PH26266A - Yeast hybrid vectors and their use for the production of polypeptides - Google Patents

Description

Field of the invention

The invention relates to DNA fragments containing the promoters of the yeast acid phosphatase genes, hybrid vectors containing said promoters capable of transforming yeast cells and yeast cells transformed with said hybrid vectors. The invention also provides processes for the prep- "aration of said DNA fragments, said hybrid vectors and said yeast cells, wherein recombinant DNA technology is applied. Furthermore, the in- vention concerns a process for the manufacture of polypeptides which are encoded by gene inserts in said hybrid vectors, and which are useful in the treatment of human and animal diseases, and of derivatives thereof.

Background of the invention

With the development of recombinant DNA technology, the controlled microbial production of useful polypeptides, especially such of med- ical interest, has become possible. Most of the recent work with re- combinant DNA technology concerns prokaryotic organisms. Methods have been elaborated and are now well established to introduce into these organ- isms DNA which codes for eukaryotic proteins. Several bacterial species, especially strains of Escherichia coli, which have been mod- ified by this new technology, are now available and permit the com- mercial production of polypeptides of utmost importance, such as insulin, human leukocyte and fibroblast interferon and human growth hormone.

- 3 =~ 26266

However, for many purposes it will be desirable or necessary in future to use cukaryotic systems in the commercial preparation of proteins, especially of pharmacologically important proteins. Since yeasts are cukaryotes they share many biological pathways with other cukaryotes, most importantly with mammalian cells. As many pharmacolo- gically important proteins arc synthesized by mammalian cells, the related- ness of the two systems can be advantageous. For example, the secre- tory pathway of yeast resembles that of higher animal cells and it is known that yeast cells have the machinery for the cleavage of signal sequences (uncharged N-terminal part of a protein, usually split off during the secretory transport) (47) . Associated with the secretory pathway is the glycosylation system, The basic steps leading to glyco= : sylated proteins are similar in all eukaryotes and it is expected that yeast cells, contrary to prokaryotic cells, can produce proteins which are faithfully glycosylated (although some final steps in the pathway will have to be modified).

Moreover, yeast cells are free of endotoxins. Contaminating endotoxins are often found in protein preparations from E. coli and have to be removed through expensive purification steps. . :

Since yeast is a microorganism, yeast cells are easy to cultivate.

The cell mass obtainable per volume of culture fluid is considerably higher for yeast than for E. coli. In addition, the fermentational behaviour of yeast is well understood and conditions for large scale fermentations are already established.

In the last few years, baker's yeast, Saccharomyces cerevisiae, has received increasing attention among molecular biologists from basic and applied research areas. To a laurie extent, this development is due to the establishment of a transformation system (Hinnen et al. (1); Beggs (2)) which allows this microorganism to be used for genetic manipulations, such as introduction and cloning of heterolo- ~~

BAD GRIGINAL J

~~ gous DNA. Similac to the prokaryotic systems, plasmids are the pre- terved vactors used to transform yeast cells, i.e. to introduce re- combinant DNA into yeast cells.

There are various patent applications and other publications which relate to vectors, suitable for transforming yeast cells, yeasts trans- formed with said plasmids, polypeptides produced by said transformed yeasts and processes for the production thereof:

The general yeast transformation protocol is disclosed by Hinnen et al. (1), Beggs (2), Hicks et al. (3) and Struhl et al. (4). ;

Expression of a human interferon gene linked to DNA fragments of the 5'-flanking sequences of the Saccharomyces cerevisiae alcohol dehydrogenase 1 (ADHI) gene in a plasmid and transformed into yeast cells is described by Hitzeman et al. (5).

The transformation of Saccharomyces cerevisiae with a plasmid contain- ing the chromosomal rabbit g-globin gene is reported by Beggs et al. (6). As set forth in the publication, the gene is incorrectly transcribed and no splicing of the primary p-globin transcripts could be detected.

Eukaryotic cells, such as yeast and especially mammalian cells, co- transformed with foreign DNA coding for a polypeptide and linked with an inducible promoter, and with unlinked DNA which permits the iden=- tification of the transformed cells, and a proccss for the production thereof is described in PCT patent application 8102425 (7).

A DNA sequence coding for a eukaryotic replication site, eukaryotic vectors conferring mitotic stability at low copy number and contain- ing a eukaryotic replication site, and yeast cells transformed with said vectors are disclosed in European patent application 48081 (8). . oa ORIGINAL 9

. hy 9 6266

Hybrid DNAs comprising, inter alia, a cukaryotic host autonomously replicating segment, a method for the production thereof and a method for high-frequency transforming eukaryotic cells, e.g. yeast, with said hybrid DNAs is disclosed in European patent application 45573 (9).

Plasmids comprising the ovalbumin gene controlled by the promoter of the Escherichia coli f-lac Z gene and capable of being transformed into yeast cells are described in German Offenlegungsschrift 2923297 (10) and French patent application 2458585 (11).

Hybrid plasmids comprising DNA of a bacterial plasmid, whole or part of the DNA of the yeast 2p plasmid and the yeast URA3 gene and yeasts transformed with said hybrid plasmids are disclosed in

European patent application 11562 (12).

Object of the invention

During the last ycars, there was great progress in the field of genetic engineering, and the first systems using genetically manipulated micro- organisms, especially strains of the enterobacteria Escherichia coli, are now working. However, there exists a need for additional aud im- proved systems, especially eukaryotic systems, such as yeasts, which are suitable for the economic and large-scale production of proteins in industry. At present, various yeast vectors are available for gene cloning. For the efficient expression of foreign genes in yeast structural coding sequences will have to be combined with strong yeast promoters which, advantageously, should show regulatory features which would allow exogenous control of gene expression. It is an object of the present invention to provide yeast promoters meeting these re- quirements. It is also an object of the present invention to provide lybrid vectors containing said promoters and foreign structural genes controlled by said promoters.

N

BAD ORIGINAL 9

Detailed description of the invention

L. DNA fragments containing yeast acid phosphatase a promoters and their preparation ) 0 26 6 ,

The present invention provides newly isolated yeast promoters having im- proved :xpression properties and a process for the production thereof.

The yeast promoters according to the present invention are derived from the genomic DNA of yeast, especially of Saccharomyces cerevisiae.

At least two structural genes (PHO3 and PHOS) and several regulatory genes (PHO2, PHo4, rioso, PHOB1, PHO85) are involved in the expression of acid phosphatase in yeast (for reference, see, for example, (13)).

PHOS and PHO3 code for a repressible (regulated) and a constitutive yeast acid phosphatase, respectively. The PHOS gene is repressed at high concentrations of inorganic phosphate and turned on (derepressed) under inorganic phosphate starvation (usually to a high extent under appropriate physiological conditions), whereas the PHO3 gene is ex~ presscd constitutively at lew levels. The repressible enzyme is glyco- sylated and has a molecular weight of about 490 Kilodaltons (14). .

