CA2048676C - Artificial promoter for the expression of proteins in yeast - Google Patents

Abstract

The invention relates to an artificial promoter for the expression of proteins in yeast, which comprises:
- a sub-sequence upstream from the TATA
component of the sequence of the promoter of the GAL7 gene of Saccharomyces cerevisiae, which comprises the upstream activation sequences UAS1 and UAS2; and - a sub-sequence of the sequence of an ADH2 promoter comprising the TATA component and the transcription initiation region.
Application: Preparation of proteins, especially urate oxidase.

Description

20~86'~6 Artificial promoter for the expression of proteins in east The present invention relates to a novel arti-ficial promoter for the expression of proteins, in par-05 ticular heterologous proteins, in yeast, to a vector for the expression of said proteins which carries said promoter, to the strains of yeast, and especially of Saccharomyces cerevisiae, which are transformed by this expression vector, and to a method of producing a re-combinant protein with the aid of these strains.
Yeast, and in particular Saccharomyces cere-visiae, a non-pathogenic microorganism whose genetics has been studied in detail, is a preferred eukaryotic host for the expression of proteins, especially hetero-logous proteins. It is therefore important to discover or construct novel promoters for the expression of said proteins which are more advantageous than the known promoters.
The structure of a yeast promoter, which is a DNA sequence located upstream from a gene and res ponsible for the transcription of said gene, is begin ning to be partially known and understood. Said promoter is known to comprise a TATA component located in an AT-rich zone, a transcription initiation region downstream from said component and, if appropriate, up-stream from said component, sequences, called upstream activation sequences (UAS) or upstream repression se-quences (URS), which regulate the strength of the promoter under the effect of an inducer or a repressor.
The Applicant constructed a novel hybrid pro-moter from two known promoters: the promoter of the GAL7 gene of Saccharomyces cerevisiae (TAJIMA et al., 1986, Molecular Cellular Biology, 6, 246-256) and a promoter with a sequence similar to that of the natural ADH~ promoter (5'-flanking region of the ADH2 gene, ~~~6'~6 The invention therefore relates to a novel artificial promoter for the expression of proteins in yeast, which comprises:
- a sub-sequence upstream from the TATA component of the sequence of the promoter of the GAL7 gene of Sac-charomyces cerevisiae, which comprises the upstream activation sequences UAS1 and UAS2; and.
- a sub-sequence of the sequence of an ADH2 promoter comprising the TATA component and the transcription initiation region.
Preferably, the sub-sequence upstream from the TATA component of the promoter of the GAL7 gene of Saccharomyces cerevisiae is the following sequence or a sub-sequence thereof.:
described by RUSSEL et al. (1983); J. Biol. Chem. 258, 2674-2682), which is called an ADH2 promoter in the M
L
a CGCGTCTAT CTTCGGAGCACTGTTGAGCGAAG CTCATTAGATATATTTTCTGTCAT
__+______ __+_________+_________+ ________+_________+_____ AGATA GAAGCCTCGTGACAACTCGCTTC GAGTAATCTATATAAAAGACAGTA

TTTCCTTAACCCAAAAATAAGGGAGAGGGTCCAAAAAGC CTCGGACAACTGTTGACCGT
____+_________+_________+_________+___ ____+_________+_____ AAAGGAATTGGGTTTTTATTCCCTCTCCCAGGTTTTTCG GAGCCTGTTGACAACTGGCA
G TCCGAAGGACTGGCTATACAGTGTTCACAAAATAGCCAAGCTGAAAATAATGTGTAGC
1 0 _ _+_________+_________+_________+_________+_________+_____ C GGCTTCCTGACCGATATGTCACAAGTGTTTTATCGGTTCGACTTTTATTACACATCG
S
P
h CTTTAGCTATGTTCAGTTAGTTTGGCATG
____+_________+_________+____ GAAATCGATACAAGTCAATCAAACC
The sub-sequence of the sequence of an ADH2 promoter comprising the TATA component and the trans 20 cription initiation region is preferably selected from the following sequence or a.sub-sequence thereof .
CCIATCACATATAAATAGA
.,_.._.. , .,....._ GTGCCAG1AGCGACTTTTITCaCACTCGAGATACTC1TACTACTGC1CTCTTGTTGTTTT
...__.............,......___,__.~____.,_.......:,...._.
CACGGTCATCGCTGAAAAAAGTGIGAGCTCTATGAGAATGATGACGAGAGAACAACAAAA
TATCACTTCTTGTTTCT1C'1TGGTAAATAGAATATCAAGCTACAAAAAGCATACAATCAA
_._,.. _.....,.........,.........,_ ___--_~~........,._-.__.
30 ATAGTGRAGAACAAAGAAGAACCATTTAICT1ATAGTTCGpTGTTT1TCGTATGT1AGTT
C

a I
CTATCAACTATTAACTA1A:~
35 __.,...._._..~....

20~486'~6 A particularly advantageous promoter is that which comprises the following sequence:

a I
CGCG1C1R1aC11CGGAGCaCTGIfGAGCCrAAGGC1CATTAGaIATAtIITCTGTCAT
..,_..._._._....._..._,...___...,.._......,_..__._..,_..__ AGatATGAAGCCTCGTGACAACTCGCTTCCGaGTAAICTATATAAAAGACaGTA
T1TCCTTRACCCAAAAATAAGGGAGAGGG1CCAAAaAGCGCTCGGACAAC1C11TGACCOT
_.._,....___._,... .,... .,._ .,. _ AAAGGAA11GGGT1ITTA1TCCC1C1CCCAGG111T1CGCGAGCCtGTTGACAACTGGCti GATCCGAAGGACTGGCTATACAG1G1TCACAAAATRGCCAAGCTGRAAATAATGIGIAdC
_._.,...._._ .,__.__..._,._...._..,......___,.__.____.,.._..
CTAGGCTTCC1GACCGaTAIGTCACAAGTG1T1TATCGGTTCGAC111TATTACRCATCCi S
n I -CTTTRGCTATG11CAG11RG1iTGGCATGCCTATCACAIAtAAAIAGA
_...,_.__...._,.._......,...._.,_.._._ ...___...

GTGCCAGTRGCGRCTTI111CACiiCTCGaGATRCTCT7ACTRC1GCTCTCTTGTTGIT1T
.~. ..... ...~ _,..... ....... _,...
CpCGGTCATCGCTGaAAAaaGTGTGAGC1CIATGAGaATGAIGACGAGAGAaCAACaAAA

_... .,.. ..~...._........_.__.,.._....._,._..._ EiTAGTGARGAACaAAGAAGARCCRITTATC1TATAG1ICGGiIGTITTTCGTATGTTi~GTT
C

CIRTCAAG1A1TRACTATAtf _...._._.__......, The promoter of the invention may be obtained by the conven-3 0 tional recombinant DNA techniques wellknawn to anyone skilled in the art.
The promoter of the invention has important advantages over the known promoters and in particular over the ADH2 promoter and the promoter of the GAL7 gene.
It permits a high maximum level of transcrip-~0~~~~6 tion -and hence of expression, in particular for A.
flavus urate oxidase, and offers the possibility of regulating said expression at three levels:
- zero level in the presence of glucose and in the 05 absence of galactose: no expression is detected, which is an identical result to that published for the promoter of the GAL7 gene in a strain of Sac charomyces cerevisiae growing under these conditions (TAJIMA _et al., 1986, Molecular Cellular Biology, 6, 246-256). The ADH2 promoter shows a low but detec-table level of expression under these conditions.
- basic level in the absence of glucose and galactose:
there is weak expression, which is an intermediate result between that observed for the ADH2 promoter (maximum level) and that published for the promoter of the GAL7 gene (zero level: TAJIMA et al., 1986, Molecular Cellular Biology, 6, 246-256).
- maximum level of expression in the absence of glucose and in the presence of galactose.
The advantage, for the expression of heterolo-gous proteins in yeast, of having a promoter which shows a zero level under certain conditions is that it affords the possibility of avoiding any selection pressure which would favor the least productive cells during the propagation of the strain. This is particu-larly important in the case where the protein is toxic to the host cell.
The advantage of a promoter with two levels of expression: the one a basic level (non-induced) and the other a maximum level (induced), lies in the ability to choose an intermediate level by varying the concentra-tion of the inducer.
The invention further relates to an expression vector for yeast which carries a gene of interest with the means necessary for its expression, its replication - 6 - ~04~6'~6 and the selection of the transformed cells, wherein the gene of interest is under the control of the promoter def fined above .
This gene of interest can be an endogenous gene 05 of yeast or a eukaryotic or prokaryotic exogenous gene.
Of particular value as eukaryotic exogenous genes are a recombinant gene coding for Asperqillus flavu urate oxidase, a recombinant gene coding for a human cyto kinin and a recombinant gene coding for hirudin.
to zn the case where the protein coded for by the exogenous gene is secreted naturally, the sequence coding for this protein is preferably preceded by a signal sequence. The function of this signal sequence, which is chosen according to the host cell, is to 15 permit export of the recombinant protein out of the cytoplasm, enabling the recombinant protein to adopt a configuration similar to that of the natural protein and considerably facilitating its purification. This signal sequence can be cleaved either in a single step 2o by a signal peptidase which releases the mature pro-tein, the eliminated sequence usually being called a pre sequence or signal peptide, or in several steps when this signal sequence comprises, in addition to the sequence eliminated by the signal peptidase, called a 25 pre sequence, a sequence eliminated later in the course of one or more proteolytic events, called a pro sequence.
The invention further relates to the strains of yeast, in particular of Sack aromyces cerevisiae, which 30 are transformed by the above expression vector, and to a method of producing a pzotein of interest, which com prises the culture of said strains in the presence of galactose.
In particular, the invention relates to the strains 35 of Saccharomyces cerevisiae which have been deposited in the depository authority named Collection Nationale de Culture de Microorganismes -Intitut Pasteur - France under the following numbers .

- 6 b i s - ~~~ti'~'6 I - 919 on December 28, 1989 I -1021 on December , 1990 I - 1022 on December , 1990 I - 1023 on December , 1990 05 The invention is illustrated, without impl;~~:,~ a limitation, by means of the Examples below .
Many of the following tech.-:igues, which are well known to those.skilled in the art, are described in detail in the work by Maniatis et al.: "Molecular cloning: a laboratory manual" published in 1984 by Cold Spring Harbor Press in New York.
The synthesis of the oligonucleotides is ca~riec' out by means of a DNA Synthetize Biosearch 4600.
ExAMPLE 1: nPrpYM~nation of the seguenrP ~r the coNA o' A flavus urate ox 1 ) Isolation of the messe~gPr RNA's 'ron As~er~'_' ~~s flavu~
The strain of A flavus which produces urate oxidase was cultivated under conditions appropriate for the production of urate oxidase, i.a. in a medium con taining uric acid and having the following composition:
glucose 15 g/1, higS04.7H'O 1 g/1, KH~P04 0.75 g/1, CaCO~ 1.2 g/l, uric acid 1.2 g/1, KOH 0.5 g/1, soy bean oil 0.66 ml/1, FeS04.7H~0 10 mg/1, CuSO4.5H~0 1 mg/1, ZnS04.7H~0 3 mg/1, rlnSO4.Ha0 1 mg/1. The medium is adjusted to pH 7 with H~S04 1 M and sterilized at 120'C
for 80 min.
In a 5 1 Erlenmeyer flask, 1.5 1 of medium are inoculated with about 1 to 3.10' spores. ' The culture is incubated for about 40 h at 30~C, with agitation (120 rpm). The mycelium is re-covered by filtration on gauze, washed with water and frozen in liquid nitrogen.
15 g of mycelium (wet weight) are thawed, re suspended in 45 ml of lysis buffer and then taken up in the same volume of beads (0.45 ~m in diameter). The lysis buffer consists of guanidine thiocyanate 4 M, _ 7 -20~6'~~;
Tris-HC1 10 mM pH 7.6, EDTA 10 mM, l3-mercaptoethanol 50 ml/1. The mycelian suspension is ground in a Zell-muhler mill (vibrogenic) for 5 min.
The ground material is recovered and the beads 05 are decanted. The supernatant is removed (about 45 ml), brought back to a final concentration of 3 t~i in respect of lithium chloride and stored at 0°C.
After two days, it is centrifuged for 60 min at 10,000 rpm. The supernatant is discarded and the re sidue is taken up in 40 ml of LiCl 3 M and centrifuged again at 10,000 rpm for 1 h 30 min.
The following are added: proteinase K (SIGi~iA) 40 ug/ml, SDS (0.1% w/v) and EDTA 20 mM. The mixture is incubated at 37°C for 3 h. Precipitation with 2 volumes of ethanol is followed by washing with 70%
ethanol. The residue is taken up in 0.5 ml of TE buf fer (Tris-HC1 10 mM, EDTA 1 mM pH 7.5), the mixture is extracted twice with chloroform and precipitation is carried out with ethanol. The RNA's are stored at -80°C in alcohol.
2) Purification of the poly A+ fraction of the RNA's About 1 mg of RNA is precipitated for 20 min at 4°C (15,000 rpm) and then washed with 70o ethanol and dried. The residue is taken up in 1 ml of TE buffer and resuspended by agitation in a Vortex. Oligo dT-cellulose type 3 (marketed by Collaborative Research Inc., Biomedicals Product Division) is prepared accor-ding to the manufacturer's recommendations. The RNA is deposited on the oligo dT, agitated gently to resuspend the beads and then heated for 1 min at 65°C.
The suspension is adjusted to 0.5 M NaCl and then agitated gently for 10 min. It is then centri-fuged for 1 min at 1000 rpm, the supernatant is removed and the residue is washed twice with 1 ml of TE buf f er containing 0.5 M NaCl. The supernatants are removed.

