WO1986000923A1

WO1986000923A1 - Yeast cloning vehicles

Info

Publication number: WO1986000923A1
Application number: PCT/US1985/001470
Authority: WO
Inventors: Anton K. Beck; Gregory P. Thill; Edward G. Bernstine; Jeffrey F. Lemontt
Original assignee: Integrated Genetics, Inc.
Priority date: 1984-07-31
Filing date: 1985-07-31
Publication date: 1986-02-13
Also published as: JPS61503002A; EP0192679A1; DK141386A; DK141386D0; EP0192679A4

Abstract

A cloning vehicle for expressing in a yeast host a DNA sequence coding for a human fertility hormone selected from the alpha and beta subunits of hCG, LH, and FSH, which includes, in phase and in order of transcription: a yeast promoter sequence and a DNA sequence coding for the immature hormone subunit, including the hormone signal peptide.

Description

I

YEAST CLONING VEHICLES Background of the Invention This invention relates to cloning vehicles for production of proteins. (The term "proteins" is used in this application to include peptides of indefinite size.) There has been increasing interest in the use of eukaryotic cells, such as yeasts, as hosts for the expression of genes or complementary DNA (cDNA) coding sequences for specific proteins. Typically, proteins exported from eukaryotic cells are processed in a secretory pathway that involves synthesis of a precursor protein (containing an amino-terminal "signal" peptide region), translocation across endoplasmic reticulum membranes, followed by specific signal peptide cleavage and further processing including carbohydrate additions (glycosylation) , and finally secretion of the mature product out of the cell. The yeast Saccharomyces cerevisiae is known to have such a secretory pathway [see Schek an and Novick, "The secretory process and yeast cell-surface assembly", in The Molecular 3iology of the Yeast Saccharomyces; Metabolism and Gene

Expression, Strathern et al. (Eds.), Cold Spring Harbor Laboratory, 1982]. Moreover, unlike an analogous secretory pathway that exists in some prokaryotic species, the yeast secretion mechanism provides components capable of glycosylating proteins. Hitzeman et al. (1983) Science 219;620-625 report expression and secretion in yeast of human interferon based on a coding region containing the inteferon signal sequence. The mature secreted protein reported by Hitzeman et al. demonstrated inaccurate cleavage of the signal peptide. Hinnen et al., at an International Congress of Microbiology (Boston, Massachusetts 1982) reported the fusion of a DNA fragment from the PH05 gene of the yeast Saccharomyces cerevisiae (containing the promoter and 80% of the 5' end of the PH05 signal sequence) to a cDNA fragment coding for the mature protein sequence and a portion of the 3' e'nd of the signal sequence of human alpha interferon. PH05 codes for the major repressible (by phosphate) acid phosphatase, a secreted enzyme in this yeast. This construction lacks the DNA coding sequence for the last 3 amino acids of the PH05 signal peptide and instead the last three amino acids of the hybrid signal are those coded for by the human cDNA sequence—i.e., the amino acids of the pre-interferon signal information. The plasmid constructed by Hinnen et al. was used to transform cells of Saccharomyces cerevisiae. The transformants were reported to express interferon activity under phosphate regulation. Localization of the interferon activity (intracellular versus extracellular) was not discussed.

Summary of the Invention In general, the invention features cloning vehicles capable of effecting the expression in a yeast host of the DNA sequences coding for a' uman fertility hormone selected from the and 3 subunits.of human chorionic- gonadotropin (hCG) , luteinizing hormone (LH) , and follicle stimulating hormone (FSH) , the cloning vehicles including, in phase and in order of transcription, a yeast promoter sequence and a DNA sequence encoding the immature hormone subunit, including the signal peptide. The most preferred hormones are α-hCG and β-hCG.

In some preferred embodiments, the promoter is substantially identical to the promoter sequence of a yeast PH05 gene, which preferably is within a 0.55 kb sequence upstream from the translational start site of the PH05 coding sequence contained within the 8 kb EcoRI PH05 genomic DNA fragment from chromosome 2 of Saccharomyces cerevisiae. In other preferred embodiments, the promoter is substantially identical to the promoter sequence of a yeast GAPD gene, preferably within a 0.6 kb sequence of the 2,1 kb Hindlll GAPD genomic DNA fragment of Saccharomyces cerevisiae. The GAPD gene encodes one species of glyceraldehyde-3'-phosphate alhydrogenase, a cytoplas iσ (i.e., nonsecreted) enzyme in Syccharomyces cerevisiae.