The promoters controlling the acid phosphatase genes have not been isolated or used in prior art recombinant DNA ‘technology: and. hence : 4 their nucleotide sequences have not been elucidated. In contrast to Co other yeast promoters used in recent recombinant DNA technology (e.g. Ca

ADM1), the DNA sequences directly following the yeast acid phosphatase ! promoters code for signal peptides which are thought to be involved in the , secretion process. It would be advantageous to link a foreign protein . coding region to a yeast signal sequence ensuring in vivo transport of the ¥ protein across the yeast cell membrane. This would result in a reduction . of product degradation and contamination of the product by host cell material and would facilitate product recovery. : [BAD ORIGINAL I -

- 7 ~— RB 26206

It is a disadvantage of the promoters hitherto used in recombinant

DNA technology that the respective genes are transcribed constitu- tively. The expressed polypeptide may be either toxic to the yeast cell (fungicidal activity) or may at least inhibit cell proliferation (fungistatic activity), or the polypeptide may be enzymatically digested within the cell, especially if it is exposed to yeast prote- ases for a long time. In all cases mentioned, the yicld of the desired polypeptide would be low. These disadvantages can be avoided by using the PHOS promoter and vectors containing said promoter. The PHOS promoter can be repressed or turned on (derepressed) at the will of the experimentator, solely by increasing or decreasing the concentra-: tion of inorganic phosphate in the medium. Thus, the promoter can be repressed during the exponential growth phase of the yeast and may be turned on only during early stationary phase at maximal cell density allowing expression of the gene controlled by the PHOS promoter. This property combined with a high level of transcription makes the nos promoter the preferred ome in the present invention.

The present invention relates especially to a DNA fragment comprising a yeast acid phosphatase promoter, such as the PHO3 promoter or, . preferably, the PIIO5 promoter, and flanking sequences.

Optionally, the yeast acid phosphatase promoter is followed by all or part of the signal sequence of the yeast acid, phosphatase’ coding region naturally linked to said promoter. In addition, said DNA RE : fragment may contain sequences which are required for efficient translation of mRNA. Also enclosed are those mutants of said DNA fragment which retain the promoter function. ‘

A DNA fragment according to the invention may be prepared, for example, by (A) preparing an acid phosphatase gene by complementing an acid phos- phatase deficient yeast strain by transformation with plasmid

DNA from a ycast gene library containing the wild-type copy of said gene and isolating said gene, : { oe SAD ORIGINAL 9d oo 26266 (B) preparing subclones of the obtained gene, and (C) identifying the location of the promoter region of the above subclones and isolating DNA fragments comprising the acid phosphatase promoter,

More especially, the following steps are involved in the preparation of said DNA fragment; (1) A yeast gene library is constructed using wild-type yeast DNA cloned into a hybrid bacterial (especially Escherichia coli)-yeast plasmid carrying appropriate markers capable of expression in both the bacterial and yeast cell (for suitable markers, see below). (2) Clones containing a yeast acid phosphatase gene are selected by transformation of an acid phosphatase deficient yeast strain using plasmid pools of the above library. . (3) Plasmids containing an acid phosphatase gene are isolated from the transformed yeast and amplified by transforming back into E. coli selecting for the phenotypic property of the bacterial marker (e.g. ampicillin resistance). ’ (2') In an alternative approach the gene library is divided into sub- pools which are used to transform acid’ phosphatase ‘deficient yeast strains, and (3') positive sub-pools are again sub-divided and trans- formed as above until a single clone is identified. ’ (4) The plasmid DNA of the identified clone is isolated, digested with suitable restriction endonucleases and the fragments are recloned into an appropriate yeast vector. (5) DNA fragments containing a.yeast acid phosphatase gene can be identified by transforming yeast acid phosphatase deficient yeast strains vith said vectors. By means of this procedure the boundaries of the acid phosphatase genes can be determined with a precision of approximately 300 base pairs.

BAD ORIGINAL 9

(6) DNA sequencing of the identified fragments serves to locate the promoter regions, the acid phosphatase protein coding regions and, additionally, the restriction site(s) which may be useful in further processing, for example, for cutting off DNA sequences which are not neccessary for promoter function, with restriction endonucleases.

Depending on the choice of the restriction endonucleases, the DNA frag- ments containing the acid phosphatase promoter may also include at the : 3' and 5' termini original flanking DNA sequences which do not affect the promoter function and may be used as connecting sequences in the sub- sequent cloning procedures. If desired, these additional sequences can be shortened by digestion with a restriction endonuclease (if possible) or with a suitable exonuclease, for example Bal3l. In addition, the fragments can be ligated to chemically synthesized

DNA linkers which preferably include the recognition sequence of an appropriate restriction endonuclease. This allows a convenient connec-— tion of the acid phosphatase promoter with foreign polypeptide coding regions. It is also possible to isolate and/or construct a DNA fragment which contains the yeast acid phosphatase promoter and_part or all of the adjacent signal sequence from the acid phosphatase protein coding region. When ligated to an appropriately cut foreign polypeptide coding region, the resulting hybrid DNA will be expressed in yeast to yield polypeptides with acid phosphotase signal sequences or fused signal sequences. of - Ce Co oo

The yeast acid phosphatase promoters according to the present inven- tion may be used to control the expression of a yeast or a non-yeast polypeptide coding region in a yeast hybrid vector. 2. Hybrid vectors containing yeast acid phosphatase promoters and _ their preparation ———

The present invention also relates to hybrid vectors comprising a yeast acid phosphatase promoter and a yeast or a non-yeast polypeptide coding region which is controlled by said promoter. (

BAD ORIGINAL J

TR

UU -— —_— - 10 . : 262606

The terms "vector", "hybrid vector", "DNA sequences' etc. used in the present application relate in particular to double stranded DNAs.

However, single stranded DNAs are also comprised. Vectors and hybrid vectors may be present in lincar or, preferably, circular form.

A yeast acid phosphatase promoter is especially one of those described in chapter 1 and refers preferably to the regulated acid phosphatase promoter PlIOS. ’

The yeast or non-yeast polypeptide coding region (gene) controlled by one of the above promoters may be derived from genomic DNA or from cDNA pre- pared via the mRNA route or may be synthesized chemically. The non-yeast polypeptide coding regions (genes) originate from viruses, prokaryotic cells or eukaryotic cells, including from higher eukaryotic cells, espe- cially from human cells. When expressed in the host yeast cell, these genes can provide for the productionof a wide variety of polypeptides including glycosylated polypeptides, such as enzymes which can be used, for example, for the production of nutrients and for performing enzy- matic reactions in chemistry, or non-enzymatic polypeptides, for example hormones, polypeptides with immunomodulatory, anti-viral and anti-cancer properties, antibodies, viral antigens, vaccines, clotting factors, foodstuffs and the like. For example, such genes code for amylases, proteases, lysozyme, viral thymidine kinase, rennin, «.