- g -The polyadenylated fraction of the RNA's (consisting of the messenger RNA's) is eluted by suspending the beads in 1 ml of TE buffer, then heating this suspension at 60°C for 1 min and subsequently agitating it for l0 min 05 on a tilting plate. It is then centrifuged for 1 min at 1000 rpm, which makes it possible to recover on the one hand the supernatant containing free mRNA's in solution, and on the other hand the residue of cellu-lose beads. The above series of operations (starting from elution) is repeated. The supernatants obtained in this way are pooled, the excess beads are removed by centrifugation and the supernatant is precipitated with ethanol containing NaCl in accordance with the usual techniques (Maniatis: op. cit.).
3) Building of the cDNA librarv The messenger RNA's isolated as described in the previous section were used to build a cDNA library in vector pTZl9R (marketed by PHARMACIA). This vector is a plasmid comprising a polylinker containing unique restriction sites.
The cloning technique used is the one described by Caput et al. (primer-adapter technique: Caput et al., Proc. Natl. Acad. Sci. (U.S.A.) (1986) 83, 1670-1674).
It consists firstly in digesting the vector with Pstl, adding a polydC tail to the protuberant 3' end and then digesting the resulting plasmids with BamHI. The fragment corresponding to the vector is purified on a column of Sepharose* CL4B (Pharmacia). It therefore comprises a polydC tail at one end, the other end being a sticky end of the BamHI type. Secondly, the messenger RNA's are subjected to reverse trans-cription starting from a primer having the sequence 5'<GATCCGGGCCCT~12~)<3. Thus the cDNA's have at their 5' end the sequence GATCC complementary to the BamHI
* - Trade-mark A

_ g -sticky end.
The RNA-DNA hybrids obtained by the action of reverse transcriptase are subjected to alkaline hydro-lysis, enabling the RNA to be removed. The single-05 stranded cDNA's are then purified by 2 cycles on a column of Sepharose CL4B and subjected to a treatment with terminal transferase so as to add polydG's at the 3' end. The cDNA's are inserted in single-stranded form into the vector prepared as described above. A
second oligonucleotide, the adapter, complementary to the primer, is necessary in order to generate an "open"
BamHI site at the 5' end of the cDNA's. After hybridi-zation of the vector, the cDNA and the adapter, the recombinant molecules are circularized by the action of the ligase of phage T4. The single-stranded regions are then repaired by means of the DNA polymerase of phage T4.
The plasmid pool obtained in this way is used to transform the MC1061 strain for ampicillin resis tance (Casabadan, Chou and Cohen, J. Bact. (1980) 143, pages 971-980).
4) Determination of the partial sequence of urate oxidase An A . f lavus urate oxidase preparation ( SIGriA ) was repurified by chromatography on a column of Red agarose 120 (SIGMA), this being followed by filtration on Ultrogel Aca 44 (IBF), an acrylamide-agarose gel.
Direct amino-terminal sequencing of the protein was attempted in order to obtain information on the amino acid sequence of the purified urate oxidase, making it possible to synthesize the probes necessary for cloning the cDNA. This sequencing was not success-ful, probably because of amino-terminal blocking of the protein.
The following strategy was therefore developed to obtain the partial sequence of urate oxidase:
- cleavage of the protein with proteolytic enzymes (using the enzymes trypsin and protease V8 of staphylo-coccus aureus) 05 - separation of the resulting polypeptides by reversed phase HPLC
- sequencing of the purified peptides.
1) Hydrolysis of the urate oxidase with trypsin, purification and seauencinct of the peptides:
The urate oxidase, at a concentration of 9 mg/ml in an ammonium carbonate buffer 100 mM pH 8.9, was digested with trypsin (Worthington, TPCK), in a ratio urate oxidase/trypsin of 30/1 by weight, at 30'C
for 24 h. After tryptic hydrolysis, 60 ~cg of digested urate oxidase were directly injected on to a reversed phase HPLC column of Brownlee G18 grafted silica (column: 10 x 0.2 cm) equilibrated with acetonitrile 1%
(v/v) and trifluoroacetic acid 0.1% (v/v) in water.
The peptides were then eluted by a linear gradient of acetonitrile in a solution of trifluoroacetic acid ( 0. 1% v/v) in water, varying from 1% to 60 0 of aceto-nitrile in 60 min, at a rate of 150 ul/min. The pep-tides leaving the column were detected by measurement of the optical density at 218 nm.
The elution profile is shown in Figure 1, in which the numbers following the letter T (trypsin) cor-respond to the peaks identified.
Each peak was collected and stored at -20°C
until analyzed on a protein sequencer (model 470 A from Applied Biosystems) equipped with a chromatograph (model 430 A from Applied Biosystems), which con-tinuously analyzes the phenylthiohydantoic derivatives formed, after each degradation cycle.
Table (1) below shows the peptide sequences of the 9 peaks identified.

- 11 - ic:0~~6~~
2) Hvdroly_sis of the urate oxidase with protease V8, purification and sequencing of the peptides:
The urate oxidase, at a concentration of 2 mg/ml in an ammonium acetate buffer 100 mM pH 6.8, was 05 digested with the protease V8 of Staphylococcus aureus (Boehringer-Mannheim), in a ratio urate oxidase/
protease V8 of 60/1, at 30 ° C for 72 h. 160 ~Cg of di-gested urate oxidase were then injected on to a rever-sed phase HPLC column of Brownlee G18 grafted silica l0 (column: l0 x 0.2 cm: particles: 7 x 0.03 Vim), equili-brated with acetonitrile 1% and trifluoroacetic acid O.lo (v/v) in water. The peptides were then eluted by a linear gradient of acetonitrile in a solution of tri-fluoroacetic acid in water (O. to (v/v)), varying from 15 1 o to 60 0 of acetonitrile in 60 min, at a rate of 150 ~1/min. The peptides leaving the column were detected by measurement of the optical density at 218 nm.
The elution profile is shown in Figure 2, in which the numbers following the letter V (protease V8) 20 correspond to the peaks identified.
Each peak was collected and stored at -20°C
until analyzed on the protein sequencer already men-tioned.
Table (1) below shows the peptide sequences of 25 the 5 peaks identified.

i i I I I 1 i I I t ~ 1 N a~ a~ a s.. n. a c~
~ a a a ~

a .=. .. c I .-~ n ~ w I 1 I I I I ( 1 I I I

4 1r GL A ~ a. T I T
, ~

R a j V
V V
~

1 1 I I 1 1 ~ I I I

G N C~ T T d Ir ~ I Sr ~ i.r ~r T tr ...~ .-r ~ N ~
H

,_." v7 c~ a .s c~ V I N H

1 I I I , , I I I I I 1 I

N ~ a a~ a> a c .-. a~ a a N T ~ ~ ~ ~ r1 t/) a~ O
rp ~ N '_' T a E X a C~. C1. c7 r2 ~ .~ a. .-a w r. sa . a~ a a n> n. c.. 1... v..
s.. c T ~-1 .C ~ ~ 'U d ~ V7 'U T QJ ~~ d H H a. a a a. .c w H v~ c.7 cn I i a rr Lr rp Y..rT a a 1-n a C d I V 1r L

TJ ~I rp d --rU ~ .-~ ~-1 T ~ C7 .-r O O
~ d O V ~ t/Wt V~ V V V H V n4 n--n ~ V~
N ~/7 t I I I 1 1 I I 1 1 1 I I I I I
I

~ ~r G a d. a O O N r0 L1. ~ ~ N N
~ C I

.t.1rD I d N d ~ L, Lr .-v n ~ T t/7 T
~r .-v i o ~ V a ~x a H w o.. z .s .s H , ,~ a I .x I

N I I ~ 1 I I I 1 I I 1 , 1 I 1 I I
; I

_ .~
I U .-rC Q. ~1 D. O O O N 'U r0 C I ~ ~
~ 6' ~ ~ I
G ~O -i I N r0 N ~ Sr 1.a T ~.r r~ .-' -n w ~ c~ I .x ~ .x o. a. a. a c ~ .s .x ! .-.
.s .. .-r H

i I

~ I I ~ I 1 I I I I I I l I I i 1 n.
I ~ c s 1.. c L. I
N N ' n.
s.

O o a p. a m. a _ -. ~ N ~ T. T V7 Q1 V7 N Q7 .--~N .-a ~

I H ~ ft .~ H r- wt ~ V H E rC H ..7 J rC "C
v7 w I

w a I a 1 a lr ~ ~ O N lr ~ ~ d ~
N N N

. ~ .-1.-t.-1 ~ rp tp La T ~ .-r T T
~ T

v ~ v ~ t5 E-rc~ v> >. ~ a. a H ~ I .... .., H a .a c_ I I I I I I , 1 1 1 I I 1 I 1 1 U
C C 1 d Lr d C O O a a L L i~ a T
S.-n N b a O .--~-a ~ ~ Q7 .G N it ~ _ O a. O ~r d O ~ ~ it d ~r i a V N I G..V7 C7. ~ D. LY ~7 ~7 H V) V .-7 CW
r4 H H S rC -7 D' d 1 1 I I I 1 I I I I 1 I 1 I I I I I

Q7 p, lr a L 1-aC C 1~ 1-~ L a. N ~ N
T ~. ~ C ~

N ~ G7 T O 7~ t N N N ~ O N I T n0 T
.--~ W ~~ Ir rp 7 Gl.I H w H H nC rat N H N fG ~ .-~ .-7 ~2 V V H V 7 rtC 7 I t I I I I I 1 1 I i I 1 I I I I

C C ~ N d N 3-rC G it .~ L, N T L T
4J a 7 lr ~I a fr N N _ O ~ .-1 T r0 ~ ~ --r .C
O d d t0 O

,Q ,Q i Z ~, Z N V C5 H ~. H H V H .-~ V
~y, .-.7 .~ t/~ ~ H

~ "

N I N N N f'1M M ~ N 1 l~1 ~D
N ~

H H I C-~E H E~ F H > > >
E

N

r 9 r p N
C r0 d O

G1 ~

O
a.

~,., ~ G~.

w w _ O ~ O

- 13 - ~~~86'i 6 5) Screening of the bacteria 1) Preparation of the labeled probes:
Two pools of probes deduced from amino acid sequences of the protein were synthesized with the aid 05 of a Biosearch 4600 DNA synthesizer. The first pool corresponds to the sequence of residues His-Tyr-Phe-Glu-Ile-Asp (part of the sequence of T 27), i.e. from 5' to 3':
A T G G G
T C G A T T C A A T A T G
T C A A A
This pool in fact consists of 24 x 3 - 48 different oligonucleotides, representing all the possible combi nations.
The second pool corresponds to the sequence of amino acid residues Gln-Phe-Trp-Gly-Phe-Leu (part of the sequence of V5), i.e. from 5' to 3':
GG A G T
A AAGCCCCA AA TG
AA C A C
T
This pool consists of 24 x 4 = 64 combinations.
The probes are labeled with terminal deoxy-nucleotide transferase (TdT) (marketed by IBI, Inc.).
The reaction is carried out on 100 ng of a 3o mixture of oligonucleotides in solution (100 mg/ml) in "Cobalt" reaction buffer (supplied as a 10-fold concen-trate by IBI, Inc.): 1.4 M potassium cacodylate - pH
7.2, 300 mM dithiothreitol, 1 ~1 of the enzyme terminal deoxynucleotide transferase (IBI, Inc.) and 50 ~Ci of deoxycytidyl triphosphate, dCTP, labeled with P32.

The reaction is carried out at 37°C for 10 min and is then stopped by the addition of 1 ~,1 of EDTA
0.5 ri.
A phenol extraction is carried out and the 05 extract is dialyzed on a column of Biogel P10 poly-acrylamide (Biorad: 150-1050).
2) --Hybridization and detection of the colonies containinct,urate oxidase cDNA:
About 40,000 colonies are screened by the in situ hybridization technique developed by Grunstein and Hogness (1975, Proc. -Natl. Acad. Sci. (U.S.A.), 72, 3961). About 6000 bacteria are plated out in Petri dishes to give isolated colonies. After incubation for 24 h at 37°C, each dish is replicated on 2 filters, each filter being intended to be treated with one of the 2 pools of probes, so that all the colonies ob-tained are tested with the 2 pools of probes in paral-lel.
The filters are hybridized with one of the 2 pools of probes in a buffer containing 6 x SSC, 10 x Denhardt's solution and 100 ~g/ml of sonicated and de natured salmon sperm DNA (SIGMA). The hybridization is carried out at a temperature of 42°C for 16 h. The 6 x SSC solution is obtained by diluting a 20 x SSC solu tion. The preparation of the 20 x SSC buffer is des-cribed by Maniatis, Fritsch and Sambrook (op. cit.).
In summary, this buffer contains 175.3 g/1 of NaCl and 88.2 g/1 of sodium citrate and is adjusted to pH 7 with a few drops of NaOH 10 N. The 10 x Denhardt's solution contains 1 g of Ficoll*, lg of polyvinylpyrrolidone and 1 g of human serum albumin per 500 ml of final volume.
After washing in the 6 x SSC solution at 42 ° C
(3 h with 5 changes of bath), the filters are wiped with Joseph paper and subjected to autoradiography.
The filters are developed after 16 h. A fraction of * - Trade-mark - 15 - 20~86'~~
about 0.5% of the colonies was found to have hybridized with the 2 pools of probes.
colonies from this fraction were taken up and purified. The plasmid DNA was prepared from each of 05 these colonies and this DNA was analyzed by digestion with either BamHI, or HindIII, or both BamHI and HindIII.
After analysis on agarose gel, the 5 plasmids obtained were found to have been linearized by BamHI
and by HindIII. The double digestions make it possible to release a fragment corresponding to the whole of the cloned cDNA. The size of this fragment is about 1.2 kb in 3 cases and about 0.9 kb in the other 2 cases. For the following determination, one of the 0.9 kb frag-ments and one of the 1.2 kb fragments were selected and recloned (see section 6 below).
6) Determination of the sequence of urate oxidase cDNA
On the one hand one of the 0.9 kb fragments (clone 9A) and on the other hand one of the 1.2 kb fragments (clone 9C) were recloned in the DNA of the replicative form of single-stranded phage M13. The DNA
of the M13 clones, containing the 0.9 kb fragment on the one hand and the 1.2 kb fragment on the other, was digested with exonuclease so as to generate a series of overlapping M13 clones (procedure: "Cyclone I Bio-system" of IBI). Said clones were sequenced by the di-deoxyribonucleotide method (Sanger et al., PNAS-U.S.A.
- 1977, 14, 5463-5467).
The nucleotide sequence of clone 9C is shown in Figure 3, which also indicates, with an arrow, the start of clone 9A and, with a nucleotide symbol fol lowed by an asterisk ~, the sequenced nucleotides of clone 9A which are not identical to those of clone 9C
(when matching the two sequences and the AccI and BamHI
restriction sites used in the subsequent constructions (cf. 2)).
It is found that:
- the nucleotide sequence of the longer frag ment (clone 9C) overlaps that of the shorter fragment 05 (clone 9A) but for two differences (see Figure 3). One of the differences is quiescent and the other corres-ponds to a change from a tryptophan residue to a gly-cine residue. These differences may be due either to differences in the messenger RNA's isolated (cf. 2) above) or to errors in the reverse transcriptase used when building the cDNA library (cf. 3) above).
In the case of the longer fragment, an ATG
codon (in position 109 in Figure 3) opens an open reading frame corresponding to a polypeptide of 302 amino acids, with a molecular weight of about 34,240 Da, whose sequence corresponds to the partial sequence of purified A. flavus urate oxidase (cf. 4)).
Figure 4 shows the DNA sequence opened by the ATG codon and the polypeptide coded for, and, with arrows opposite the polypeptide coded for, the se quenced peptides (cf. 4)) obtained by hydrolysis of A.
'avus urate oxidase with trypsin and protease V8.
It is found that the sequence of the poly peptide terminates in the triplet Ser-Lys-Leu, which is typical of peroxisomal location enzymes (could S.J. et al., J. Cell. Biology 108 (1989) 1657-1664).
EXAMPLE 2: Construction of three expression vectors for urate oxidase cDNA in breast: plasmid pEMR469 carrvinct an ADH2 promoter and plasmid gE'~T'd73 and plasmid pEMR515 carrvina the artificial promoter of the invention The strategy employed uses fragments obtained from pre-existing plasmids available to the public, and fragments prepared synthetically by the techniques now in common use. The cloning techniques employed are those described by T. MANIATIS, E.F. FRITSCH and J.
SAMBROOK in "Molecular Cloning, a laboratory manual"
(Cold Spring Harbor Laboratory, 1984). The oligo nucleotides are synthesized with the aid of a Biosearch 05 4600 DNA synthesizer.
The following description will be understood more clearly with reference to Figures 5, 6 and 7, which respectively show restriction maps of plasmids pEMR414, pEMR469 and pEMR473. The symbols used in these Figures will be specified in the description below. In the case where a site has been blunted by Klenow polymerase, it carries the index "°"; where the sites have been eliminated by ligation, they are indi-cated in brackets.
1) Construction of~lasmid ~EMR469:
This plasmid was constructed from the shuttle vector E. coli-yeast pEMR414, constructed by successive ligations of the following components:
- the PstI-HindIII° fragment - symbolized by , ++++ in Figure 5 - of plasmid pJDB207 (BEGGS, 1978:
Gene cloning in yeast - p. 175-203 in: Genetic Engi-neering, vol. 2 - WILLIAMSON - Academic Press - London UK), comprising the upstream part of the ampicillin resistance gene AmpR of pBR322 (Sutcliffe, 1979, Cold Spring Symp. Quart. Biol. 43, 779) and an endogenous 2~
fragment, B form, carrying the LEU2 gene of S. cere-visiae partially modified by the deletion of its promoter (called LEU2d), the locus STB (REP3) and the origin of replication of the 2~ fragment (HARTLEY and DONELSON, 1980, Nature, 286, 860-865). The HindIII end of this fragment has been blunted by the action of Klenow polymerase. It is denoted by HindIII° in Figure 5.
- the HindIII-SmaI fragment - represented by ~ in Figure 5 - of yeast chromosome V, containing - is -the URA3 gene with its promoter (ROSE et al., 1984, Gene, 29, p. 113-124). This HindIII-SmaI fragment originates from plasmid pFLl (CHEVALLIER et al., 1980, Gene 11, 11-19). The HindIII end of this plasmid has 05 been blunted by the action of Klenow polymerase.
- an SamI-BamHI fragment - symbolized by in Figure 5 - containing a synthetic version of the promoter of the ADH2 gene which differs from the natural version described by RUSSEL and SMITH (RUSSEL
et al. (1983) J. Biol. Chem. 258, 2674-2682) only by a few base pairs intended for introducing restriction sites. (The natural sequence could be used with only slightly different results.) The sequence of this fragment is given below:

- 19 - 2o~ss~s S M
m 1 a I I
~GGGACGCGTCTCCTCTGCCGGAACACCGGGCRTCTCCAAC~TRTAAGTTGGAG
.______,_________._________,_________,_________,______ 05 . CCCTGCGCAGAGGAGRCGGCCT1GTGGCCCGTAGAGGTTGAATATTCAACCTC
AAATAAGAGAATTTCAGATTGAGAGAATGAAAAAAAAAAAAAAAAAAAAGGCAGAGuAGr'~-___,_________,_________,_________,_________,_________,______ TTTATTCTCTTAAAGTCTAACTCTCTTACTTTTTTTTTTTTTTTTTTTTCCGTCTCCTCT
S
to p h I
GCATAGARATGGGGTTCACTTTTTGGTAAAGCTATAGCATGCCTATCACATATAAATAGA
___,_________,_________,_________,_________,_________,______ CGTATCTTTACCCCAAGTGAAAAACCATTTCGATATCGTACGGATAGTGTATATTTATCT
15 ~,TGCCAGTAGCGACTTT1TTCACACTCGAGATACTCTTACTACTGCTCTCTTGTTGTTTT
_ _ _ ~ _ _ _ _ _ _ _ _ _ , _ _ _ _ _ _ _ _ _ , _ _ _ _ _ _ _ _ _ , _ _ _ _ _ _ _ _ _ , _ _ _ _ _ _ _ _ _ i _ _ _ _ _ _ CACGGTCATCGCTGAAAAAAGTGTGAGCTCTATGAGAATGATGACGAGAGAACAACAAAA
TATCACTTCTTGTTTCTTCTTGGTAARTAGAATATCAAGCTACAAAAAGCATACAATCAR
___,_________,_________,_________~_________,_________,______ ATAGTGAAGAACAAAGAAGRACCATTTATCTTATAGTTCGATGTTTTTCGTATGTTAGTT
20 ' C a 1 m a H
I
CTATCAACTATTAACTATATCGATACCATATGGATCCGTCGACTCTAGA~~;,TCGTC
_ _ _ , _ _ _ _ _ _ _ _ _ , _ _ _ _ _ _ _ _ _ , _ _ _ _ _ _ _ _ _ ; _ _ _ _ _ _ _ _ _ , _ _ _ _ _ _ _ _ _ , _ _ _ B
a n H
GACTCTAGAG i _ __ -_-CTGAGATCTCCTAG
- the BgIII-HindIII fragment - symbolized by in Figure 5 - carrying the 3' end of the yeast PGK gene. This fragment originates from complete di-gestion with BgIII of the HindIII fragment of the yeast - 2 0 - ~U~6'76 chromosomal DNA, carrying the PGK gene described by HITZEMAN et al. (1982, Nucleic Acids Res., 10, 7791-7808), which has only one BgIII site. This digestion makes it possible to obtain two HindIII-BgIII fragments 05 of which the smaller, of about 0.4 kb, which carries the 3' end of the yeast PGK gene, is retained. The sequence of the latter fragment is described by HITZE-MANN et al. (op. cit.). The BgIII site is cloned in the BamHI site of the previous fragment (the BamHI and BgIII sites therefore disappearing), and the HindIII
site, blunted by the action of Klenow polymerase, is cloned in the PvuII site of the PvuII-PstI fragment of pBR322, described below.
- the PvuII-PstI fragment - symbolized by xxx in Figure 5 - of pBR322, containing the origin of rep lication and the downstream part of the ampicillin resistance gene AmpR.
Plasmid pEl~t414 formed in this way therefore contains the following components:
- an origin of replication and an ampicillin resistance gene AmpR permitting the replication and selection of the plasmid in E. coli cells. These com-ponents permit transformation in E. coli cells.
- an origin of replication for the yeast (ARS), the locus STB and the LEU2 gene of S. cerevisiae without promoter and the URA3 gene of S. cerevisiae with its promoter. These components permit the repli cation and selection of the plasmid in S. cerevisiae cells and a sufficient partition efficacy in cells containing the endogenous 2~ plasmid.
Plasmid pEMR414 was completely digested with the restriction enzymes NheI and ClaI. The small NheI-ClaI fragment containing the URA3 gene, hereafter called fragment A, was purified.
Plasmid pEMR414 was completely digested with 21 2~~6'~'Ci the enzymes NheI and BamHI. The large NheI-BamHI
fragment containing especially the LEU2d gene and the origin of replication of plasmid pBR322, hereafter called fragment B, was purified.
05 The synthetic ClaI-AccI fragment, containing the start of a gene coding for the protein deduced from the urate oxidase cDNA sequence (clone 9C), was also prepared. This fragment contains modifications, rela-tive to clone 9C, introduced for the purpose of inser-ting codons which are customary in yeast (q.v. SHARP et al., 1986, Nucl. Ac. Res., vol. 14, 13, pp. 5125-5143) without changing the amino acids coded for. The se-quence of this fragment, hereafter called fragment C, is as follows (the underlined nucleotides are those modified relative to clone 9C):
C A
l c a c I I
CGATATACACAATGTCTGCTGTTAAGGCTGCTAGATACGGTAAGGACAACGTTAGAGT
___~_________+_ _ ._.+ _ .__._+.____ .__+____ ._.
TATATGTGTTACAGACGACAATTCCGACGATCTATGCCATTCCTGTTGCAATCTCAGA
The plasmid of clone 9C (cf. Figure 3) was digested with the enzymes AccI and BamHI. The AccI-BamHI fragment, which contains the end of urate oxidase cDNA, hereafter called fragment D, was purified. This fragment has the following sequence:

22 r2~'~~~~~
Accl.
~ CTACAAGGTTCACAAGG:\CGAGAAG
__+_________t_________, ~:cTTccA~c:c:~
vbv~VlW V
ACCGG~G.CCAGACGV~G~ACGAGATCACC G~C~GTGTGCTTCTGGAGGG~G:,G;.TTG:\G
_________+_________+_________~. _________+_________t_________t TGGCCACnGGTCTGCC..~CA:GCTCTACTCG CAG.~CACnCGAAGACCTCCC:,CTCTAACTC

_________+_________~._________.f _____.___+_________+_________ TGV..C.,..~G~GG~~CCGVC~V~~G~CGC..G TA:,G.GCG~~GC~~G..GV~A...~C-.V~VV
ATTTAC:.TG1CCGCCA:.GCACAACCCCGTT ACTCCTCCCC:.GCTCTTCGG:.TCC:\TCCTG
_________.f_________.f_________+ _________+_________+_________.f TAAATGTAGTGGCGGTTCGTCTTGGGGCAA TGAGGAGGGCTCGACAAGCCGAGGTAGGAC
GGCACACACT TCATTGAGAAGTACAACCAC ATCCA~GV.CG..TC:\CGTCAAC:.T:GTCTGC
1 0 _________+_________+_________+ _________+_________+_________+
CCGTGTGTGAAGTAACTCTTCATGTTCGTG TAGcTACGCCGAGTGCAGTTGTAACAGACG
C.\CCG\. ~ GGnCCCGG.,~ Gw\CA~ ~ G..CGV V. I1.,G\.G.CnCCC ~ C..C ~ CC ~ ~ C.. ~
........c.\C
_________+_________+_________; _________+_________+____.____+
G~uV\.GnCC~GGGCC~ACC~G~AAC~GCCv ~~CGV~GvGGVAG~G..GG/V1G~..GV..VV..V

_________+___._____+_________+ _________+_________f_________+
TCGCTCCTCTTCGCCTTACACGTCCACCTG CACCAGCTCCCGTTCCCGTAGCTATAGTTC
TCGTCTCTGTCCCGCCTGACCGTGCTCAAG AGCACCa.ICTCGC:\GTTCTGGGVUTTCCTG
_________t_________.~._________.t _________+_________+_________ AGCACAGACAGGCCGGi.CT CGCACGACT T C T CG T GGT T CAGCG T C.1AC:\CCCCG:v\GGAC
CCTG.>CGACTACACCACACTTAAGC:.CaCC TGGGaCCCTATCCTGAGCACCGACGTCG:.T
_________*_________t_________+ _________i_________+_________+
GC.:.CTGCTCATGTGGTGTGAATTCCTCTGG ACCCTGGCATAGG:.CTCG:GGCTGC:\GCTa GCCACTTGGCAGTGGAAGAATTTCAGTGGA CTCCAGGAGGTCCGCTCGCACGTGCCTAAG
2 0 _________+_________+_________+ _________t_________+_________+
CGV~GIV\CCGTCACCTTCTTNV1GTCACCT GAGGTCCTCCAGGCGAGCGTGCACGCaTTc TTCGATGCTACCTGGGCCACTGCTCGCGAG GTCACTCTGF.AGACTTTTGCTGA.1G:\TAAC
AAGCTACGATGGACCCCGTGACGAGCGCTC CAGTG:.G:.CTTCTCAAA:.CG:.C::CTATTG
AGTGCG.CCGTGCAGCCCACTATGTACAAG ATGCCAGAGCAAATCCTGGCCCGCCAGCAG
2 5 Tc:.ccGTccc:.cc:cccc:cATACATcTTc TACCC:c:ccTTTACG~ccccGCGGTcGTc CTGATCGAGACTGTCGAGTACTCG:TcCCT AACAAGCACTATTTCGAAATCG:.CCTGAGC
_________+_________+_________+ ._____.__+_______._+____.____+
GACTAGCTCTGACAGCTCATGAGCAACCGA TTGTTCGTGATAAAGCTTTAGCTCCACTCG
TGCCACAAGGCCCTCCAAAACACCCGCAAG AACGCCGACGTCTTCCCTCCTC:.GTCGG:.C
_________+_________+_________+ _________+_________+_________+
ACCGTGTTCCCGCAGGTTTTGTGCCCGTTC TTGCCGCTCCAGAAGCGAGG:.GTC:.GCCTG
3O CCCAACGGTCTGATCAAGTGTACCGTCCGC CGGTCCTCTCTCAAGTCTAAATTGTAA.ACC
_________+_________+_________+ _________+_________+_________+
GGGTTGCCAGACTAGTTCACATCGCAGCCC GCCACGAGAG:,CTTCAGATTT.1ACATTTCG
AACATGATTCTCACGTTCCGGAGTTTCCAA GGG'lllJICTGTATATAGTCTGCGATAGGGTA
____+_________+_________+ _________+______.__;_________+
TTGTACTAAGAGTGCAAGGCCTCA.1AGGTT CCGTTTGACATATATCAGACCCTATCCCAT
TAGC.~TTCATTCACTTGTTTTTT.1CTTCCA
3 5 _____ _t_________+_________+ _________+_________+_________+
ATCGTAAGTAAGTGAACAAAAAATCAACGT TTTTTTTTTTTTTTTTTTTTTTTTTTTTTT
AAAGGGCCCC~ BamNl _________+_________t________ TTTTTTTTTTTTTTTTTTTTTCCCCGGCCT AG