Description of the Preferred Embodiments We first briefly describe the drawings. Drawings

Fig. 1 is a diagram representing the PH05 expression vector.

Fig. 2 is a diagram representing the GAPD expression vector. Fig. 3 is a diagram representing the PH05- α

(A) and PH05- 3 (B) cloning vectors.

Fig. 4 is a diagram representing the GAPD- α (A) and GAPD- 3 (B) cloning vectors. PH05 Expression Vector Structure and Construction PH05 expression vector pl25N (Fig. 1) includes the following DNA segments (counterclockwise from the EcoRI site) :

1. The functional TRP1 gene from yeast contained within a 1.45 kb EcoRI genomic DNA fragment from chromosome 4, described in Kingsman et al. (1979) Gene 1_, 141-152 and in Tschumper et al. (1980) Gene 10, 157-166. This fragment contains a 103 bp functional TRP1 promoter region, a 672 bp coding sequence, and a 678 bp 3' untranslated region, which functions not only as a transcription termination sequence but also as a weak DNA replication origin (or replicon) called the arsl sequence; plasmid YRp7 has been described by Tschumper et al., iά_{^ r} (1980) and consists of this 1.45 kb EcoRI fragment inserted into pBR322 at the EcoRI site in an orientation.such that TRP1 and the gene for ampicillin resistance are transcribed in the same direction; 2. The pBR322 sequence, EcoRI-NruI, 3389 bp, carrying the gene for ampicillin resistance in E.coli, as well as a DNA replication origin that functions in E.coli;

3. ^! The 1.45 kb Hindi fragment from the B form of the 2 micron circle, a naturally occurring plasmid endogenous to most strains of Saccharomyces cerevisiae, described in Broach, The Molecular Biology of the Yeast Saccaromyces; Life Cycle and Inher tance, Cold Spring Harbor Laboratory, Cold Spring Harbor, 1981. The Hindi fragment contains a "strong" origin of DNA replication, (i.e., "stronger" than arsl because it confers a higher plasmid copy number in yeast) and confers a much lower rate of plasmid loss per mitotic cell division. This is also partly due to the presence of endogenous 2-micron plasmids carried in most strains of yeast. The orientation of the vector fragment in plasmid pl25N is not important for this function;

4. The pBR322 597 bp^' sequence from NruI-BamHI; and 5. The 0.55 kb PH05 promoter sequence.

The PH05 expression vector pl25N is constructed using the conventional recombinant DNA procedures described in Maniatis et al. Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, 1982. Plasmid YRp7 is partially digested with EcoRI such that only one of the two EcoRI sites is cleaved. Most of fhe product of this digestion consists of 5.8 kb linearized YRp7 molecules. Approximately half of these linear molecules are cleaved at one EcoRI site while the other half are cleaved at the other EcoRI site.. This mixture of linear molecules is separated from uncleaved circular molecules and from shorter linear molecules, arising by occasional cleavage of both EcoRI sites, by gel electrophoresis, as described in Maniatis et al. (1982) , followed by elution from the gel. The 5' protruding EcoRI ends are then filled in with dNTP's using DNA polymerase I (Klenow enzyme). (Alternatively, the 5' EcoRI protruding ends can first be removed with a single-strand-specific 5' to 3' exonuclease.) The flush-ended molecules thus generated are rejoined (circularized) with DNA ligase, which results in the loss of one of the EcoRI recognition sequences. The plasmid of interest, YRp7', which retains the EcoRI site adjacent the TRP1 promoter, is identified by restriction mapping. Next, the 1.45 kb Hindi fragment from the yeast 2 micron cirjcle is ligated into the Nrul site of YRp7'. The plasmid that is generated, YRp7'N, is then digested with both EcoRI and BamHI. The resulting large fragment, 6.87 Eb, which is isolated by gel electrophorsis, is used to make pl25N, as described below.