B~lactamase, glucose isomerase; secretin, thymosin, relaxin, Co - calcitonin, somatostatin, human or bovine growth hormone, insulin, luteinizing hormone, parathyroid hormone, adrenocorticotropin,

B-endorphin, melanocyte-stimulating hormone, B-lipotropin, urogastrone; interferon, such as human interferon, e.g. a human interferon-a or -B polypeptide derived from human leukocyte, lymphoblastoid or fibroblast cells, or human interferon-y; lymphokines, tumour necrose factor; anti-rennin antibody, hepatitis A virus antigen, hepatitis B virus (HBV) surface or core antigens, hepatitis non-A non-B virus antigen, human histocompatibility antigens, food and mouth disease virus antigen, influenza haemagglutinin, fowl pest virus haemag-

BAD ORIGINAL 9

- 11 ~- ' 6266 glutinin; serum albumin, ovalbumin, thaumatin, 2 or plasminogen activators.

A chosen polypeptide coding region may optionally include a signal sequence or a part thereof. As indicated above, this can give rise to a fused protein containing the PHOS signal sequence or a hybrid signal sequence containing part of the PHOS signal sequence and part of the signal sequence of the foreign polypeptide together with the foreign mature polypeptide. In both instances, those combinations are favoured which lead to the cleavage of the signal sequence upon

LU maturation of the foreign polypeptide.

Apart from an acid phosphatase promoter and a yeast or a non-yeast poly- peptide coding region, the hybrid vectors according to the invention may contain additional DNA sequence(s) which are inessential or less im- portant for the function of the promoter, i.e. for the expression of the polypeptide coding region, but whichmay perform important functions, for example, in the propagation of the yeast cells transformed with said hybrid vectors. The additional DNA sequence (s) may be derived from prokaryotic and/or eukaryotic cells and may include chromosomal “and/or © extra-chromosomal DNA sequences. For example, the additional DNA scquences may stem from (or consist of) plasmid DNA, such as bacterial or cukaryotic plasmid DNA, viral DNA and/or chromosomal DNA, such as bacterial, yeast or higher eukaryotic chromgsonial DNA. Preferred hybrid vectors contain additional DNA sequences derived from bacterial plasmids, especially Escherichia coli plasmid pBR322 or related plasmids, bacteriophage As yeast 2p plasmid, and/or yeast chromosomal

DNA. ‘

Preferably, the additional DNA sequences carry a yeast replication ‘ origin and a selective genetic marker for ycast. Hybrid vectors con- taining a yeast replication origin, e.g. the chromosomal autonomously replicating segment (ars), are extrachromosomally maintained within the yeast cell after transformation and are autonomously replicated upon lps mom J

' 26266 mitosis. llybrid vectors containing scquences homologous to yeast 2p plasmid DNA can be used as well. These hybrid vectors will get integrated by recombination into 2p plasmids already present within the cell or will replicate autonomously. 2p sequences are especially suitable for high-frequency transformation plasmids and can give rise to high copy numbers.

In addition, the hybrid vectors according to the invention may include a DNA sequence of a gene present in the host yeast chromosome (e.g. ros), the promoter of which may be linked to the yeast or non-yeast polypeptide coding region. By virtue of the homologous sequence the whole vector can be stably introduced into the host chromosome by recombination. Thus, during propagation the progeny cells will retain the introduced genetic material even without selective pressure.

As to the selective gene marker for yeast, any marker gene can be used which facilitates the selection for transformants due to the pheno- typic expression of the marker. Suitable markers for yeast are - particularly those expressing antibiotic resistance or, in the case of auxotrophic yeast mutants, genes which complement host lesions.

Corresponding genes confer, for cxample, resistance to the antibiotic cycloheximide or provide for prototrophy in an auxotrophic yeast mutant, for example the URA3, LEU2, 1IS3 or TRP1 gene. It is also possible to employ as markers structural genes which are associated with an autonomously replicating segment providing that the host to be transformed is auxotrophic for the product expressed by the marker.

Advantageously, the additional DNA sequences which are present in the hybrid vectors according to the invention may also include a replica- tion origin and a selective genetic marker for a bacterial host, especially

Escherichia coli. There are useful features which are associated with the presence of an E. coli replication origin and an E. coli marker

BAD ORIGINAL 9 or 26266 in a yeast hybrid vector: Firstly, large amounts of hybrid vector

DNA can be obtained by growth and amplification in E. coli and, secondly, the construction of hybrid vectors is conveniently done in

E. coli making use of the whole repertoire of cloning technology based on E. coli. E. coli plasmids, guch as pBR3I22: and the like; con- tain both E.coli replicationorigin and E.coli genetic markers. confer- - ring resistance to antibiotics, for example tetracycline and ampicillin, and are advantageously employed as part of the yeast hybrid vectors. }

The additional DNA sequences which contain, for example, replication : origin and genetic markers for yeast and a bacterial host (see above) are hereinafter referred to as "vector DNA" which together with the acid phosphatase promoter and the yeast or non-yeast polypeptide coding region is forming a hybrid vector according to the invention.

The hybrid vectors can be prepared by methods known in the art, for example by introducing into a vector DNA a yeast acid phosphatase promoter and a yeast or a non-yeast polypeptide coding region which is controlled by said promoter.

Conveniently mapped linear or, preferably, circular vector DNA, for cxample bacterial plasmid DNA or the like (see above), having at lcast one restriction site, preferably two or more restriction sites, can be employed. Advantageously, the vector DNA already contains replication origins and gene markers for yeast and/or a bacterial host.

The vector DNA is cleaved using an appropriate restriction endonuclease,

The restricted DNA is ligated to the DNA fragment containing the acid phosphatase promoter and to the DNA segment coding for a yeast or non- yeast polypeptide. Prior to or after linking of the promoter and the polypeptide coding region (or simultaneously as well), it is also possible to introduce replication origins and/or markers for yeast or a bacterial host. At all events, the restriction and annealing conditions are to be chosen in such a manner that there is no interference with v (54D ORIGINAL 9 ——

the essential functions of the vector DNA and of the promoter. The hybrid vector may be built up sequentially or by ligating two DNA segments comprising all sequences of interest.

Various techniques may be used to join DNA segments in vitro.

Blunt ends (fully base-paired DNA duplexes) produced by certain restriction endonucleases may be directly ligated with T4 DNA ligase.