2 3 ~~,~~~6 Fragments A, B, C and D were ligated to give plasmid pEMR469 shown in Figure 6, in which the symbols have the same meanings as in Figure 5, the novel ClaI
AccI and AccI-BamHI fragments being symbolized by Plasmid pEMR469 carries a sequence coding for the protein deduced from the sequence of urate oxidase cDNA, under the control of a promoter called an ADH
promoter, similar to the natural ADH2 promoter, com-prising the sequence H

a _r tGCGTCTCCTCTGCCGGAACACCGGGCATCTCCAACTTATAAGTTGGfaG
_._,.____....,._....__.,_..______,_________,._____ ~AGAGGAGACGGCC1TGTGGCCCGTAGAGGTTGAATA1TCAACC7C

___,_....___.,_.... .._,......__.,.__._____,____..._.,_.____ S
P
h TATA canw_ nent I

.,...____..,......................._.____,_.._.._ .,_.____ CGTATCTTTA;,CCCAAGTGAAAAACCA111CGATATCG~TACGGATAGIG1ATATTTATCT

._.,..___....,......_..,.........,.....___.,._.._..._,_..___ CACGGTCATCGCTGAaAAAAGIGTGAGCTCTATGAGaATGATGACGAGAGaACAACAAAA

.__,__.__.__.,....._.._,..__._._.,_________,_...._.._,____._ ATAGTGAAGAACAAAGRAGAACCA1TTA1~TTA1AGTTCGATGTTTTTCG1ATGTTAGTT
C Transcription initiation region a CTATCAACTATTAACTA1AT~
-_.,.___...__,_._..__ 2 ) Construction of nl asmid oEt~iR473 Plasmid pEt~'IFt469 was completely digested with the en2ymes MluI and Sphl. The large MluI-SphI frag-ment, containing the urate oxidase gene, was then 05 ligated with the synthetic fragment, whose sequence is given below, corresponding to a part (200 bp) of the sequence upstream from the TATA component of promoter GAL7 of S. cerevisiae, said part comprising two high-affinity upstream activation sequences called UAS1 and UAS2, which are boxed off below (q.v. R.J. BRAri et al.
(1986) E2dB0 J., vol. 5, n~ 3, p. 603-608).
M
L
a CGCGTCTAT CTTCGGAGCACTGTTGAGCGAAG CTCATTAGATATATTTTCTGTCAT
__+______ _+_________+_________+ ________+_________+_____ AGATA AAGCCTCGTGACAACTCGCTTC GAGTAATCTATATAAAAGACAGTA

TTTCCTTAACCCAAAAATAAGGGAGAGGGTCCAAAAAGCGCTCGGACAACTGTTGACCGT
____+_________+_________+_________+___ ____+_________+_____ AAAGGAATTGGGTTTTTATTCCCTCTCCCAGGTTTTTC CGAGCCTGTTGACAACTGGCA
GA CCGAAGGACTGGCTATACAGTGTTCACAAAATAGCCAAGCTGAAAATAATGTGTAGC
_________+_________+_________+_________+_________+_____ . C AGGCTTCCTGACCGATATGTCACAAGTGTTTTATCGGTTCGACTTTTATTACACATCG
S
P
h I
CTTTAGCTATGTTCAGTTAGTTTGGCATG
3 0 ____+_________+_________+____ GAAATCGATACAAGTCAATCAAACC
Plasmid pEMR473 obtained in this way is shown in Figure 7, in which the symbols have the same meanings as in Figure 6, the novel MluI-SphI fragment introduced being symbolized by Plasmid pEMR473 therefore carries a sequence - 25 - 2~
coding for the protein deduced from the sequence of urate oxidase cDNA, under the control of the artificial promoter of the invention, which comprises the sequence M

a CGCGTCTAT CTTCGGAGCACTGTTGAGCGAAGGCTCATTAGATATATTTTCTGTCAT
AGATA GCCTCGTGACAACTCGCTTCCGAGTAATCTATATAAAAGACAGTA

1 0 TTTCCTTAACCCAAAAATAAGGGAGAGGGTCCAAAAAGC,GCTCGGACAACTGTTGACCGT
I
AAAGGAATTGGGTTTTTATTCCCTCTCCCAGGTTTTTCC~CGAGCCTGTTGACAACTGGCA
GATCCGAAGGACTGGCTATACAGTGTTCACAAAATAGCCAAGCTGAAAATAATGTGTAGC
CTAGGCTTCCTGACCGATATGTCACAAGTGTTTTATCGGTTCGACTTTTATTACACATCG
15 -~ S
P
h I TATA cannonent CTTTAGCTATGTTCAGTTAGTTTGGCATGCCTATCACATATAAATAGA
GAAATCGATACAAGTCAATCAAACCGTACGGATAGTGTATATTTATCT
GTGCCAGTAGCGACTTTTTTCACACTCGAGATACTCTTACTACTGCTCTCTTGTTGTTTT
CACGGTCATCGCTGAAAAAAGTGTGAGCTCTATGAGAATGATGACGAGAGAACAACAAAA
TATCACTTCTTGTTTCTTCTTGGTAAATAGAATATCAAGCTACAAAAAGCATACAATCAA
' ' ATAGTGAAGAACAAAGAAGAACCATTTATCTTATAGTTCGATGTTTTTCGTATGTTAGTT
< >
C Transeriptian l initiation region a I
CTATCAACTATTAACTATAT
GATAGTTGATAATTGATATAGC
3) Construction of plasmid ~EMR515:
Plasinid pEMR473 was partially digested with the enzyme XbaI and totally digested with the enzyme MluI.
The large XbaI-MluI fragment was purified. This frag ment contains especially the sequences of the origin of replication and the locus STB of the 2~ fragment, the LEU2d gene, the ampicillin resistance gene AmpR, the origin of replication of pBR322 and the expression cas-sette for urate oxidase. On the other hand, it con-tains neither the URA3 gene nor that part of the 2~c fragment which is between the XbaI and NheI sites.
05 The large XbaI-Mlul fragment was recircularized via the following sequence adapter containing MluI and modified XbaI sticky ends:
modified XbaI

MIuI
Plasmid pEMR515 obtained in this way has only 15 one of the three components of the target FRT site of the recombinase coded for by the FLP gene of the 2u fragment.
Plasmid pEMR515 therefore carries a sequence coding for the protein deduced from the sequence of 20 urate oxidase cDNA, under the control of the artificial promoter of the invention.
EXAMPLE 3: Transformation of the EMY761 yeast strain by plasmids pEMR469J, pEMR473 and p.EMR515 -Transformation of the EMY500 and GRF18 yeast 25 strains by plasmid,~EMR515 - Transformation with selection for the prototrophy of leucine Three non-isogenic strains of Saccharomyces cerevisiae were used as recipient strains:
30 - the EMY761 strain (Mats, leu2, ura3, his3, gal) - the EMY500 strain (Mats, leu2, ura3, pep4) - the GRF18 strain (Mats, leu2, his3) The GRF18 strain is well known to those skilled in the art (terry FINK, MIT, USA). The EMY761 and 35 EMY500 strains are related to the GRF18 strain. They were obtained by successively crossing the GRF18 strain with a ura3 strain derived from the FL100 strain (depo-sited in the ATCC under n° 28 383 ) and with the 20B12 strain (Mats, tspl, pep4) described by E.W. JONES (E.i9.
05 JONES et al. (1977) Genetics, 85, 23).
The GRF18 strain can be obtained by curing plasmid pEMR515 of the GRF18 pEMR515 (leu~) strain deposited in the CNCt~i under reference n° I-920 on 28 December 1989, and the Et~tY500 strain can be obtained by curing plasmid pEMR515 of the EMY500 pEM.R515 (leu+) strain deposited in the CNCM under reference n° I-919 on 28 December 1989.
These strains contain mutations (leu2 and ura3) capable of being complemented by the LEU2d defective selection marker and the URA3 selection marker, which are present in each of plasmids pEriR469 and pEMR473.
The transformation technique used is a variant of that described by Beggs et al. (Beggs et al. (1978), Nature 275, 104-109). It consists in subjecting yeasts to a protoplastization treatment in the presence of an osmotic stabilizer, namely sorbitol at a concentration of 1 M.
The precise transformation protocol is speci-fied below:
a) 200 ml of liquid YPG medium (cf. Table I) are inoculated with about 5 x 106 cells of a culture in the stationary phase, and the culture inoculated in this way is agitated overnight at 30°C.
b) When the density of the culture reaches about 10' cells per ml, the cells are centrifuged at 4000 rpm for 5 min and the residue is washed with sorbitol 1 M.
c) The cells are suspended in 5 ml of sorbitol solution 1 M containing 25 mM EDTA and 50 mM dithio threitol, and are incubated for 10 min at 30°C.

- 2 $ - ~Q~B~'~'~
d) The cells are washed once with 10 ml of sorbitol 1 M and suspended in 20 ml of sorbitol. Zymo-lase-100T (a preparation obtained by partial purifi-cation of Arthobacter luteus culture supernatant on an 05 affinity column and containing 13-1,3-glucan laminari-pentahydrolase, marketed by SEYKAGAKU KOGYO Co. Ltd.) is added up to a final concentration of 20 ~g/ml and the suspension is incubated at room temperature for about 15 min.
e) The cells are resuspended in 20 ml of a medium containing sorbitol, called sorbitol YPG medium (cf. Table I below), and incubated for 20 min at 30°C, with gentle agitation.
f) The cells are centrifuged for 3 min at 2500 rpm.
g) The cells are resuspended in 9 ml of trans-formation buffer (sorbitol 1 M, Tris-HC1 10 mM pH 7.5 and CaCl2 l0 mM).
h) 0.1 ml of cells and 5 ~cl of DNA solution (about 5 ~Cg) are added and the suspension obtained is left for 10 to 15 min at room temperature.
i) 1 ml of the following solution is added:
polyethylene glycol PEG 4000 20%, Tris-HCl 10 mM pH 7.5 and CaCl2 10 mM.
j) 0.1 ml of the suspension obtained in i) is poured into a tube containing leucine-free solid re-generation medium (cf. Table I below) which has been melted beforehand and kept liquid at about 45°C. The suspension is poured into a Petri dish containing a solidified layer of 15 ml of leucine-free solid re-generation medium.
k) Step j) is repeated with the remainder of the cell suspension obtained in h).
The transformed strains start to appear after three days.

~o~s~~

A transformed strain EMY761 pEMR469 (leu~), three transformed strains EMY761 pEhiR473 (leu~) (clones 1, 2 and 3 ) , a transformed strain EMY761 pEt-iR515 ( leu+) , a transformed strain EMY500 pEMR515 ( leu+) and 05 a transformed strain GRF18 pEMR515 (leu~) were thus retained.
TABLE I
Principal media used in Examples 3 4, Obis 6 and 7 - uracil-free solid medium 6.7 g of Yeast nitrogen base without Amino Acids ( from DIFCO ) 5.0 g of casein hydrolyzate (Casamino acids from DIFCO) 10 g of glucose g of agar Mix all the ingredients in distilled water and make up the final volume to 1 1 with distilled water.
Autoclave for 15 min at 120°C.
- uracil-free liquid medium Use the formulation of the uracil-free solid medium 20 without the agar. Autoclave for 15 min at 120°C.
- leucine-free solid medium 6.7 g of Yeast nitrogen base without Amino Acids ( from DIFCO ) 20 mg of adenine 20 mg of uracil 20 mg of 1-tryptophan 20 mg of 1-histidine 20 mg of 1-arginine 20 mg of 1-methionine mg of 1-tyrosine 30 mg of 1-isoleucine 30 mg of 1-lysine 50 mg of 1-phenylalanine 30 100 mg of 1-glutamic acid 150 mg of 1-valine 400 mg of 1-leucine 20 g of glucose 20 g of agar Mix all the ingredients in distilled water. Make up the final volume to 1 1 with distilled water. Auto slave for 15 min at 120°C. After autoclaving, add 200 mg of 1-threonine and 100 mg of 1-aspartic acid.