The other major component of pl25N is the PH05 promoter. The PH05 gene of Saccharomyces cerevisiae codes for the major phosphate-repressible acid phosphatase enzyme (APase), corresponding to the peptide called p60, and is contained within an 8 kb EcoRI genomic DNA fragment from chromosome 2 described in Bostian et al. (1980) PNAS 77, 4504-4508 and in Kramer and Andersen (1980) PNAS 11_, 6541-6545. The PH05 promoter lies within the approximately 0.55 kb sequence that preceeds the ATG translational start triplet at the beginning sequence of the PH05 coding sequence and is excised and isolated from the PH05 gene as a DNA fragment using Kpnl endonuclease. (These Kpnl sites occur 1400 bp upstream (i.e. on the 5¹ side) of the ATG triplet and 51 bp downstream (i.e. on the 3' side) of the ATG triplet. There is a BamHI site 550 bp upstream of the ATG triplet.) Then, this Kpnl fragment is digested with Bal-31 exonuclease, such that approximately fifty to ninety base pairs from either end of the DNA fragment are cleaved. Digestion with Bal-31 exonuclease generates a series of fragments of varying lengths. These fragments are combined with DNA polymerase I (Klenow enzyme) to assure that all the molecules are blunt-ended, after which EcoRI linkers are added to the blunt-ended fragments and ligated to both ends with T4 DNA ligase. Subsequent reaction with BamHI and EcoRI produces a "sticky-ended" DNA ^"fragment having cohesive termini which can be incorporated into a plasmid vector having BamHI or EcoRI insertion sites. Fragments approximately 0.5 kb, containing the PH05 promoter region, are identified and isolated using gel electrophoresis.

After the PH05 promoter region has been isolated, it is combined with the 6.87 kb vector fragment, above, into a circular plasmid using DNA ligase. Two such plasmids have been made, each employing a PH05 promoter containing fragment of a different length. Plasmid pl25N contains the longer of the two fragments and includes, downstream from the PH05 promoter, the major PHOS start site. Plasmid plllN contains a shorter fragment which does not include the major PH05 -start site. pl25N and plllN were used to transform E.coli and transformants selected on the basis of ampicillin resistance. Plasmids pl25N and plllN, isolated from transformants, have been deposited in the NRRL, Peoria, IL, and given NRRL number B-15805 and B-15804, respectively.

GAPD Expression Vector Structure and Construction Referring to Fig. 2, plasmid pBGAP is an expression vector containing the promoter from the major gene encoding glyceraldehyde-3'-phosphate dehydrogenase (GAPD) in S__j_ cerevisiae. Plasmid pBGAP is similar in structure to pl25N, described above, with the GAPD promoter being substituted for the PH05 promoter; a somewhat diffferent method, however, was used to construct the two plasmids. The GAPD promoter^" is contained within a 2.1 kb Hindlll genomic DNA fragment from S_-_ cerevisiae, which includes the DNA coding sequence for the GAPD gene, as well as the 5' and 3' flanking regions. To isolate the promoter region, the 2.1 kb Hindlll fragment is completely digested with Hpal endonuclease to* ield a 1.4 kb and a 0.7 kb fragment. The 1.4 kb fragment is then isolated and digested completely with TaqI; from these digests a 650-bp fragment containing the GAPD promoter region is isolated. The ends of this fragment are filled in using DNA polymerase I (Klenow enzyme) and dNTP's to form a blunt-ended fragment, after which EcoRI linkers are added to form a fragment having cohesive ends. Subsequent reaction with EcoRI produces a DNA fragment having cohesive ends, which can be iserted into the EcoRI site q plasmid pBR322. (Orientation is not important.) This plasmid is then partially digested with EcoRI _r such that only one of the two EcoRI sites is cleaved. After separation and elution of the resulting linear molecules from an agarose gel, the 5' protruding EcoRI ends are filled in with dNTP's using DNA polymerase I (Klenow enzyme). Then, BamHI linkers are ligated to both ends with T4 DNA ligase. Subsequent reaction with BamHI and EcoRI produces a 650-bp DNA fragment, which can be isolated from a gel and then inserted into the previously described 6.87 kb vector fragment containing BamHI and EcoRI insertion sites, and ligated to form a circular plasmid. Two kinds of BamHI-EcoRI-ended promoter fragments can be isolated in this way: One with the EcoRI linker at the 5' end of the promoter and the BamHI linker at the 3' end, and another having a 5' BamHI end and a 3' EcoRI end. The latter type was used to construct the GAPD expression vector plasmid pBGAP, which has been deposited with the NRRL and has been given accession number B-15806. PH05- hCG and PH05- 3 hCG- Cloning Vectors

Plasmid pl25N can be used for the production and secretion of the α and 3 subunits of hCG in Saccharomyces cerevisiae; transcription is under the control of the promoter sequence of PH05, and the signal sequence is that of immature α or 3 hCG. cDNA clones coding for and 3 -hCG, including the signal peptide sequence at the amino-terminal end, are obtained as follows. First, RNA is extracted from placental tissue by the following method: Homogenization of the tissue is carried out in a 1:1 mixture of phenol:lOOmM Na-acetate (pH 5.5) containing ImM EDTA, that has been warmed to 60° for 20 min. After cooling on ice for 10 in., the phases are separated by centrifugation. The hot phenol extraction is repeated twice more -followed by two extractions with chloroform. RNA is precipitated from the final aqueous phase by the addition of 2.5 volumes of ethanol.