More usually, DNA segments are linked through their single-stranded cohesive ends and covalently closed by a DNA ligase, e.g. T4 DNA ligase. Such single-stranded "cohesive termini'' may be formed by cleav- ing DNA with another class of endonucleases which produce staggered ends (the two strands of the DNA duplex are cleaved at different points at a distance of a few nucleotides). Single strands can also be formed by the addition of nucleotides to blunt ends or staggered ends using terminal transferase ("homopolymeric tailing') or by simply chewing back one strand of a blunt-ended DNA segment with a suitable exonuclease, such as A —exonuclease. A further approach to the pro- duction of staggered ends consists in ligating to the blunt-ended DNA segment a chemically synthesized linker DNA which contains a recogni- tion site for a staggered-end forming endonuclease and digesting the resulting DNA with the respective endonuclease. : Co

In order to be efficiently expressed,’the gene aoding for a yeast or a non-yeast protein must be properly located with respect to sequences a containing transcriptional (acid phosphatase promoter) and transla-. tional functions (ribosome binding sites). Firstly, the ligation of the

DNA scgment comprising the promoter with the polypeptide coding region has to be achieved in the proper orientation. If two orientations are possible the correct one can be determined by conventional restriction analysis. Hybrid vectors containing an incorrectly oriented gene insert can be re-oriented by excising the gene insert with a suitable : (BAD ORIGINA, 9

T——

~ 15 - restriction endonuclease and re-ligating the 26.26.06. vector fragment. In any case improper orientation can be avoided by ligating two DNA segments cach with different restriction sites at their ends. Furthermore, the construction of the hybrid vector should be done in such a way that it allows correct transcription initiation and termination. As to the latter point, the transcript should preferably end in a DNA sequence derived from yeast chromosomal

DNA or yeast 2p plasmid. Advantageously, the transcript ends in a

DNA sequence containing transcription termination signals of a yeast gene, e.g. of PHOS or PHO3. Secondly, a proper reading frame must be established. Ordinarily, the nucleotide sequence of both pro- moter region and polypeptide coding region is known prior to ligation or can easily be determined (e.g. (15)) so that there are no problems in establishing the correct reading frame. In addition, specific second- ary DNA structures might be needed for even more efficient expression of the gene,

A preferred region for joining the acid phosphatase promoter to a foreign coding sequence is between the major acid phosphatase mRNA start and the ATG of the acid phosphatase coding region, for example, when using the PHOS promoter, within a stretch of about 40 bp between the major PHO5 mRNA start and the ATG of the PHO5 acid phos- phatase coding region. For a junction in this region the foreign coding sequence should have its own ‘ATG' for trans?ation initiation, or else it has to be provided by an additional synthetic oligonucleo- tide.

Since many polypeptides of higher organisms are primarily expressed as pre-polypeptides consisting of signal peptides attached to the

N-termini of the mature polypeptides, it may be useful to include a signal sequence in the gene insert. Suitable signal sequences are those naturally linked to the polypeptide gene to be expressed or to the acid phosphatase promoter. Alternatively, fused signal sequences may be constructed by ligating part of the acid phosphatase signal sequence with part of the polypeptide signal sequence. If the v \ ao ORIGINAL JP direct expression of a mature polypeptide is desired, signal sequences or parts thereof optionally following the promoter region or optionally preceding the mature polypeptide coding region have to be eliminated, for example by digestion with an exonuclease, e.g. with Bal3l.

Intermediate products, such as vectors still lacking one or more essential Functions, as well as the final hybrid vectors according to the invention may be transformed into a bacterial host, especially

E. coli, for the above rcasons (e.g. production of large amounts of intermediate products and hybrid plasmids, respectively). Bacterial 19 vectors, such as the E. coli plasmid pBR322 and those fragments : thereof which contain a bacterial replication origin and gene marker (s) arc the most preferred vectors for that reason. When using such a bacterial vector, the final steps for the preparation of the yeast hybrid vectors preferably also include the introduction of a genetic marker and a replication origin for yeast.

DNA segments, which may be inserted into the bacterial vector in order to produce the hybrid vectors according to the invention, such as an autonomously replicating segment (ars, cf. (4)), sequences of yeast 2p plasmid (2) or yeast marker DNA (cf. 16), can be isolated from yeast chromosomal DNA and yeast 2p plasmid DNA, respectively, in a conventional manner. The geéne.coding for a yeastor anon-yeast polypeptide may be isolated from chromosomal or extrachromosomal DNA, derived from cDNA prepared via the mRNA route (see above) using con- ventional techniques (e.g. 17, 18) or may be synthesized chemically.

In a preferred embodiment of the invention, the method for the preparation of the hybrid vectors comprises the steps of (1) constructing a yeast gene library using wild-type yeast DNA, (2) isolating the acid phosphatase gene, especially the PHO5 gene, and cloning it into a bacterial plasmid, such as pBR322, or a biolugically functional, in particular an intact replication origin and selection marker containing fragment thereof, “3 N- »

BAD Orit :

CRE

-17 - CL 26266 (3) inserting into said plasmid a genetic marker for yeast, such as the TRP1 gene, and a yeast replicationorigin, such as a chromosomal autonomously replicating segment or alternatively yeast 2p plasmid sequences into an appropriate restriction site, 9 (4) inserting a DNA segment coding for a yeast or a non-yeast poly- peptide, such as human interferon or HBV surface antigen, in such a manner that the acid phosphatase promoter controls said poly- peptide coding segment, and (5) optionally inserting a DNA sequence containing transcription termination signals of a yeast gene, e.g. of PHOS, downstream from the polypeptide coding region.

It is likewise possible to alter the order of steps, such as steps 3 to 5, for example, by first introducing the polypeptide coding segment and subsequently inserting the genetic marker and the replication origin for yeast into the recombinant plasmid obtained as a product of step 2.

Prior to inserting a gene marker for yeast, a yeast replication origin and a polypeptide coding segment, inessential functions, such as the acid phosphatase structural gene, may optionally be excised from the recombinant plasmid obtained in step 2.

Especially, the DNA segment coding for a yeast or non-yeast poly~ peptide is joined to the aeid phosphatase promoter (step 4) in the i region between the major acid phosphatase mRNA start and the ATG of the acid phosphatase coding region. Optionally, a synthetic linker containing an appropriate restriction site is introduced to allow a junction between said DNA segment and the acid phosphatase promoter. lutermediate hybrid vectors comprising the yeast acid phosphatase promoter and still lacking the yeast or non-yeast polypeptide coding’ sequence are also an object of the present invention and can be prepared by the above successive steps (1), (2), (3) and optionally (5), wherein the acid phosphatase promoter is preferably terminated in the region between the major acid phosphatase mRNA start and the ATG of i I” or ‘BAD ORIGINAL gi hee the acid phosphatase gene and/or, optionally, a synthetic linker con- taining an appropriate restriction site is introduced to allow the insertion of a DNA segment coding for a yeast or non-yeast polypeptide. 3. Transformation of yeast with hybrid vectors containing yeast acid phosphatase promoters

Another aspect of the present invention involves a process for the production of transformed yeast cells capable of producing yeast or non- yeast polypeptides, which processcomprises transforming yeast with any of the hybrid vectors described in chapter 2.

Useful yeasts include species of the genera Saccharomyces, Schizo- saccharomyces, Torulopsis and related genera (cf. (19)), especially strains of Saccharomyces cerevisiae.