- leucine-free solid regeneration medium Use the formulation of the leucine-free solid medium, mixing in 30~g of agar instead of 20 g and adding 182 g of sorbitol to the mixture.
- leucine-free liquid medium 05 Use the formulation of the leucine-free solid medium without the agar. Autoclave for 15 min at 120°C.
After autoclaving, add 200 mg of 1-threonine and 100 mg of 1-aspartic acid.
- liquid YP medium g of yeast extract (Bacto-yeast extract from 10 DIFCO) g of peptone (Bacto-peptone from DIFCO) Mix the ingredients in distilled water. Make up the final volume to 1 1 with distilled water. Autoclave for 15 min at 120'C.
- liquid YPG medium Use the formulation of the liquid YP medium, adding, 15 after autoclaving, glucose at a concentration of 20 g/1.
- sorbitol YPG medium Use the formulation of the liquid YPG medium, adding, after autoclaving, sorbitol at a concentration of 1 M.
20 - ethanol-glycerol YP medium Use the formulation of the liquid YP medium. After autoclaving, add 10 ml of ethanol 100% (1% final con-centration) and 30 g of glycerol.
- ethanol-glycerol-galactose YP medium Use the formulation of the liquid YP mediu~a. After autoclaving, add 10 ml of ethanol 100%, 30 g of glycerol and 30 g of galactose.
EXAMPLE 4: Expression of urateoxidase by the EMY761 ~EMR469 lleu;7 and EMY761 pEMR473 (leu~l jclones 1, 2 and 3) strains - Immunodetec-tion by Western blot - Assay of the urate oxidase activity and the soluble proteins 1) Expression of urate oxidase:
a) Transformed strains In a first stage, a colony of each of the EI~iY761 pEriR469 ( leu+) and EZ~iY761 pEt~iR473 ( leu; ) ( clones 1, 2 and 3 ) strains was cultured in 25 ml of leucine-free liquid medium (cf. Table I, Example 3). This made it possible to obtain and maintain a large number of 05 copies of plasmids by carrying out the selection for complementation of the leu2 mutation by the LEU2 gene carried by plasmids pEt~t469 and pE2~473.
After 22 h at 30°C, with agitation, the two cultures were centrifuged for 10 min at 7000 rpm. The residues were taken up in 10 ml of sterile distilled water and centrifuged again for l0 min at 7000 rpm.
Expression of the urate oxidase was induced by taking up the cells in 20 ml of ethanol-glycerol YP medium for the Et~iY761 pEMR469 (leu+) strain and in 20 ml of ethanol-glycerol-galactose YP medium (cf. Table I, Example 3 ) for the EMY761 pEMR473 ( leu~) strain. The cultures were incubated again at 30°C for 27 h, with agitation.
c) Control strain The non-transformed EMY761 strain, i.e. the EMY761 strain without plasmid, was cultivated as above except that the first culture was carried out in liquid YPG medium. It was subjected on the one hand to induc-tion in 10 ml of ethanol-glycerol liquid YP medium and on the other hand to induction in 10 ml of ethanol-glycerol-galactose YP medium.
2) Preparation of the samples:
a) The cells cultivated in la), lb) and lc) were centrifuged and the supernatant was removed. The residues were taken up in 10 ml of distilled water and centrifuged for 10 min at 7000 rpm. The residues washed in this way were taken up in about 1 ml of tri-ethyleneamine buffer, TEA, of pH 8.9. About 300 ~1 of cells taken up in said buffer were lyzed in the pre-sence of glass beads (from 400 to 500 ~m in diameter), representing about half the final volume. This nixture was agitated vigorously in a Vortex 4 times for 1 min, the samples being placed in ice for 30 s between grin-ding operations. The liquid was withdrawn from the 05 tubes with a Pasteur pipette and transferred to a microtube. The glass beads were washed once with about 200 ~1 of TEA buffer of pH 8.9. The beads were agi-tated in a Vortex once for 1 min and the liquid was withdrawn with a Pasteur pipette and added to the above lyzate. The lyzate was then centrifuged in a microtube for 5 min at 7000 rpm. The supernatant was cautiously withdrawn and stored at -20°C for Western blot, assay of the urate oxidase activity and assay of the total soluble proteins. The residue of the lyzed cells was stored separately at -20°C for Western blot (cf. 3) below).
Furthermore, samples of the cultures prepared in la) and lb) were taken in the following manner before induction: 2 ml of culture were centrifuged for 10 min at 7000 rpm. The residues were taken up in 500 ul of distilled water and centrifuged again for 5 min at 7000 rpm. The residues were taken up in about 200 ~cl of TEA buffer of pH 8.9 and lyzed as above in the presence of glass beads. The supernatants and the residues of the lyzed cells were stored separately at -20°C. Assay of the oxidase activity and assay of the total soluble proteins were performed on the super-natants.
3) Immunodetection of the urate oxidase by Western blot:
a) Procedure The residues and the supernatants of the dif-ferent samples were subjected to a Western blot - a technique well known to those skilled in the art -which comprises the following steps:

solubilization of the residue by boiling for 10 min in a buffer, called a loading buffer, consisting of Tris-HC1 0.125 M pH 6.8, SDS 40, bromophenol blue 0.0020, glycerol 20%, f3-mercaptoethanol 10% (accor-05 ding to the protocol described by LAEt~iLI (U. K.
LAEMMLI, Nature, 227 (1970), 680-685)) (step perfor-med solely for the residues);
- electrophoretic separation of the different proteins contained in the solubilizate, according to the protocol described by LAEMMLI ( U . K . LAEMriLI , Nature , 227 (1970), 680-685); and - transfer of said proteins contained in the gel on to a nitrocellulose filter (according to the technique of H. TOWBIN et al., Proc. Natl. Acad. Sci. USA 76 (1979) 4350-4354).
Immunodetection, performed according to the technique of BURNETTE (W.W. BURNETTE, Ana. Biochem. 112 (1981) 195-203), involves the following successive operations:
~ rinsing the nitrocellulose filter for 10 min with a buffer A (Tris-HC1 10 mM, NaCl 170 mM, KC1 1 mri);
bringing the nitrocellulose filter into contact with a buffer B (buffer A with bovine serum. albumin added at a rate of 3 g per 100 ml) for 30 min at 37°C;
~ bringing the nitrocellulose filter into contact with an immune serum (polyclonal antibodies recognizing A.
lavus urate oxidase) for 1 h at 37°C;
rinsing the nitrocellulose filter with buffer B;
bringing the nitrocellulose filter into contact with a solution of protein G, labeled with iodine 125 at a rate of 0.1 microcurie/ml, for 1 h at 37°C;
rinsing the filter with buffer A;
drying the filter between two absorbent sheets;
bringing the filter into contact with an X-ray film;
and 34 ~~~~r~~
developing the film.
b) Results It is found that the EbiY761 pEMR469 (leu~) and EMY761 pEMR473 (leu+) (clone 1) strains produce a pro 05 tein with an apparent molecular weight of about 33 kDa, which is recognized by antibodies directed against A.
flavus urate oxidase (prepared in rabbits by techniques well known to those skilled in the art: q.v. VAITU-KAITIS et al. (1981) "Methods in enzymology", Academic Press, New York, vol. 73, p. 46) and which is absent from the control strain.
Comparison between the amounts of this protein for the residues and the supernatants makes it possible to deduce that about 800 of said protein is in soluble form in the lyzate.
4) Assay of the urate oxidase activity:
The urate oxidase activity was measured on the supernatants of the lyzed cells.
a) Principle The conversion of uric acid to allantoin is followed by the decrease in absorbance at 292 nm. The reaction is as follows:

H H
~ N 0 H zNCONH N 0 HN~ 5 ~ Urate oxydase 2 4 ~ >
0~ N 0 H
H20 + 0~ H20 + C02 Uric acid Allantoin (absorbs at 292 nm) b) Rea eq nts a) TEA 0.05 M pH 8.9/EDTA buffer - 7.5 g of TEA (reagent for analysis - Prolabo ref.
287.46.266) are dissolved in 400 ml of distilled water;

- 0.372 g of Complexon III (Merck'- ref. 8418) is dis-solved in 50 ml of distilled water;
- the two solutions are combined and made up to 500 ml (solution 1);
05 - the pH of this solution is adjusted to 8.9 with HC1 0.2 N; and - the volume is made up to 1000 ml with distilled water (solution 2).
b) Uric acid stock solution - 100 mg of uric acid (Carbiochem - ref. 6671) are dis-solved in 50 ml of solution 1;
- the pH is adjusted to 8.9 with HC1 0.2 N; and - the volume is made up to 100 ml with distilled water.
The solution obtained can be stored for one week at 4°C.
c) Uric acid substrate solution - 1.5 ml of uric acid stock solution (Carbiochem - ref.
6671) are taken and diluted to 100 ml with TEA buffer (reagent for analysis - Prolabo ref. 287.46.266).
This solution must be used the same day.
c) Procedure The following volumes are introduced into the quartz cell of a spectrophotometer set to 292 nm and thermostated at 30°C:
- 600 ~1 of uric acid substrate solution (preheated to 30°C); and - 100 ul of the above supernatants to which 200 ~cl of TEA pH 8.9 have been added (preheated to 30°C).
After mixing, the change in optical density (sometimes abbreviated to OD hereafter) is read off every 30 s for 5 min. Q E, the variation in optical density per minute, is deduced from these readings.
d) Expression of the results The urate oxidase enzymic activity A, expressed in U/ml, is calculated from the aE measurement with the aid of the formula A = DE x Vr x d I x VpE

in which the symbols Vr, d, I and Vp~ respectively represent the reaction volume (0.9 ml), the dilution factor (2), the extinction coefficient of uric acid at 292 nm (12.5) and the volume of the test sample (0.1 ml).
5) Assay of the total soluble proteins in the lvzates:
The protein assay kit from BIORAD was used for assaying the total proteins present in the supernatant of the lyzed cells. It is based on the observation that the maximum absorbance of an acid solution of Coo-massie brilliant blue g-250 changes from 465 nm to 595 nm when proteins become attached thereto (q. v. Reisner et al., Anal. Biochem., 64, 509 (1975)).
procedure The following volumes are introduced into the cell of a spectrophotometer set to 595 nm:
- 10 ~1 of sample to which 790 ul of distilled water have been added; and - 200 ~C1 of concentrated Dye reagent (Biorad).
The ingredients are mixed and the optical den-sity is read off at 595 nm. A calibration range with increasing concentrations of BSA (bovine serum albumin) was prepared in this way. The unknown concentration of the total proteins in the lyzates is read off on the calibration curve obtained.
6) Results:
The results obtained are collated in Table (II) below, which specifies, for each strain, the culture medium, the carbon and energy source of the culture, the urate oxidase activity in U/ml, the amount of total soluble proteins in mg/ml and the percentage of urate oxidase in the total soluble proteins. This last para-meter is calculated by assuming that the specific activity of the recombinant protein is identical to 05 that of the urate oxidase obtained from A. f lavus : 30 U/mg.

2~~~'~

i N

ro I
p G

i w .c O

O +n M O O O O O ~ tf11f1 O
L

d C tl' tT.-~d0 0 0 0 0 0 .X 0 0 0 Ovl1 ppr.s p .--I .r N rr .-.

G N

V ro .-t U ~
O

Yr -, N

d x n.
O

i N

~
G

_ O~ O~ o e~ tC1 N o .-..I.-~.-. tD ao ~ o ar O
E
N d w E d' M \D tt1 ~ ~!'N M M M vf1 M ~f1 V N

r0 L
.~

w a.

O

I

d N

roT

'~
+u .--.

~x o 0 0 0 0 0 ~n o 0 0 o M
>
E

w O O O O O O O 00 O O O O t~ ~J N O

v +~ v v v v v v ro S~

p _ w ~ _w w T .--n.-1 -r .-a 0, O O O O O O O O O O

L. La 1,. 1W . L 1r 1, ~. i'. 1..
p v ~ Q~ U U p v O 'U V O J

G w U U U U U U U U U U

Q) fy ~ ~ ~ ~ T ?- T T T T T T T T
~

w tr N C7 V7 N N

'v ~ ~ ~ O ~ tn v~ v~ v~ v~ v. a7 tT O v~ a~
O v' d v~ a~
~

C y J U W ~ v W W W n v N v N W V7 W N

.-~ ro a ~ a ~ a .-,.-~ -. _ -. .~ o -. o -. o -.
a~ o .~ o -.

CO U O --~ -~ --~--n -~ O O O O O O .-I O .~ O w O
w O y C lr p~ D' p~ p~ p~ C C C C C G U G U G U C U C
~ U V

M a ro ro ro ro ro ro ro ro ro ro ro g ro ro ~
ro m o ~ s ~ t t _ . .c .~ ~ .~ .c --.
~ _.

,d ~.r r.r_ r-~ru r.., ..~ ro w ro w ro N .u E~ V p O1 d V N U D' U C~ CJ D' V
P O tT

~ ~ ~ ~ ~

d d O d CJ

_ _ _ _ _ N N V7 N N

'D 'D 'C7'~ '~ O
O
O
O
O

c ~ ~ ~ ~ ~ E ~ ~ ~ _ _ _ _ _ o ~ ro w w ro ro V V V V V _ _ _ ..~ v v 'v w A. a. A. a. ro ro ~~ w w ro > a ~ >. >. v, v. o. v. v.

O 1.. a a a a ~ v w v w w _ w ~ ~ w .--~-r .~ -. -i _ ro -. -. -.

.~70 0 0 0 O ~ O ~ O

.~. E a~ a~ a~ a~ ~ a~ O a~ O
~. a>

GY d d d d V V U U U U U U U U

d d ~ ~ T T T T T T T T T T

N O .-v 1-..1r L C7 .--wr ~ .-r.--m -a .-1 W w w w w 0. tT v~ U ~ v , ~ ~ v.
a ~ C~ ~ U
~ ,F, ~ ,F

w~L1 n v I I Y w w w ~ w ~
J
~
wwt w O O U p _ ~.r .--~~ .-r _ _ _ ....~ ...-, ..--~
.-v G G G G '~ O O O O O O '~ O 'O O '~J O
TJ O '~

N _ _ _ _ _ c m ~ ~

V U U U a r ~ ~ ro E ro E ro a a a a o- v .c .c ro E ro_ E
t a~ a~ a> a~ .-~~ .-~.-~.-~ .-~ o, .-. o, a .-~
o. ' a.

.-, ...,.~ .--~~ a> ai a> a~ ~ a> >. a> >. a> >.
a> >- am.

N M M ~ N ~ N M

d d d d d d d d d C G C G C G C C G

O O O O O O O
O O

U U _ U V U _ . . U ~r . U U U
. . . . .
. . .
. .

- _ _ .-. - -_ -.

~D t~ tD

1 ~.. t~ 1 4 1 t~ i v 1 a a ~ a a. >..a a a a r a a a a G

Q1 M M M Q1 M M M d M M M Q~
ro V ~ ~ r~ V r~ e~ e~ t~ ~ ~ V
~

L.1 ct' ~ Q' d' p' Q Q' d' V' ~T d' ~1' w G CG tx G' O O GG p'..G: G'r O CC C
.' ~ P.' C~

_ ~3 ~ ~3 N N ~3 ~i ~i ~3 _ G3 Npl~f~~3 R G1 O..D. G C L1 G. GL GL C G1. L1 Q. Q.

w ro ro _ .-- mr .r .~ H it ~ rr .--v.~ S-a ~D ~D tD ~D w w tD tD ~O ~D a~ ~D ~D ~O ~D

n n n I I f~ e~ A r~ I n n n t~

> a. >. ~. c G ~-. r ~- >. c a >. >.
c ~ ~ ~ c c ~

2~~~'~~

This Table shows that:
a) in the presence of glucose, the ("repres-sed") level of urate oxidase is not detectable in the case of the artificial promoter of the invention 05 (strain EMY761 pEMR473 (leu+) (clone 1, 2 or 3)), whereas it is detectable in the case of the ADH2 promoter (strain EMY761 pEMR469 (leu+)). In the presence of glucose, therefore, the artificial promoter permits better repression than the ADH2 promoter.
b) in the absence of glucose but in the pre-sence of ethanol/glycerol, the level of urate oxidase is high for the ADHZ promoter ( about 14% of the total soluble proteins) and low but detectable for the arti-ficial promoter.
c) in the absence of glucose but in the pre-sence of ethanol/glycerol/galactose, the level of urate oxidase retains a value little different from that of the previous case for the ADH2 promoter (about 13.5% of the total soluble proteins), but reaches a high value (about 180 of the total soluble proteins) for the arti-ficial promoter.
The artificial promoter of the invention there-fore permits a high level of production of recombinant protein and has three levels of expression:
- zero level a) - basic level b) - maximum level c) EXAMPLE Obis: Expression, in an Erlenmeyer flask, of urate oxidase cDNA by the EMY761 pEMR515 (leu~), EMY500 pEMR515 (leu+~ and GRF18 pEMR515 fleu+) strains A colony of each of the above three strains was cultured in 20 ml of leucine-free liquid medium.
After one night at 30°C, with agitation, the three cultures were centrifuged for 10 min at 7000 rpm.