In order to enrich for poly A+ mRNA, placental RNA is passed over oligo (dT)-cellulose in 0.5M NaCl buffered with lOmM Tris-HCl, pH 7.5, and washed with the same solution. Poly A+ mRNA is eluted with lOmM

Tris-HCl (pH 7.5), ImM EDTA, 0.05% SDS and precipitated twice with ethanol. Typical initial yields are 1.5-2.0 mg of total RNA per g of tissue, of which about 2% is poly A+ mRNA. Placental cDNA libraries are constructed by reverse transcription of placental mRNA, second strand synthesis, using E^ coli DNA polymerase I (large fragment), treatment with SI nuclease, and homopolymer tailing (dC) with terminal deoxynucleotidyl transferase; all such procedures are by conventional techniques.

In a typical preparation, 20-30% conversion of mRNA to single strand • (ss) -cDNA; 70% resistance to digestion with nuclease SI after second strand synthesis; and dC "tails" of ten to twenty-five bases in length, are obtained. These cDNA molecules are then annealed to DNA fragments of the plasmid pBR322 which has been digested with PstI, and to which dG "tails" have been added. These recombinant plasmids are then used to transform E^_ coli cells to generate a cDNA library (transformed cells are selected on the basis of tetracycline resistance).

In order to identify the human alpha hCG clone, a 219 bp fragment of a mouse alpha thyroid stimulating hormone (TSH) clone is used as a hybridization probe. This probe has 77% sequence homology with the human clone. It is radioaσtively labeled by nick translation and hybridized to the cDNA library under conditions that take into account the extent of ho ology. Strongly hybridizing clones are analyzed by restriction .mapping and clones containing the complete coding sequence of alpha.hCG are verified by DNA sequencing.

A 579 bp cDNA fragment encoding beta hCG is described in Fiddes et al. (1980) Nature, Vol. 286, pp. 684-687.

To insert either the α or 3 -hCG cDNA fragment into the pl25N or plllN PH05 expression vector, EcoRI linkers are first ligated to both ends of the cDNA fragments and then cut with EcoRI to generate cohesive ends. The modified cDNA fragment is then inserted into the PH05 expression vector by cleaving the expression vector with EcoRI, combining the α or -hCG cDNA with the now-linear expression vector sequence, and ligating these DNA fragments together such that the hCG cDNA is inserted in a counterclockwise transcriptional direction adjacent to and downstream-from the PH05 promoter sequence, as shown in Figs. 3A and 3B, respectively. Although mRNA transcripts initiated by the PH05 promoter will likely extend and terminate well into the TRPI region, translation will preferentially be initiated at the first AUG codon, i.e., in the α -hCG or 3 -hCG cDNA. hCG Production

The resulting expression vectors are used to transform host yeast cells using standard techniques, e.g. that described by Beggs, Nature 275: 104-109 (1978). The transformed yeast cells are cultured in a standard culture medium using standard techniques, e.g.. that summarized by Botstein and Davis, "Principles and Practice of Recombinant'DNA Research with Yeast", _in The Molecular Biology of the Yeast Saccharomyces; Metabolism -and Gene. Expression, pp. 607-636, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY, 1982.

Selection of transformants is then carried out as follows. Any one of the above-described PH05 promoter-containing expression vectors is used to transform a strain of Saccharomyces cerevisiae carrying a mutant trpl gene which causes tryptophan auxotrophy. Cloning vectors PH05- α and PH05- 3 each carries an expressible yeast TRP1 gene allowing the transformed yeast (Trp+) to grow in the absence of tryptophan (Trp prototrophy) . Trp transformants are isolated by single-colony isolation on synthetic growth medium lacking tryptophan. This medium, SC-TRP, is described in Table 1, below.