The transformation of yeast with the hybrid vectors may be accomplished by procedures known from the literature, e.g. according to the method described by Hinnen et al (1). This method can be divided into three steps: (1) Removal of the yeast cell wall. - : (2) Treatment of the "naked" yeast cells (spheroplasts) with the. transforming DNA in the presence of PEG (polycthyleneglycol) and ca’? ions. (3) Regeneration of the cell wall and selection of .the transformed cells in a solid layer of agar. oo . - referred methods: ad (1): The yeast cell wall is removed enzymatically using various preparations of glucosidases, such as snail gut juices (e.g. clugulased or telicased) or cn yme mixtures obtained from micro- oo organisms (e.g. Zymolyase ®y {4 osmotically stabilized solutions - (e.g. 1 M sorbitol). \ rr ——

- 19 - 9 6 9 ‘ 66 ad (2): The ycast spheroplasts aggregate in the presence of PEG and local fusions of the cytoplasmic membranes are induced. The generation of "fusion-like" conditions is crucial and many transformed yeast cells will become diploid or even triploid during the process of trans-— formation. Procedures which allow selection of fused spheroplasts can be used to enrich for transformants, i.e. transformed cells can easily be screened for among preselected fusion products. ad (3): Since yeast cells without cell wall do not divide the cell wall has to be regenerated. This regeneration is conveniently done by embedding the spheroplasts into agar. For example,molten agar (about 50°C) is mixed with the spheroplasts. Upon cooling the solution to yeast growth temperatures (about 30°C), a solid layer is obtained. This agar layer is to prevent rapid diffusion and loss of essential macromolecules from the spheroplasts and thereby facilitates regeneration of the cell wall. lowever, cell wall regeneration may also be obtained (although at lower efficiencies) by plating the spheroplasts onto the surface of preformed agar layers.

Preferably, the regeneration agar is prepared in a way to allow : regeneration and selection of transformed cells at the same time.

Since yeast genes coding for enzymes of amino acid biosynthetic path- ways are generally used as selective markers (cf, chapter-2), the. . regeneration is preferably performed in yeast minimal oo Co oo medium agar. However, if very high efficiencies of regeneration are required a two step procedure miglit be advantageous: (1) regeneration of the cell wall in a rich complex medium, and (2) selection of the transformed cells by replica plating the cell layer onto selective agar plates. i '5AD ORIGINAL 9g [FUR re 26266

If the hybrid vector does not contain any marker gene the transformed cells can also be identified by means of alternative methods. Such methods include, for example, in situ hybridization with a labeled

DNA fragment homologous to sequences of the hybrid vector (e.g. accord- ing to Hinnen et al. (1)), in situ immunoassays provided that the antibody of the product of the introduced gene is available, or other screening methods which measure gene products encoded by the trans- forming plasmid(s).

Alternatively, the yeast can be co-transformed with a hybrid vector according to the invention and a second vector containing a genetic marker for yeast. If the two different vectors have DNA sequences in common (these can be bacterial sequences present on the vectors), recombination can take place leading to a fused selectable hybrid molecule. .

The invention also relates to yeast hosts transformed with hybrid vectors containing a yeast acid phosphatase promoter and a yeast or a non-ycast polypeptide coding region. | : 4, Cultivation of transformed yeast cells and induction of poly- peptide synthesis

To a varying extent, yeast cells trans formed with autonomously repli- cating plasmids, for example, plasmids containing yeast 2p plasmid

DNA, tend to lose the introduced hybrid plasmid (cf. (16)). :

For this reason, such yeast cells have to be grown under selective conditions, i.e. conditions whic# require the expression of a plasmid- encoded gene for growth, Most selective markers currently in use are genes coding for enzymes of amino acid or purine biosynthesis. This makes it necessary to use synthetic minimal media deficient in the corresponding amino acid or purine base. However, some genes conferring antibiotic resistance may be used as well (e.g. genes conferring resistance to cycloheximide or to the amino-glycoside G 418 (21)). Yeast cells transformed with vectors containing antibiotic resistance genes may

QD

(DAD ORIGINAL 9

- 21 ~ be grown in complex media containing the corresponding antibiotic whereby faster growth rates and higher cell densities can be reached.

Yeast cells transformed with DNA integrating into the chromosomes do not require sclective growth conditions. These transformed cells are sufficiently stable to allow growth without selective pressure. For the above reason, these cells are advantageously grown in complex media.

Yeast cells containing hybrid plasmids with a constitutive acid phos- phatase promoter (e.g. PHOJ) express the yeast or non-yeast protein gene attached to said promoter without induction. However, if the yeast or non-yeast protein gene is under the control of the regulated acid phosphatase promoter PHO5, the composition of the growth medium has to be adapted in order to obtain maximum levels of mRNA transcripts, i.e. the growth medium must contain low concentration of imorganic phosphate for derepression of the PIIO5 promoter. 5. Isolation and purification of the expressed polypeptide

The invention also concerns a method for producing a yeast or a non- yeast polypeptide, such as human interferon or HBV surface antigen, comprising the steps of (1) culturing a yeast strain transformed with a hybrid vector con.’ taining a yeast acid phosphatase promoter and a yeast or a non-yeast polypeptide coding region under appropriate nutrient conditions, and (2) isolating and purifying said polypeptide.

The transformed yeast strains according to the present invention ’ are cultured in a liquid medium containing assimilable sources of carbon and nitrogen and inorganic salts. ooo oman 2)

BAD ORIGINAL . \

. 26266

Various carbon sources ran be used. Examples of preferred caivbon sources are assimilable carbohydrates, such as glucose, maltose, mannitol or lactose, or an acetate, which can be used cither alone or in suitable mixtures. Suitable nitrogen sources include, for example, amino acids, such as casamino acids, peptides and proteins and their degradation products, such as tryptone, peptone or meat extracts, furthermore yeast extract, malt extract, corn steep liquor, as well as ammonium salts, such as ammonium chloride, sulphate or nitrate, which can be used either alone or in suitable mixtures. Inorganic salts which may be used include, for example sulphates, chlorides, phosphates and carbonates of sodium, potassium, magnesium and calcium.

Additionally, the nutrient medium may also contain growth promoting substances and/or substances exerting a selection pressure in order to prevent the loss of the hybrid plasmid. Substances which promote growth include, for example, trace elements, such as iron, zinc, manganese and the like, or individual amino acids. .

If the hybrid plasmid contains a gene conferring resistance to an antibiotic substance, cells containing such a hybrid plasmid will survive in a medium supplemented with the antibiotic substance whereas cells which have lost said hybrid plasmid as weld as contaminating antibiotic-sensitive microorganisms will not. If the hybrid ‘plasmid contains a gene providing for prototrophy in an auxotrophic yeast mutant, e.g. the LEUZ or HIS3 gene, a selection pressure can be exerted by omitting the gene product, such as leucine or histidine, in the nutrient medium. sa ORIGINAL GY

1f the cultured yeast strain has been transformed with a hybrid plasmid containing the regulated acid phosphatase promoter PHO5, the content of inorganic phosphate must be reduced in the nutrient medium after the pre-culture phase in order to ensure maximum levels of mRNA transcripts and, consequently, maximum yields of polypeptides.