..~ 2s~s~s The cell residues were taken up in 10 ml of sterile distilled water and centrifuged again for 10 min. Ex-pression of the urate oxidase was induced by taking up the cells in 20 ml of ethanol-glycerol-galactose YP
05 medium (cf. Table I, Example 3). The cultures were incubated again at 30°C for about 20 h, with agitation.
As a control, a culture of each non-transformed host strain was prepared.
The cells of each of the six cultures are re deposited by centrifugation and the supernatant is removed. The residues were taken up in 10 ml of dis tilled water and centrifuged for 10 min at 7000 rpm.
The residues washed in this way were taken up in about 1 ml of TEA buffer of pH 8.9 and the grinding and re moval of the particles by centrifugation were carried out in the manner described in Example 4, 2). The supernatant of each culture is used, as before, for assay of the urate oxidase and the total proteins. The principal results obtained are collated in Table III.
below:

- 41 - 2o~s~s T?.BLE III
Strain/culture conditionsUrate oxidaseTotal soluble% of orate activity proteins obidase ~
(U/nl) (~g/al) in the sol~le protei:a GRF18 pEHR515 (leu')/a)< 0.1 2.2 < 0.05 EHY500 pEHR515 (leu')/a)< 0.1 0.9 < O.C5 EHY761 pEMR515 (leu')/a)< 0.1 1.8 < 0.05 GRF18 pEMR515 (leu')/b)38 5.4 23 EMY500 pEHR515 (leu')/b)20 2.5 26 EHY761 pEHR515 (leu')/b)33 4.2 26 a): the strains are cultivated in the presence of glucose (non-induction conditions) 1 5 b): the strains are cultivated in the absence of glucose and in the presence of galactose (induction) The promoter according to the invention there-fore permits a high level of expression of orate oxidase in three non-isogenic strains.
20 EXAMPLE 5: Construction of two expression vectors for f3-qalactosidase in yeast: plasmid oEr1R429 carrying an ADH2 gromoter, and olasmid pEMR437 carrying the artificial pronoter of the invention 25 The strategy employed uses fragments obtained from pre-existing plasmids available to the public, and fragments prepared synthetically by the techniques now in common use. The cloning techniques employed are those described by T. MANIATIS, E.F. FRITSCH and J.
30 Sp~gROOK in "Molecular Cloning, a laboratory manual"
(Cold Spring Harbor Laboratory, 1984). The oligo-nucleotides are synthesized with the aid of a Biosearch 4600 DNA synthesizer.
1) Construction of plasmid pEMR429:
35 Plasmid pEMR414 (cf. Figure 5) was completely digested with the restriction enzyme BamHI. The BamHI
site, located between the promoter sequences (ADH~) and terminator sequences (PGK), is unique. The linear DNA
of the plasmid is purified by elution from an agarose 05 gel after electrophoresis. Plasmid pt~iC1403 (ref. Casa-daban et al. (1980), J. Bacteriol., 143, 971-980) was completely digested with the enzymes BamHI and EcoRI, which made it possible to release a BamHI-EcoRI DNA
fragment of about 3 kb containing the essential part (upstream part) of the sequence coding for E. coli f3-galactosidase. The complete sequence was reconstituted with the aid of the synthetic fragment of the following sequence:
EcoR I~ r,r n,.nrr~~~ ~r 5 ~ nATT T CAGE, i GnG~,v.~~uT~u AGTCGACTCGCGGCCAGC
CTACCATTACCAGTTGGTCTGGT
GATGGTAATGGTCAACCAGACCA
GTCAAAAATAATAATAAG
CAGTTTTT:,TTATTATTCCTAGv BamHl The BamHI-EcoRI fragment originating from the double digestion of pMC1403 is ligated at EcoRI with the above synthetic fragment.
The BamHI-BamHI fragment obtained is then ligated with the linear DNA of plasmid pEMR414 digested with BamHI, to give plasmid pEMR429 shown in Figure 8, in which the symbols have the same meanings as in Figure 5, the BamHI-EcoRI and IcoRI-BamHI fragments introduced being represented by .
In this plasmid, the sequence coding for B-4 3 - ~~~b~! 6 galactosidase is under the control of the ADH2 promoter described in Example 2.
2) Construction of plasmid pEMR461:
Plasmid pEMR429 was completely digested with 05 the enzymes MluI and SphI. The large MluI-SphI frag ment, containing the 13-galactosidase gene, was then ligated with the synthetic fragment whose sequence, given below, comprises the upstream activation se quences (UAS) of the promoter of the GAL7 gene of S.
cerevisiae and MluI and SphI sticky ends.
~A
U
CGCGTCTA1ACTTCGGAGCACTGTTGAGCGnAGGCTCA11AGATATATTITCTG1CAT
__,._.__._..,....._.__,__...___.,._.______,_______._,.____ AGATATGAAGCCTCGTGACAACTCGCTTCCGAGTAATCTA1ATAAFiGiG~CAG1A
TTTCCTTAACCCAAAAATAAGGGAGAGGGTCCAAAAAGCGCTCGGACAaCTGTTGACCGT
..._,___._____,.._._____,_____.___,____.____._________,_____ AAAGGAATTGGGTTTiTATTCCCTCTCCCRGGiTTITCGCGnGCCTGTTGt~CnACTGGCa GaTCCGaaGGACTGGCTATACAGTGTTCACAar~ATAGCCAaGCTGAr~AATAATG1GTAGC
__._,,_____._._._..______,_._._..__,___..____,__._.____,_____ ~ S
P
h I
CTTTAGC1ATG1TCAGTTAGTT1GGCATG~
__..,_...__...,._.._._._,_....
GAAATCGATACAAGTCAATCAAAC G
Plasmid pEMR461 obtained in this way is shown in Figure 9, in which the symbols have the same mea-nings as in Figure 8, the novel MluI-SphI fragment introduced being symbolized by ~.
In this plasmid, the sequence coding for B
galactosidase is under the control of the artificial promoter of the invention, described in Example 2.

- 44 - ~iQ~ .~7~~~
EXAMPLE 6: Transformation of the DBY746 S. cerevisiae strain by plasmids pEMR429 and pE2dR461 Transformation with selection for the proto-05 trophy of uracil:
A colony of the DBY746 strain, which is (t~fata, his3, leu2, ura3, trpl, cyhR) (ROSE et al. (1981), PNAS
USA, 78, 2460-2464), was used to inoculate 100 ml of a medium called liquid YPG medium (cf. Table I of Example 3). When the cell density had reached 10' cells per ml, the cells were treated with lithium acetate 0.2 i-i for transformation by a technique well known to those skilled in the art and described by ITO et al. (ITO et al., 1983, J. Bacteriology 153, 163-168).
The DBY746 cells were transformed in parallel with about 1 ~g of each of plasmids pEMR429 and pEl~iR461. The transformed cells are selected for the auxotrophic character of uracil (ura~) on a medium called uracil-free solid medium (cf. Table I of Example 3). A transformed strain DBY746 pEMR429 (ura+) and a transformed strain DBY746 pEMR461 (ura~) were thus retained.
EXAMPLE 7: Production of fi-galactosidase with the aid of the DBY746 pEMR429 (ura~l and DBY746 pEMR461 (uraW strains 1) ~~ression of f3-galactosidase:
A transformed colony DBY746 pEMR429 (ura~) and a transformed colony DBY746 pEMR461 (ura+) were each used to inoculate 20 ml of uracil-free liquid medium to which tryptophan (10 mg/1) had been added beforehand.
After one night at 30°C, with agitation, 1% of glucose is added and culture is allowed to continue for 4 h. A
check is then made to see that there is still some glucose in the cultures. An aliquot is taken in order to assay the B-galactosidase.

- 45 - 2~~~el~i After one night at 30'C,~with agitation, the two cultures were centrifuged for 10 min at 7000 rpm.
The residues were taken up in 10 ml of sterile distil-led water and centrifuged again for 10 min at 7000 rpm.
05 Expression of !3-galactosidase was induced by taking up the cells in 20 ml of ethanol-glycerol YP medium (cf.
Table I, Example 3) for the DBY746 pEMR429 (ura+) strain and in 20 ml of ethanol-glycerol-galactose YP
medium (cf. Table I, Example 3) for the DBY746 pEMR461 (ura+) strain. The cultures were incubated again at 30°C overnight, with agitation.
2) Preparation of the samples and assay:
The cells cultivated above were centrifuged and the supernatant was removed. The residues were taken up in 10 ml of distilled water and centrifuged for 10 min at 7000 rpm. The residues washed in this way were taken up in about 1 ml of B-galactosidase assay buffer (EDTA 2 x 10-3 M; Na2HP04 7 x 10-2 M; NaH~P04 3 x 10-2 M; MgS04 10-3 M; MnS04 2 x 10-3 M). About 300 ~C1 of cells taken up in said buffer were lyzed in the pre-sence of glass beads (from 400 to 500 ~Cm in diameter), representing about half the final volume. This mixture was agitated vigorously in a Vortex 4 times for 1 min, the samples being placed in ice for 30 s between grin-ding operations. The liquid was withdrawn from the tubes with a Pasteur pipette and transferred to a microtube. The glass beads were washed once with about 200 ~C1 of TEA buffer of pH 8.9. The beads were agita-ted in a Vortex once for 1 min and the liquid was with-drawn with a Pasteur pipette and added to the above lyzate. The lyzate was then centrifuged in a microtube for 5 min at 7000 rpm. The supernatant was cautiously withdrawn and stored at -20°C for Western blot, assay of the urate oxidase activity and assay of the total soluble proteins. The residue of the lyzed cells was stored separately at -20'C.
The f3-galactosidase activity was assayed by the technique of PARDEE (PARDEE et al., J. Mol. B. (1959), 1, 1656-178).
05 Furthermore, the total soluble proteins were assayed using the BIORAD protein assay kit, as des-cribed in Example 4.
The results obtained are collated in Table IV
below:
TABLE IV
Strain Carbon and B-GalactosidaseTotal solubleCulture energy ~ediu~

source of activity proteins the culture U/nl pg/nl DBY746 glucose 59 375 liquid ~ediun pEHR429 (ura~) without uracil +

tryptophan (10 ng/1) +
glucose (1%) 2 0 DBY746 ethanol/glycerol6500 1700 ethanol-glycerol pE.u.R429 YP nediun (ura~) DBY746 glucose 0 350 liquid nediun pEHR461 (ura~) without uracil +

tryptophan (10 ng/1) glucose (1%) DBY746 ethanol/glycerol/400 520 ethanol-glycerol-pEHR461 (uraa) galactose galactose YP

aediun This Table shows that:
_ in the presence of glucose, the EMY746 pEMR429 (ura+) strain produces a small amount of f3-galactosidase, whereas this protein is not detected for the DBY746 pEMR461 (ura+) strain. The artificial promoter of the invention therefore permits better repression than the ADH2 promoter.

- under induction conditions, the artificial promoter leads to a high level of expression of ~3-galactosidase, although under these conditions it is lower than the level obtained with the ADH~ promoter.
S
M
In a CGCGTCTATACTTCGGAGCACTGTTGAGCGAAGGCTCATTAGATATATTTTCTGTCAT
. . . _ _ _ . _ . .f _ _ _ _ _ . _ . _ ~. _ _ . . _ . _ _ _ y . . . . _ . . _ . .~. . . _ . . . . _ _ i . . . . .
AGATATGAAGCCTCGTGACAACTCGCTTCCGAGTAATCTATATAAAAGACAGTA
TTTCCTTAACCCAAAAATAAGGGAGAGGGTCCAAAAAGCGCTCGGACAACTGTTGACCGT
_ . . _ . _ . . _ i. . . . _ . _ . _ . + _ . . . . _ . _ _ + . _ _ _ _ . . _ _ ~. . _ . . _ _ _ _ _ .~ . . _ _ _ rIAAGGAATTGGGTTTTCATTCCCTCTCCCAGGTTTTTCGCGAGCCTGTTGACAACTGGCA
GATCCGAAGGACTGGCTATACAGTGTTCACAAAATAGCCAAGCTGAAAATAATGTGTAGC
15 ._..+.._.__.__y....____.+________.+..._.....+______._.+.....
CTAGGCTTCCTGACCGATATGTCACAAGTGTTTTATCGGTTCGACTTTTATTACACATCG
P
h CTTTAGCTATGTTCAGTTAGTTTGGCATGCCTATCACATATAAATAGA
_ . . _ ~. . . . . . . _ . . ~. _ . . . . . . _ _ + _ . . . _ . + . . . . . .
. . . ~. . . _ _ _ _ GAAATCGATACAAGTCAA-rCAAACCGTACGGATAGTGTATATTTATCT
GTGCCAGTAGCGACTTTTTTCACACTCGAGATACTCTTACTACTGCTCTCTTGTTGTTTT
. . . . _ . . . . + _ _ . . _ . _ . . + _ . . . . _ _ _ . .~ _ . . . . _ . . .
~. . _ _ . . . . . . ~. . _ . _ . .
CACGGTCATCGCTGAAAAAAGTGTGAGCTCTATGAGAATGATGACGAGAGAACAACAAAA
TATCACTTCTTGTTTCTTCTTGGTAAATAGAATATCAAGCTACAAAAAGCATACAATCAA
_.__._.._.F..._._... r..._...._~.._._____.~......__...H__..._ ATAGTGAAGAACAAAGAAGAACCATTTATCTTATAGTTCGATGTTTTTCGTATGTTAGTT
C

