that constructed for the yeast expression vector in Example 2.

E. coli expression vector ptrp3 was constructed by replacing the EcoRI to CIal region of pBR322 (coordinates -3 to 28, see Reference 16a) with an EcoRI to Clal fragment containing the E.. coli 25 tryptophan promoter and ribosome binding site. The nucleotide sequence of this region is shown in Table V; the EcoRI, Clal and HindlII sites have been identified in Table Y. 5 0 Table V GAA TTC CGA CAT CAT AAC GGT TCT GGC AAA TAT TCT GAA ATG EcoRI AGC TGT TGA CAA TTA ATC ATC GAA CTA GTT AAC TAG TAC GCA AGT TCA CGT AAA AAG GGT ATC GAT AAG CTT Clal HindlII The HindlII cassette of the glucoamylase gene, described above in this example, was cloned into the HindlII site of ptrp3. Transformants were screened by DNA restriction fragment mapping in order to identify clones where the glucoamylase gene was in the same . orientation as the promoter; one such clone was selected for further study as pGC24.

In order to examine expression of the glucoamylase gene using the trp promoter, plasmid pGC24 was transformed into coli host MH70 which had been obtained from the E. coli Genetic Stock Center, Yale University (their collection number is CGSC 6153). MH70 is a mal" E_. coli strain whose genotype is araD139, a(argF-lac) 205, flbB5301, ptsF25, relAl?, rpsL150, ma!Q63, bg!R15, deoCl? The malQ mutation is in the amylomaltase gene; a mutation in this gene makes Z. coli unable to hydrolyze maltose to glucose.

A sample of the MH70 transformed with pCG24 was deposited with the American Type Culture Collection on December 16, 1983, and has been assigned the ATCC Deposit No. 39,537.

The MH70/pGC24 transformant and strain MH70 were grown at 37°C and 200 rpm in 5 ml of the following medium containing tryptophan at 50 mg/1. 51 25X Bonner-Vogel Salts 40 ml Ampicillln 50 mg Glucose 2 g Vitamin B1 10 mg 5 Casamlno Acid 2 g Water to 1000 ml 25X Bonner-Vogel Salts (Methods 1n Enzymology, XVIIA:5): Glass Distilled Water 670 ml MgS04*7H20 5 g 10 Citric Acld'HgO 50 g K2HP04 ' 250 g NaNH4HP04a4H20 87.5 g Glass 01 stilled Water to 1000 ml After overnight incubation, the cells were harvested by 15 centrifugation at 3000g for 5 minutes and resuspended in 5 ml of the same medium but without tryptophan. The cells were then subcultured in 20 ml of the medium without tryptophan to an Aggg of 0.05-0.07. This culture was grown at 37°C and 250 rpm to an Aggg of 0.05. The cells were harvested from 10 ml of culture by centrifugation as above 20 and resuspended In 1 ml of sonlcatlon buffer (15% sucrose, 50 mM Tris pH 7, 40 mM EDTA). The samples were sonicated for 3 minutes (on ■pulse) In a cup sonlcator (Sonlfier Cell Disrupter #350, Branson Sonic Power Co.). The cell lysates were centrifuged for 5 minutes in an Eppendorf Microfuge and the clear supernatants were removed for fur-25 ther analysis. The clear lysates were electrophoresed on an polyacryl amide sodium dodecyl sulfate (SOS) gel and analyzed by Western analysis as described In Example 2C. A protein band of approximately 69,000 molecular weight, the size expected for an unglycosylated form of gl ucoaraylase, was detected 1n the MH70/pGC24 30 clear lysate but not 1n the MH70 lysate.

To further demonstrate that an active glucoamylase enzyme was produced in Z, coll. MH70/pGC24 and MH70 were streaked on 52 MacConkey Agar (Difco Co., Detroit, MI 48232) plates containing 1% maltose and incubated overnight at 37°C. The fermentation of maltose results in a pH change in the media that is indicated by a shift from a colorless to red color in the colonies; nonfermenting colonies 5 remain colorless. Since MH70 is malQ", its colonies were colorless. The expression of the k. awamori glucoamylase in MH70/pGC24 permitted the hydrolysis of maltose to glucose and the fermentation of the glucose resulted in red colonies. Therefore, an active glucoamylase 1s produced in E. coli. 10 While preferred embodiments of the present invention have been described herein, it will be understood that various changes and modifications may be made without departing from the spirit of the invention. For example, while the examples all demonstrate autonomous replication in the host, using integrative transformation of the host 15 as described in References lc and Id where the gene and promoter are integrated into the chromosome is also possible.