The cultivation is carried out employing conventional techniques.

The culturing conditions, such as temperature, pH of the medium and fermentation time are selected in such a way that maximal levels of polypeptides are produced. A chosen yeast strain is preferably grown under aerobic conditions in submerged culture with shaking or stirring : at a temperature of about 25° to 35°C, preferably at about 30°C, at a pH value of from 4 to 8, for example at approximately pH 7, and for about 4 to 20 hours, preferably until maximum yields of polypeptides are reached.

After the transformed yeast cells have been grown to a satisfactdry cell density, the first step for the recovery of the expressed poly- peptide consists in liberating the polypeptide from the cell interior.

In most procedures the cell wall is first removed by enzymatic diges-— tion with glucosidases (cf. section 3). Subsequently, the resulting spheroplasts are treated with detergents, such as Triton. Alterna- [BAD ORIGINAL 9 tively, mechanical forces, such as shearing forces (for example

X-press, French press) or shaking with glass beads, may be used to break cells. The resulting polypeptide mixture can be enriched for the desired polypeptide by conventional means, such as precipitation with ammonium sulphate or trichloroacetic acid, gel electrophoresis, dialysis, chromatography, for example, ion exchange chromatography, size-exclusion chromatography, HPLC or reverse phase HPLC, and the like.

The final purification of the pre~purified product can be achieved, for example, by means of antibody affinity chromatography. In principle, the purification steps (except the lysis of the cells) can be accom- plished according to the method of Staehelin et al. (22) developed for the purification of human leukocyte interferon.

For example, the isolation and purification of the desired polypeptide can be performed using the following steps: (1) lysis of the yeast cells with glucosidase, (2) treatment with a detergent, . (3) removal of most of the non-proteinaceous material by treatment with polyethyleneimine, (4) precipitation of the polypeptides by saturating the solution with ammonium sulphate, Lo Co. Co (5) dialysis in an appropriate buffer mixture, CL Co (6) column chromatography on DEAE-cellulose, (7) affinity chromatography on a monoclonal antibody column, and (8) molecular sizing on a suitable Sephadex “~~ column. r

In order to obtain a sufficiently pure product additional purification steps may turn out to be necessary, e.g. cation or anion exchange chromatography, adsorption on hydroxylapatite,reverse phase HPLC etc.

On the other hand, one or more of the above steps may be omitted if possible, or the order of steps may be altered.

BAD ORIGINAL 5)

In the case where the desired polypeptide is secreted by the yeast cell into the periplasmatic space, a simplified protocol can be used:

The polypeptide may be recovered without cell lysis by enzymatic re- moval of the cell wall or by treatment with chemical agents, e.g. thiol reagents or EDTA, which give rise to cell wall damages permitting the polypeptide. to be released. In the case where the poly- peptide is secrcted into the culture broth, it can be recovered directly therefrom.

The polypeptides obtainable according to the present invention are useful and valuable in the treatment of human and animal diseases or in preventing them (e.g. interferon, HBV surface antigen, etc.) or can be used as foodstuffs, feed, feed additives or in enzymatic reactions (see 2 above). It is to be understood that the production of naturally occurring derivatives of said polypeptides, such as proteolytically cleaved polypeptides and/or glycosylated polypeptides, is also com- prised by the present invention.

The invention concerns furthermore polypeptides and naturally occurring derivatives thereof, whenever prepared according to the methods of the present invention.

The invention concerns also the new polypeptides per se obtainable according to the inventive process.’ ! . - Co

The invention concerns especially the DNA fragments, the hybrid vectors, the transformed yeast, the polypeptides and the processes for their preparation as described in the Examples.

I ——

BAD ORIGINAL ¢}) —

Brief description of the drawings 2 0 20 0

In the following experimental part various embodiments of the present invention are described with reference to the accompanying drawings in which:

Figure 1 is a partial restriction endonuclease map of che plasmids pJIDB207.PHOS5, PHO3 and pBR322PHO5Bam-Sal used as sources of the PHOS gene or for DNA sequencing, respectively.

Figure 2 shows the localization of the PHOS and the PHO3 acid phospha- ’ tase genes within a 5.1 Kb BamHl fragment isolated from a yeast gene library.

Figures 3a and 3b provide the DNA sequences of the promoter region of

PHO5 and PHO3, respectively.

Figure 4 is a schematic diagram showing the construction of the plasmids p30IFN2(8)) and P3OIFN2' (8). ‘

Figure 5 illustrates the ligation of the PHO5 promoter DNA with the.

IFN-8) cDNA in the construction of plasmid p30IFN1(8;).

Figure 6 schematically illustrates the .construction.of plasmid = pJDB207/1FN2' (81). : oo

Figure 7 is a schematic outline of the construction of recombinant

DNA molecules containing MNamalwa cDNA. © Figure 8 schematically illustrates the techniques used to synthesize the IFN mRNA specific 13mer DNA primer.

Figure 9 is a schematic diagram showing the identification of clones containing human lymphoblastoid IFN cDNA. [BAD ORIGINAL 9 ved 26266 i

Figure 10 to 14 provide the DNA and corresponding amino acid : sequences of the cDNA inserts of the plasmids CC-pBR322 HLycIFN-1'b, - i; _a' -5

By ty 8, and 3,

Figure 15 depicts the construction of the plasmid CG-pBR(AP) /LyIFN-a-1 and figure 16 shows the DNA and the amino acid sequences of its cDNA insert.

Figure 17 depicts the construction of the plasmid CG-pBR(AP) /LyIFN-a-3 and figure 18 shows the DNA and the amino acid sequences of its cDNA insert.

Figure 19 shows the DNA and the amino acid sequences of the cDNA insert of the plasmid CG-pBR(AP)/LyIFN-a-2. . Figure 20 is a schematic outline of the construction of plasmid p31 containing a PHO5 termination fragment.

Figure 21 shows the nucleotide sequence of the Sau3A-Pstl PHOS ° transcription termination fragment. }

Figure 22 is a schematic outline of the construction of plasmids p3L/LFLES)), pIL/IF2(5)), p31/1F3(5,) and p31/1IF2(1'b). Co

Figure 23 is a schematic diagram showing the construction of the plasmid p3L/1F(8)). vyigure 24 schematically illustrates the construction of a correct

PHOS5-HBVs junction in plasmid pBR322,/PHO5/HBVsAl4. ’

Figure 25 shows the DNA sequence in the vicinity of the PHOS promoter and HBVs coding region fusion point in plasmid pBR322/PHO5/HBVs.

Figure 26 is a schematic. diagram showing the construction of the yeast expression plasmids pJDB207/PHO5/UBVsAl4 and pJDB207,/7H0>1UBVsAlat.,

BAD ORIGINAL P)

Figure 27 is a schematic diagram sliowing the construction of the yeast expression plasmids pJDB207/TF2(1'b)A and pJDB207/1F2(5 )A72.