~S 1 C'T'ATCAACTATTAACTATAT
_....._..~.....__ GATAGTTGATAATTGATATAGC
3~

- 47a -EXAMPLE 8 . Construction of two vectors for the expression and secretion of the human cytokinin gro-(3 in yeast plasmids pEMR575 and pEMR583 carrying the artificial promoter of the invention The examples described above concern proteins whose localisation is intracellular. Now, it is known that yeast can secrete recombinant proteins in the culture medium. The use of the metabolic pathway leading to secretion of the protein has several important advantages .
1 - It enables a reasonably pure and correctly matured product to be recovered from the culture supernatant.
2 - It enables the protein to benefit from the modifications associated with the secretion pathway, such as the formation of disulfide bridges, glycosylation etc.
There are several proteins or polypeptides naturally secreted by yeast. In the majority of known cases, these proteins are synthesized in the form of a longer precursor whose NH2-terminal sequence is decisive for entry into the metabolic pathway leading to secretion. In certain cases, these NHS-terminal sequences can be used for the secretion of heterologous proteins. Among these sequences, it is known to use the pre-pro system of the alpha pheromone. The alpha sex pheromone of yeast is a peptide of 13 amino acids which is secreted in the culture medium by S. cerevisiae yeasts of the Mata sex type.
The alpha factor arrests the cells of the oppo-- H ~ -20,~0,~~
site sex type (t~fata) in the G 1 phase and induces the biochemical and morphological changes necessary fcr conjugation of the 2 types of cells. Kurjan, J. a.::d Herskowitz, I. (1982), Cell, 30, 933-943, cloned the 05 structural gene of the alpha factor and deduced from the sequence of this gene that this actor of 13 a::,ino acids is synthesized in the for:a of a pre-pro prec~.:rsc=
protein of 165 amino acids. The precursor contains a hydrophobic amino-terminal sequence of 22 amino acids followed by a sequence of 61 amino acids containing 3 glycosylation sites, followed finally by 4 copies of the a factor. These 4 copies are separated by space=
sequences and the mature protein is released fron the precursor by virtue of the ~ollowing enzymic activi ties:
1 - an endopeptidase of the cathepsin B type (product of the KEx2 gene, called yscF) which cleaves Lys-Arg dipeptides at the carboxy terminal end.
2 - an exopeptidase of the carboxypeptidase type (product of the KEX1 gene) which cleaves the basic residues at the carboxy terminal end of the excised peptides.
3 - a dipeptidylaminopeptidase (called A) {prcduct of the STE13 gene) which removes the Glu-Ala and Asp-Ala doublets.
A first example of a protein which is secreted by this system and uses the promoter of the invention is the human cytokinin gro-f3. The cDNA of this pro-tein, which is called either gro-a (S. Haskill et al., 1990, Proc. Natl. Acad. Sci. USA, 87, 7732-7736) or MIP-2a {P. Tekamp-Olson et al., 1990, J. Exp. tiled., 172, 911-919) was recently cloned and sequenced, gro-f3 belongs to a family of cytokinins whose members appear to be involved in modulation of the inflammatory res-ponse and in activities of the growth factor type.
The Applicant tested the use of the promoter ~s,~s,~s for the secretion of gro-B by ~. cerevisiae. To do this, it replaced the natural signal sequence of gro-!3 with the pre-pro sequence of the alpha pheromone and placed the precursor of gro-!3 behind the promoter of 05 the invention.
1 - Construction of plasmid pEMR530 (cloning vector):
Plasmid pEMR4~3 described above (ExaMple 2 2) -Figure 7) was digested with the enzymes Xhoz and Sall and the large fragment, hereafter called fragnent E, was isolated. This large fragment comprises the sequences of the URA3 gene, the origin of replication and the locus STB of the 2~ fragment, the arapicillin resistance gene AmpR, the origin of replication of plasmid pBR322 and the UA5 of the promoter of the GAL7 gene of ~. cereyisiae, as well as the terminator of the PGK gene.
A double-stranded oligonucleotide sequence of about 400 base pairs, called fragment F, was synthe-sized in the form of an Xhol-SalI fragment. This sequence brings the TATA region and the initiation region of the ADH~ promoter, which is extended by a synthetic sequence preceding the start of the pre-pro region of the alpha pheromone. The sequence of this pre-pro region of the alpha pheromone differs from that described by Kurjan and Herskowitz, 1982, Cell, 30, 933-943, in the introduction of a HindIII site by the silent mutation of the TCT codon - corresponding to serine 81 of the precursor of the pheromone - to AGC.
The whole sequence of the fragment is given below:

x h I
CGAGATACTCTTACTACTGCTCTCTTGTTGTTTTTATC~CTTCTTGTT~C
_ i ~ _ ~ _ _ _ . ~ . . _ _ ~ _ _ ~ _ i ~ _ ~ _ ~ _ _ _ _ y ~ ~ _ _ _ _ _ _ _ a ~ _ _ _ _ _ _ _ _ TATGAGAATGATGACGAGAGAACAACAAAAATAGTGAAGAACRAAG

TTCTTGGTAAATAGAATATCAAGCTACAAAAAGCATACAATCAACTaTCAATCAGATCTA
y.......__~___~__~~~a~~_~~~~~~a~________y_________~_________ AAGAACCATTTATCTTATAGTTCGATGTTTTTCGTATGTTAGTTGr7TAt;TTAGTCTAGr=.T
ATATTAATAAAAAATGAGATTTCCTTCFiATTTTTACTGCAGTTTTATTCGCA~;CATCCTC
y____~____a__~~~_...y.........~_~_~__~_~~~~_______._________ TATAATTATTTTTTACTCTAAAGGAAGTTAAAAATGACGTCAAAATr'iAI,~CGTCGTAGGr'aG
CGCATTAGCTGCTCCAGTCAACACTACAACArAAGATGAAACGGCIaCAC,ATTCCGGCTGG~
y _ _ _ ~ ~ _ _ _ _ ~ _ _ _ ~ _ ~ _ _ _ ~ _ ~ _ ~ ~ _ _ . . y . . . _ _ _ _ ~
~ a _ ~ _ _ _ _ _ _ _ a ~ _ _ _ _ _ _ _ _ GCGTAATCGACGAGGTCAGTTGTGATGTTGTCTTCTACTTTGCCGTGTTTAAGGCCGr~C~
AGCTGTCATCGGTTACTTAGATTTAGAAGGGGATTTCGATGTTGCTGTTTTGCCaTTTTC
y.....____~__~~_____y~_____~~~~_~_______~._____.__~_________ TCGACAGTAGCCAATGAATCTAAATCTTCCCCTAAAGCTACAACGACAAAACGGTAAaAG
CAACAGCACAAATAACGGGTTATTGTTTATAAATACTACTATTGCCAGCATTGCTGCTAA
GTTGTCGTGTTTATTGCCCAATAACAAATATTTATGATGATAAC~GTCGTAACGaCGaTT
H
i n d S
I a I I
AGAAGAAGGGGTAAGCTTGCATGCCTGCAGG ~
a_._______i_________.._______~-i__ ___~
TCTTCTTCCCCAT TCGAACGTACGGaCGTCCpGC~
Fragments E and F were ligated to give plasmid pEMR530 shown in Figure 10, in which the symbols have the same meanings as in Figure 7, the novel Xhol-Sall _fragment (fragment F) introduced being represented by:
This plasmid comprises the artificial promoter of the invention, the sequence of which was given in 51 ,~'~'~6"7S
Example 2.
2 - Construction of plasmid pEMR583 (expression plasmid for the human Cytokinin gro-f~) Plasmid pEMR530 was completely digested with 05 the enzymes Nhel and HindIII. The small NheI-HindIIT_ fragment, containing the artificial promoter and the pre-pro region of the a pheromone, as well as the URA3 gene, was purified (hereafter called fragment G).
Plasmid pEMR473 was completely digested with the restriction enzymes NheI and BainI~iI. The large fragment (hereafter called fragment H), comprising the origin of replication and the locus STB op the 2u frag ment, the LEU2d gene, the ampicillin resistance gene, the origin of pBR322 and the terminator of the PGK
gene, was purified.
The cDNA of gro-13 was cloned and sequenced according to the method described by Tekamp-Olson et al., op. cit. The sequence of the cDNA of gro-B
(described in detail in said reference - in particular Figure 2) has an EcoRI site (sequence: 5'-GAATTC) which covers the ATT codon (isoleucine in position +18 of the mature sequence of gro-f~, Tekamp-Olson et al., op.
cit.) and overlaps the two flanking codons GGA and CAC.
The cloned cDNA of gro-I3 terminates at the 3~ end in a polyA tail flanked by a BamHI restriction site.
The EooRI-BamHI fragment, comprising the major part of the coding sequence of gro-13, followed by the 3' sequence corresponding to the non-translated end of the mRNA flanked by the polyadenylated tail, was iso-lated by the customary techniques described in Maniatis et al. (op. cit.). The fragment is hereafter called fragment I.
A synthetic HindIII-IrcoRI fragment, containing the end og the pre-pro region of the a pheromone (cor responding to the residues Ser Leu Asp Lys Arg) and the start of the sequence coding for the mature protein of gro-f~, was also prepared. The sequence of this frag-ment, called fragment J, is given below. It will be noted that the 'sequence Corresponding to the start or 05 the cDNA op gro-13 has been modified relative to the sequence of the cDNA described by Tekamp-olson et al., op. cit., so that the codons used are among those most frequently used by S. cerPvisiae (q. v. Sharp et al., 1.986, Nucleic Acids Research, vol. 14, 13, 5125-5143).
Hind III
AGCTTGGATAAAAGAGCGCCTTTGGCTACTGAATTGAGATGTCAATGTTTGCAAACCTTGCAAGG

EcoRI
Fragments G, H, I and J were ligated to give plasmid pEMR583 shown in Figure 11, in which the 2p symbols have the same meanings as in Figures 7 and lo, fragment E being represented by ~ and fragment F by The nucleotide arid peptide sequences of the start of the mature protein of gro-f3, and of the end of 25 the pre-pro region v~ the a pheromone, are as follows:

G T A.A G C.T T G.G A T.A A A.A G A.~G C G.C C T.T T G.
Val. Ser. Leu. Asp. Lys. Arg. (Ala. Pro. Leu.
05 End of the pre-pro region of the Start of the a pheromone (mature protein Cleavage site of the endopeptidase of cathep-lo sin B type: yscF (product of the KEX2 gene) EXAMPLE 9: Secretion of the cvtokinin aro-l3 by lea=
1 Transformation of the EMY751 yeast strain by plasmid 15 pEMR583 and expression of gro-I3 by the transformed strain. The EMY761 strain (Mata, leu2, ura3, hi~3) described in Example 3 was transformed by plasmid pEMR583 far the prototrophy of leucine by the tech nique already described 1n Example 3). Two trans 2o formed strains, hereafter called EMY767. pEMR583 (1) and EMY761 p~NIR583 (2), were retained.
2 Expression, in an Erlenmeyer flask, of the cDNA og gro-B by the EMY761 pEMR583 (1) and EMY761 pEMR583 (2) strains. Detection of the protein in the culture 25 medium on pol.yacrylamide gel/sodium dodecylsulfate SDS ) according to the protocol described by LAEMtdLz (U. K. LAEMMLI, Nature, 22~ [1970] 680-685).
culture a/ A colony of each of the EMY761 pEMR583 (1) arid 30 EMY761 pEMR583 (2) strains was cultured in 50 ml of uracil-free liquid medium. This medium con-tains the following per liter:
6.7 g of Yeast Nitrogen base without amino acids (from DIFCO) 35 5.0 g of casein hydrolyzate (Casamino acids from .M. - 5 4 DIFCO) g of glucose.
After one night at 30'C, with agitation, the two cultures were centrifuged for 10 min at 7000 rpr~.
05 The residues were taken up in 10 ml of sterile distilled water and centrifuged again for 10 min at 7000 rpm. Expression of gro-D was induced by taking up the cells in 50 ml of ethanol-glycerol-galactose YNB medium. The ethanol-glycerol-10 galactose YNB medium contains the following per liter:
6.7 g of Yeast Nitrogen base without Amino acids (from DIFCO) 5.0 g of casein hydrolyzate (casamlno acids fron DIFCO) 30 g of glycerol 30 g of galactose 10 ml of ethanol The cultures were incubated again at 30'C for 24 h, with agitation.
b/ Control strain:
The non-transformed EMY761 strain, i.e. the EMY761 strain without plasmid, was cultivated as above.
It was subjected on the one hand to preculture in 50 ml of uracil-free liquid medium to which uracil had been added (20 ug/ml), and on the other hand to induction in 50 m1 of ethanol-glycerol-galactose YNB medium to which uracil had been added (20 ug/ml).
Preparation of the samples:
The cells cultivated in 1 a/ and 1 b/ were centri fuged for 20 min at 10,000 rpm and the supernatant was collected. 5 ml of 50$ trichloroaCetic acid containing 2 mg/ml of deoxycholate were added to 10 ml of supernatant.