Nutrient Amount per liter of medium

SC-TRP P₃₀(SC-TSP)

Dextrose 20 g 20 g

Ammonium sulfate 5 g 5 g Magnesium sulfate 0.5 g' 0.5 g Sodium chloride 0.1 g 0.1 g Calcium chloride 0.1 g 0.1 g Biotin, Folic acid 2 μg each 2 pg each

Calcium pantσthenate Nicotinic acid Pyridoxine HC1 400 pg each 4p0 pg each Thia ine HC1 Manganese sulfate Zinc sulfate^•

P-aminόbenzoic acid Riboflavin Ferric chloride 200 pg each 200 pg each Sodium molybdate

Boric acid 500 μg 500 pg Copper sulfate 40 pg 40 pg Potassium iodide 100 pg 100 pg Inositol 2 mg 2 mg Threonine 150 mg 15θ^"mg

Adenine, arginine 20 mg each 20 mg each histidine, methionάne, uracil

Leucine, Lysine 60 mg eacrT 60 mg each

KH₂P0₄ 1.5 g 30 mg KC1 0 1.5 g Transformants are incubated in LP_3Q (SC-TRP) medium, a low-phosphate medium (Table 1) . Cells retaining cloning vector —PH—0—5.- or —PH—05— 3 can grow in this medium, and begin to synthesize α or 3 -hCG when 5 the intracellular level of inorganic phosphate begins to decrease. After two days of incubation at 30°C, virtually all of the antigenically active α -hCG is found in the culture medium. Typical levels are 0.2 mg/1 or greater, as determined by radioimmunoassay.

10 Antigenically active 3 -hCG is found in both the culture medium and in the cells.

To isolate the αor 3 -hCG secreted into the culture medium, the yeast cells are at first removed from the harvested cell suspension by any of several

15 convenient means such as centrifugation or filtration. Then, the cell-free fermentation broth is subjected to conventional protein purification procedures designed to isolate pure α or 3 -hCG. To recover the intracellular products, the cells are suspended in a convenient buffer

^•20 solution and then broken by shaking the suspensions with 0.45 mm galss beads in a Braun Homogeriizer. Crude extracts are prepared by low-speed centrifugation to remove cell debris. GAPD- g hCG and GAPD- 3 hCG Cloning Vectors

25 The structure of the GAPD- α hCG and GAPD- 3 hCG cloning vectors is shown in Figs. 4A and 4B, respectively. These vectors are identical to the PH05- α hCG and PH05- 3 hCG vectors derived from pl25N, with the GAPD promoter replacing the PH05 promoter.

30 They are constructed using the same methods as described above for the PH05- α hCG and PH05- 3 hCG vectors. Protein Synthesis The production of and 3 -hCG using the GAPD expression vector is as' described above, except that the high phosphate SC-TRP medium of Table 1 is used as the culture medium. Other embodiments are within the following claims. For example, other human fertility hormones, e.g., the α and 3 subunits of LH and FSH, can also be produced.

Claims

Cla ims

1. A cloning vector capable of effecting the expression in a yeast host of a DNA sequence coding for a human fertility hormone selected from the α and 3 subunits of hCG, LH^J, and FSH, said cloning vehicle comprising, in phase and in order of transcription: a yeast promoter sequence and a DNA sequence coding for the immature hormone subunit, including the hormone signal peptide.

2. The hormone of claim 1, consisting of -hCG.

3. The hormone of claim 1, consisting of 8 -hCG.

4. The cloning vehicle of claim 1 wherein said yeast promoter sequence is substantially identical to the promoter sequence of a -yeast PH05 gene.

5. The cloning vehicle of claim 4 wherein said yeast promoter sequence is substantially, identical to a fragment within a 0.55 kb sequence upstream from the translational start site of the PH05 coding sequence contained within the 8 kb EcoRI PH05 genomic DNA fragment from chromosome 2 of Saccharomyces cerevisiae.

6. The cloning vehicle of claim 1 wherein said yeast promoter sequence is substantially identical to the promoter sequence of a yeast GAPD gene.

7. The cloning vehicle of claim 6 wherein said yeast promoter sequence 'is substantially identical to a fragment within a 0.6 kb sequence of the 2.1 kb Hindlll GAPD genomic DNA fragment of Saccharomyces cerevisiae.

8. A cloning vector having NRRL accession number B-15705.

9. A cloning vector having NRRL accesstion number B-15884.

10. A cloning vector having NRRL accession number B-15806.