Figure 28 displays the nucleotide sequences of plasmids pJDB207/1F2(5,)

A72 and pJDB207/1F2(5,) 082 around the XhoI junction between the 3' npontranslated region of LFN-5, and the PHOS transcription termination region.

Figure 29 is a schematic diagram showing the construction of plasmid

CG-pBR322HLycIFN(a-3)-252.

Figure 30 shows the structures of plasmids CG-pBR322/HLycIFN(a-2)-261 and CG-pBR322/HLycIFN(a-1)-258.

Figure 31 displays a schematic outline of the process for deleting the PHOS signal sequence in expression plasmid p31 and specifically shows the construction of plasmid p31/R.

Figure 32 schematically shows the collection of clones obtained in the process outlined in fig. 31. .

Figures 33 and 34 display the nucleotide sequences of the BamHI~-EcoRI1 restriction fragments containing the. PHO5/R and PHOS/Y promoter regions. :

Figures 35 to 37 schematically display the process of inserting 1FN-a-3, -a-2 and -a-1 DNA into plasmid p31/R.

Figure 38 is a schematic diagram showing the construction of plasmid pJDB207R/IF (a-3)

The following Examples serve to illustrate the present invention but should not be construed as a limitation thereof.

BAD ORIGINAL 9

I cn =

~ 29 -

Experimental part 2 6 26 6

The following abbreviations are uced in the Examples:

EtBr: ethidium bromide

BSA: bovine serum albumin :

DIT: 1,4-dithiothreitol (1,4-dimercapto-2,3-butanediol)

EDTA: ethylenediaminetetraacetic acid

SDS: sodium dodecyl sulphate

TNE: solution containing 100 mM NaCl, 10 mM Tris+HCl (pH 7.5), and 1 mM EDTA.

TriselCl:: tris- (hydroxymethyl) -aminomethane, pil adjusted with liC1

PMSF: : phenylmethanesulphonylfluoride

TE: solution containing 10 mM TriseHC1 (pl 7.5) and 1 mM EDTA

I'’xample 1: Construction of a yeast gene library : _— rn SC Jee Gene library

Thirty pg of total high molecular weight yeast DNA (23) from wild type Saccharomyces cerevisiae strain $S288C is incubated for 30 min at 37°C with 2 units of EcoRI methylase (New England Biolabs) in 250 pl of EcoRI methylation buffer as recommended by the supplier.

DNA is precipitated by ethanol, resuspended in 500 pl of 25 mM Tris<HCl pH 8.5, 2 mM MgCl, (EcoRI* huffer) ( 24) and digested-with

LcoRI (Boehringer) until the size distribution of the DNA frag= oo ments has a maximum in the 30-50 kb range (a Xhol digest of ,\ DNA provides appropriate 33 kb and 17 kb markers). The yeast DNA di- gested under FcoRI* conditions is size-fractionated on a sucrose gradient (5-20% sucrose in 10 uM TriseHCl pH 7.5, 1 mM EDTA) for 6 hrs at 38'000 rpm in a SW 40 rotor. Thirty fractions of 0.4 ml cach are collected from the top of the gradient. Fraction 16 con- tains DNA fragments of 30-40 kb in size. The DNA of this fraction (3 pg) is precipitated with ethanol and ligated for 16 hours at . [BAD ORIGINAL Yi

, 26266 15°C in a total volume of 15 pl to 1 pg of cosmid vector pYel (25), linearized by EcoRI. Ligation is carried out with 300 U T4

DNA ligase (New England Biolabs) using the buffer system described by the supplier. The DNA is packaged in vitro into bacteriophage A (26) and the assembled phages are used to transduce E. coli strain

B101 (x, ms leu , pro, recA). The efficiency of transduction is about 5000 ampicillin-resistant colonies per ug of pYcl vector, : 3000 amp" colonies are picked and grown individually in the wells of microtiter dishes in LB medium [10 g Bacto-Tryptone (Difco), 5 g Bacto Yeast Extract (Difco), 10 g NaCl] containing 100 4g /ml ampicillin. :

Example 2: Isolation of the regulated acid phosphatase gene PHOS

Replicas of the gene library are grown on LB agar plates (LB medium plus 15 g/1 agar) containing 100 pg/ml ampicillin. The cell material from 500 colonies is washed off the plates and pooled. DNA is isol- ated from individual pools using the following protocol: .

The cells are harvested by centrifugation (Sorvall, GSA rotor, 10 min at 6000 rpm, 4°C), resuspended in 100 ml TE (10 mM TriseHCIL, 1 mM EDTA, pH 8.0) and centrifuged again under the above conditions.

The cell pellet is resuspended in 3 ml Tsuc [50 mM TriseHCI, pH 7.5, 25% (wv) sucrose] and transferred to $S-34 polypropylene Sorvall tubes. All subsequent steps are carried out on ice: 0.3 ml of lyso- zyme solution (10 mg/ml, purchased from Worthington, 11'000 U/mg) is added, after 5 min 1.2 wml EDTA (500 mM, pH 8.0), and after another 5 min 4.8 ml detergent [0.17 Triton X-100 (Merck), 50 mM EDTA, 50 mM Tris+HCl, pH 8.0] are added. After 5 min the lysate is centri- fuged in a precooled SS-34 rotor for 40 min at 4°C. The supernatant is carefully removed and solid CsCl is added (8.3 g CsCl to 8.7 ml of supernatant). After the addition of ethidium bromide (Sigma) (BAD ORIGINAL p)

~- 31 - 26266 (final concentration 1 mg/ml supernatant) the solution is trans- ferred to 13.5 ml Quick Seal polyallomer tubes (Beckman) and centrifuged in a Beckman Ti50 rotor for 40 hrs at 40'000 rpm. Two fluorescent bands can be visualized with long wave UV (366 nm). The lower band contains supercoiled plasmid DNA which is collected by puncturing the tube from the side with a 2 ml syringe (18G ncedle).

The ethidium bromide is removed by extracting 5 times with equal volumes of isopropanol (saturated with CsCl) and the product is transferred to 30 ml Corex tubes. 2.5 volumes of TE is added and the DNA is precipitated with ethanol. The solution is then kept for 12-15 hrs at -20°C. The precipitated DNA is collected by centrif- ugation in a Sorvall liB-4 rotor for 30 min at 12'000 rpm at 0°C and redissolved in 200 pl of TE. 50-100 pug of hybrid plasmid DNA are recovered from a 100 ml culture. .