The mixture was cooled at *4°C for 3o min and then centrifuged for 30 min at 10,000 rpm. The residue was taken up in about 1 ml of cold acetone (T4'C) and centrifuged again for 30 min at 10,000 rpr~.
05 After having been dried, the residue is taken up in about 20 ul of a so-called loading buffer con-sisting of Tris-HC1 0.125 M pH 6.8, SDS 4%, brono-phenol blue 0.002%, glycerol 20%, B-mercapto-ethanol 10% (according to the protocol described by LAEMMLI [1970 0 . The residue is solubilized by boiling for 15 min and then neutralized by the addition of 10 N sodium hydroxide solution until the bromophenol blue turns blue.
The samples are deposited on polyacrylamide gel/
SDS:
1/ Size marker 2/ Non-induced non-transformed EMY761 : 20 ul deposited 3/ Non-induced EMY761 pEMR583 ( 1 ) . 20 ~tldeposited 4/ EMY761 pEMR583 (1) induced for 24 h . 15 ~C1deposited 5/ EtdY~61 pEMR583 induced (1) for 24 h : 5 ul deposited 6/ Size marker 7/ EMY761 pEMR583 (2) induced for 24 h : 5 ~1 deposited 8/ EMY761 pEMR583 (2) induced for 24 h . 15 ~Cldeposited 9/ Non-induced EMY761 pEMR583 (2) . 20 ~1 10/ Induced non-transformed EMY761 : 20 ~1 deposited After electrophoresis, the proteins are stained with Coomassie blue.
$~SULTS:
Analysis of the geI obtained shows that a 05 supernumerary protein with an apparent molecular weigrt of about 8 kDa is produced by the EldY761 pEMR583 (1) strain (lanes 4 and 5) and EMY761 pEMR583 (?.) strain (lanes 7 and 8) and is not produced in the culture supernatants of the non-transformed EMY761 strain (lanes 2 and 10). It is also apparent that the syn-thesis of this supernumerary protein is associated with induction of the promoter by growth on ethanol-glycerol-galactose (bands absent in lanes 3 and 9).
Analysis of the amino-terminal sequence of this protein purified by HPLC made it possible to verify that it was the mature gro-B protein described by Tekamp-Olson et al., op. cit.
It is therefore apparent that the cytokinin gro-f3 can be secreted under the control of the promoter of the invention.
The EMY761 pEMR583 (1) strain has been depo-sited at the Institut Pasteur under the number I - 1021.
EXAMPLE 10: Construction of a vector fob the expression and secretion of IL-8 '~ veast~ plasmid pEMR611 carry~~ the artificial pror.!o er og the invention A second example of a secreted protein whose expression can be regulated by the promoter of the in-vention is the human cytokinin IL-8. This cytokinin of about 8000 Da, produced by monocytes, has been des-cribed by several teams: Yoshimura et aI. (1987), J.
Immunol., 139, 788-793, ShrtSder et al. (1987), J.
Immunol., 139, 3474-3483, and Walz et al. (1987), ~io-chem-Hiophys. Res. Gommun., 149, 755-761. IL-8 acts as a chemical attractant of neutrophils. IL-8 has remar-- 57 - 2fl'~~'~~
kable structural similarities with l3-thromboglobin (Van Damme et al. (1989), Eur. J. Biochem. 181, 337-344).
The cytokinin IL-8 exists in several forms which differ from one another in their NHS-terminal end. The major 05 form is composed of 72 amino acids but 5 other minor forms are also produced, 3 of which have a truncated NHa-terminal end compared with the major form, and 2 of which have respective extensions of 5 and 1 amino acids compared with this major form.
The cDNA of IL-8 was cloned and sequenced according to the method described by ?datsushima et al., 1988, J. Exp. Med. 167, 1883-1993. The sequence of this cDNA contains a single HindIII site (5'-AAGCTT) which covers the 42nd and 43rd codons of the mature part (form 72 aa: q.v. Matsushima et al., op. cit.).
Furthermore, the clone of the cDNA of~IL-8 used in this cloning has a single BamHI site directly flanking the end (3') of the polyA tail. The BamHI-HindIII fragment, carrying the 3' end of the cDNA of IL-8, was purified.
A cloning vector was prepared in the following manner:
Plasmid pEMR583, described in Example 8, was digested with HindIII and BamHI.
This double digestion releases 5 fragments: the heaviest fragment, IiiridIII-BamHI, corresponds to the cloning vector (about 7760 base pairs). The other 4, HindIII-HindIII (with sizes of 169 base pairs, 92 base pairs, 529 base pairs and 187 base pairs), correspond to the sequence of the cDNA of gro-J3. The HindIII site of the cloning vector is located slightly upstream from the insertion site at the end of the pro sequence of the alpha pheromone (and covers the sequence of the codon - serine 81 - of the precursor). The BamHI site is located upstream from the terminator of the PGK.

-5a- 2~, The DNA of the HindTII-BamHI fragment was purified. The HindIII-BamHI fragment, containing the 3' part of the sequence of the cDNA, was ligated with a synthetic HindIII-HindIIZ DNA fragment. The sequence 05 of this fragment, which is given below, is intended for reconstituting the sequence of the major mature form of IL-8 (72 amino acids), preceded by the sequence of the Cleavage site (Ser-Leu-Asp-Lys-Arg). The novel HindIII-BamHI fragment was ligated with the HindIII-l0 BamHI fragment corresponding to the cloning vector.
The sequence of the synthetic HindIII-HindIII fragment is as follows:
5'-AGCTTGGaTAAAAG~TCTGCT~GG~ATTG~GATGTCAATGT~TC~G~.CTTACTC:~GCC~TTCC
1 5 ACCTAT:'TTCTAG~CG~TTCCTTAACTCTACAGTTAC~T~GT:C:G~TG~G~TTCCGT~1GG
dCCCrI~G:T. CnTC~CGc'lciTTGaG~GTTc'~TCG~ATCTGGTCC~ICaCTGTGCTr~CaCTGr~c'~r~TTA~
T CG G : ii T C~rIGTr'~G TTC~, T'"lAAC T CTCc~ATc'~G CTTc'1G?~C
Cc~GGTGTGr~Cc~CGc~T i GTGnC i ~TdAT~1 GC~TTCGa- 5' The plasmid obtained in this way, carrying the cDNA of IL-8 preceded by the pre-pro sequence of the a 25 pheromone and the promoter of the invention, whose sequence was specified in Example 2, is called pEt~R611.
~,~,I~PLE 11 : Tr ansformation of the EMY761 yeast strain by plasm ~,gEMR611, and secretio of IL-8 The EMY76I strain (Mata, ura3, leu2, his3), 30 described in Example 3, was transformed into the (leu-~) strain by the DNA of plasmid pEMR611 according to the technique already described. Of the (leu~) colonies obtained, one was removed at random and cultivated in order to study the secretion of rL-e. This strain is 35 hereafter called EMY761 pEMR611.

The protocol for analysis of the proteins sec-reted by the yeast is that described in Example 9. The proteins secreted by EMY761 pEMR611 were compared with those secreted by EMY761. This revealed a superriu-05 merary mayor band corresponding to a protein of 8000 Da secreted by the EMY761 pEMR611 cells induced by galac-tose. This protein is specifically recognized by rabbit antibodies directed against human IL-8 (supplied by Endogen: Anti human IL-8 polyvalent P801) according to the results of analysis by Western blot - a method which is well known to those skilled in the art and whose protocol is given in detail in Example 4.
Expression plasmid pEMR611 therefore per:aits a high level of expression of IL-8 in transformed yer~sts.
This expression is under the control of the crt~~icial promoter of the invention. ' The EMY761 pEMR611 strain has been deposited in the Institut Pasteur collection under the number I-1023.
EXP.MPLE, 12: Cons ruction of yector for the expr~on and secretion of iruain: pl, ~ml4c~pE2~T?5~c -~; ncr the ar~~ fic~ aI p r~motcr F the invention Naturally produced by the leech, rirudin is a very specific and very effective inhibitor of thrombin. A number of variants have been identified and designated by HV1, HVz and HV~ (Dodt J. et al.
(1986), FEBS Lett. ~, 373, 377). Some of these natural variants and other analogs have subsequently been prepared by genetic engineering in a variety of host cells. The present Example concerns the variant rHV~-Lys47 described in the patent publication EP-A-0273800. More particularly, the expressed sequence codes for a precursor of this variant rHVs-Lys47, which contains a signal sequence: Met - Arg - Phe - Ser -Thr - Thr - Val - Ala - Thr - Ala - Ala - Tyr - Ala -20~~'~~
Leu - Phe - Phe - Thr - Ala - Ser - Gln - val - Ser -Ala, directly preceding the start of the sequence of mature hirudin. .
The construction and the structure of this pre 05 cursor are described in the patent publication FR
2646437. The expression of this precursor permits the release of the variant rHV2-Lys47 in the culture super natant of the transformed cells.
Plasmid pTG3867, whose construction and to description are given in detail in the patent application FR 2 646 437, is a secretion vector for hirudin. In this construction, the hirudin is synthesized in the form of a precursor containing a signal sequence. The hirudin is placed behind the 15 promoter of the MFal gene, which is a constitutive promoter in yeast strains or the a conjugation type.
Construction of p~a5mid pEMR~47 Plasmid pEt~IR547 is derived fzom plasmid pt;MR515 (cf. Example 2) by deletion of the small sequence ori 20 ginating from plasmid 2~ and located between the end of the LEU2d gene and the EcoRI site bordering the se-quence of plasmid pBR322. Plasmid pEMR515 was digested partially with BsptdI and totally with EcoRI. The large BspMI-EcoRI fragment, corresponding to plasmid pEtrLR515 25 from which the 3' part of LEU2 and the small adjoining sequence of the 2u fragment have been deleted, was ligated with a synthetic Bspbtl-EcoRI° sequence intended for reconstituting the 3' region Of the LEU2d gene.
This synthetic sequence is as follows:

- sl - 2U~6"~6 EcoRI° AATTGCCCGGGACGTCTTRTGTACAAATATCATAAAAAAAGAGAATCTTTTT
+_________+_________+_________+_________+_________ ~CGGGCCCTGCRGAATACATGTTTATRGTATTTTTTTCTCTTAGaAaaA
AAGCAAGGATTTTCTTRACTTCTTCGGCGACRGCATCRCCGACTTCGGTGGTACTGTTGG
+_________+_________+_________+_________+_________+_________ 05 TTCGTTCCTAAAAGAATTGAAGAAGCCGCTGTCGTAGTGGCTGAAGCCACCATGr'1CAACC
B
s P
M
I
AACCACCTRAATCACCAGTTCTGATACCTGCRTCC
+_________+_________+_________.________ TTGGTGGATTTAGTGGTCAAGACTATGGACGTAGGTTTT
The reconstituted plasmid is called pEt~iR547.
Co struction of g],.asr~id pEMR5~8 Plasmid pEMR568 is derived from pEMR547 in the following manner:
The DNA of plasmid pEMR547 was digested with Mlul and Nhel. This double digestion makes it possible to linearize plasmid pEMR547 (6.6 kb), the 2 sites MluI
and NheI being situated within a few base pairs of one another. A double-stranded synthetic oligonucleotide of the sequence GCTCGAGTTCGAATGCGC
was inserted between these 2 sites.
The resulting plasmid is called pEMR568.
Construe ion of ,plasm'~pEMR576, an a nressi~Q, asmid for hirudin The sequence of the variant rI-iv2-Lys47 of hirudin was obtained from plasmid pTG3867, whose con-struction and description are given in detail in the patent application FR 2 646 437. In this construction, the hirudin is synthesized in the form of a precursor composed of a signal peptide of 23 amino acids (inclu-ding the methionine corresponding to the initiation codon). The messenger RNA coding for the precursor is 05 transcribed with the aid of the MFal promoter, which is a constitutive promoter in yeast strains of the alpha conjugation type.
AccI-SalI double digestion of plasmid pTG3867 releases several fragments, the shortest of which, numbering about 200 base pairs, is readily puri~iable on 2~ agarose gel. The AccI site which borders this fragment is located a few base pairs from the star of the mature sequence, according to the sequence AccI
ATTACGITA TACAGAC...
T.yATGC ATlATGTCTG
~0 Mature sequence The SalI site is located downstream from the stop codon of the coding sequence of hirudin.
The AccI-SalI fragment of 200 base pairs carries the information for the greater part of the mature sequence of hirudin. The complementary infor-mation (signal sequence and start of the mature sequence) is provided by a synthetic sequence o~ about 90 nucleotides, which is specified below:

- 63 - ~~
CGATATACACAATGCGTTTCTCTACTACAGTCGCTACTGCAGCTACTGCGCTATTTTTCACAGCCTCC
TATATGTGTTACGCAAAGAGATGATGTCAGCGATGACGTCGATGACGCGnTAAAAAGTGTCCGr~CG
CCAAGTTTCAGCTATTACGT

Vector pEMR468 was linearized with Sall and partially digested with ClaZ. The Clal-SalI fragment of about 5.6 kb, corresponding to this vector from which the sequence of urate oxidase has been deleted, was ligated with the AccI-Sall fragment of about 200 base pairs, which carries the information for the sequence of mature hirudin, and with the small syn-thetic ClaI-ACCI sequence intended for reconstituting the sequence of the precursor of hirudin. The resul-ting plasmid is called pEMR576.
EXAMPLE 13: Cor'rPYion of h'_ruc~in 1) Transformation of the EMY761 strain by plasmid pEMR576 2p The EMY761 strain (Mats, ura3, his3, leu2) was transformed into the (leu+) strain by plasmid pEMR576 according to the technique already described.
A transformed strain, hereafter called EMY761 pEMR576, was isolated.
2) Expression of hirudin As a negative control, a (leu~) strain derived from EMY761, hereafter called EMY761 (leu~), was con-structed. The technique used is described in detail in the patent application FR 2 646 437. The EMY761 pEMR576 and EMY761 (leu+) strains were Cultivated in parallel in the manner described below:
The precultures take place in medium of the following composition: Yeast Nitrogen base (Difco) 0.7~, histidine 50 ~1g/ml and uracil 50 ~g/ml. After 24 h, the cultuxes are inoculated with l0a cells in medium having the following composition: Yeast nitrogen Base (Difco) 0.7$, ethanol 1~, casamino acids 0.5~, uracil 100 ug/ml_, glycerol 3~ and galactose 1% . After culture for 72 h in the latter medium, the supernatant 05 is separated from the cells by filtration on 0.2 u.
The inhibitory activity of the supernatant on t!~.ro.::bin is measured by using the colorimetric test described in FR 2 646 437 (proteolytic activity of thrombin on a synthetic substrate: chromozyme TH marketed by Boehringer Mannheim).
The Table below shows the results of the assays in ~g of hirudin per ml of supernatant, at an optical density of 1, i.e. 0.3 x 10' cells/ml.
1~

Strain I ug/ml EMY761 (leu~) non-detectable EMY761 pF..MR576 0.5 It is therefore apparent that hirudin can be secreted under the control of tha promoter of the invention.
The EMY761 pEMR576 strain has been deposited in y5 the CNC2-i under the number I - 1022.

Claims (10)

The embodiments of the invention in which an exclusive property or privilege is claimed are defined as follows:

1. An artificial promoter for the expression of proteins in yeast, which comprises:
- a sub-sequence upstream from the TATA component of the sequence of the promoter of the GAL7 gene of Saccharomyces cerevisiae, which comprises the upstream activation sequences UAS1 and UAS2; and - a sub-sequence of the sequence of an ADH2 promoter comprising the TATA component and the transcription initiation region.

2. A promoter according to claim 1, wherein the subsequence of the sequence of the GAL7 promoter of Saccharomyces cerevisiae is the following sequence:

3. A promoter according to claim 1 or claim 2, wherein the sub-sequence of the sequence of an ADH2 promoter comprising the TATA component and the initiation region is the following sequence:

4. A promoter according to any one of claims 1 to 3, which comprises the following sequence:

5. An expression vector for yeast, carrying a gene of interest with the means necessary for its expression, its replication and the selection of transformed cells, wherein this gene of interest is under the control of the promoter according to any one of claims 1 to 4.

6. An expression vector according to claim 5, wherein the gene of interest is a recombinant gene coding for a protein which is toxic to yeast.

7. An expression vector according to claim 5, wherein the gene of interest is a recombinant gene coding for urate oxidase.

8. A strain of yeast which is transformed by an expression vector according to claim 6 or claim 7

9. A strain of Saccharomyces cerevisiae which is transformed by an expression vector according to any one of claims 5 to 7.

10. A method of producing a protein of interest, which comprises culturing a strain of Saccharomyces cerevisiae according to claim 9, in the presence of galactose.