Plasmid DNA from these pools is used to transform S. cerevisiae strain

AH216 (a, his3, leud, pho3, pho5) according to the procedure described by Hinnen et al. ( 1 ). Yeast transformants are replica- plated on low P.-minimal medium (as "Difco yeast minimal medium with- out amino acids" supplemented with 20 g-1 glucose, but prepared from the components according to the recipe of Difco (Difco Manual,

Difco Laboratories, Detroit, USA) except that 0.03 gl KH, PO, plus 1 g/1 KCl is used instead of 1 g-1 KH, PO, ] and stained for acid phos- phatase activity by overlayering them with staining agar [1% Difco agar in 100 mM acetate buffer pH 4.0, 2 mg/ml Fast Blue B Salt (Scrva) and 0.2 mg/ml a-naphthyl phosphate (Serva)}. Colonies with a functional PHOS gene stain red upon derepression of the gene on low P,-medium. By repeated subpooling (17) of the gene library 3 independent clones exhibiting repressible acid phosphatase activity arc obtained.

One of these clones (pG7) is further analysed. The hybrid plasmid has a size of 42 kb. EcoRI and BamHI fragments of pG7 are subcloned 3 oo BAD ORIGINAL jf be dd in pBR322/11S3 (16) and pJDB207 (28 ) rcspectively. Restriction digests are as recommended by the supplier (New England Biolabs) and ligations are performed in 20 ul with 150 U T4 DNA ligase (New

England Biolabs) and 20 pg/ml of the individual digested plasmids (conditions as suggested by New England Biolabs). A 5.1 kb Bamll frag- ment which is part of a 8 kb EcoRI fragment is subcloned in yeast vector pJDB207 and, upon transformation of yeast strain Ali216, this hybrid plasmid (pJDB207/PHOS,PHO3, see fig. 1) elicites high phosphatase activity under derepressed (low P.-) conditions (PHOS gene) and low levels of activity in normal yeast minimal medium (expression of the PHO3 gene).

Example 3: Localisation of the PHO5 and PHOJ genes and DNA sequence analysis a, The PHOS gene

For the localisation of PIO3 and PHOS within the BamHI fragment advan- tage is taken of the pattern of Sau3A restriction sites and a unique Pstl site. Digestion of the BamHI fragment with restriction endonuclease Sau3A (New England Biolabs) generates 6 fragments (A-F, fig. 2). Subcloning of a partial Sau3A digest into the BamHI site of self-replicating yeast vector pJDB207 leads to plasmids with different combinations of Sau3A fragments. These plasmids are then used to transform the pho3, phoS mutantiyeast S. cerevisiae AH216. Trans- formants are checked for acid phosphatase activity after growth on either : lov P, ~ or normal minimal medium plates. Clones containing at least Sau3A fragments A and B (fig. 2, No. 1-4) express acid phosphatase at the same level (qualitative estimates after overlayering with acid phos- phatase staining agar, as described in Example 2) as the entire5.1 kb

BamHI fragment. Expression is regulated normally by the concentra- tion of inorganic phosphate in the medium. Clones with Sau3A-fragment

A only (fig. 2, No. 5, 6) express low levels of acid phosphatase, which is not influenced by the inorganic phosphate concentration in the medium, This indicates that information carried by the Sau3A fragment A is sufficient for constitutive acid phosphatase (PHO3) [BAD ORIGINAL 9 expression. Sau3A fragment B (14g. 2, No. 7) alone does not lead to any expression of acid phosphatase under either repressed or de- repressed conditions. However, a subclone with the complete se- quence between the BamMI and Pscl sites (fig. 2, No. 10) shows regu- lated, but not constitutive synthesis of acid phosphatase. This subclone must therefore contain the yeast PUOS5 gene (16).

The cxact localisation of the PHOS gene is determined by DNA sequenc- ing using the method of Maxam and Gilbert (15). A 623bp BamHI-Sall restriction fragment is cloned into plasmid pBR322 (sce fig. 1), replacing the BamlHI-Sall fragment which extends from position 375 to 650 (pBR322 nomenclature), using digestion and ligation conditions as described above (all enzymes are from New England Biolabs). DNA fragments of the BamllI-Sall DNA insert are asymmetrically labelled at their 5' ends at the following sites: BamHI (- 541, Sau3A (-200) and Sall (+82), (for numbering see fig. 3a). The nucleotide sequence of the 623bp BamHI-Sall DNA insert is depicted in fig. 3a. It reveals that the insert contains the PHO5 promoter region and part of the PHOS phosphatase protein coding region. : b. The PHO3 gene

The exact localisation of the PHO3 gene is determined by DNA sequence analysis according to the manual "M13 cloning and DNA sequencing. system" published by New England Biolabs. A 416 bp (5')PstI-Rsal(3') : fragment is subcloned in vectors M13mp8 and M13mp9 (49), using unique

PstI and Smal restriction sites. The nucleotide sequence of the 416 bp Psti-Rsal DNA insert is shown in fig. 3b. It reveals that the insert contains the PHO3 promoter region and part of the PHO3 acid phosphatase protein coding sequence. pro cnciia. J)

Vo oC

Example 4: Construction of plasmid p30 (see fig. 4) a) Flimination of the Ball restriction site in plasmid pBR322

The scheme outlined in fig. 4 requires elimination of the unique Ball restriction site in plasmid pBR322. 3 pg of pBR322 are digested to completion with restriction endonucleases Ball (BRL) and Pvull (Bio- labs) according to the rccommendations of the suppliers. The Ball/

Pvull double digest of pBR322 results in two restriction fragments of 3738 bp and 622 bp in size. The two fragments are separated on a 17 low melting agarose gel (Sigma) in TBE (90 mM Tris<HC1l pH 8.3, 2.5 mM EDTA, 90 mM boric acid) buffer. The DNA bands are stained with ethidiumbromide and visualized under long wave UV light at 366 nm. The piece of agarose containing the 3738 bp fragment is cut out from the gel, liquified at 65°C, adjusted to 500 mM NaCl and incubated at 65°C for 20 min. One volume of phenol (equilibrated with 10 mM Tris+HC1l pl 7.5, 1 mM EDTA, 500 mM NaCl) is added. The aqueous phase is rcextracted twice with phenol and once with chloroform. The DNA is precipitated with 2.5 volumes of cold absolute ethanol and collected by centrifugation. The DNA pellet is washed with cold 80Z ethanol and then dried in vacuum. The DNA is resuspend- ed in TE at a concentration of 0.15 mg/ml.

The isolated 3738 bp DNA fragment has. two blunt ends resulting from i the Ball and Pvull double digests. The DNA is circularized by blunt end ligation. 0.6 ug of DNA are incubated over night at room tem— perature in 30 ul of 60 mM Trise<HCl pH 7.5, 10 mM MgCl,, 10 mM DTT, 4 mM ATP, and 900 U of T4 DNA ligase (Biolabs). 5 pl aliquots of the ligation mixture are added to 50 pl of calcium treated, transforma- tion competent E. coli HB1Ol cells, prepared by the method of Mandel et al. (29). The mixture is kept on ice for 5 min, then incubated for 2 min at 37°C and left 10 min at room temperature before plating on LB agar plates containing 100 ug/ml of ampicillin. Six amp® colonies are picked and grown individually in 100 ml of LB (as above but with- out agar) medium containing 100 pg/ml ampicillin, Plasmid DNA is

Fr

BAD ORIGINAL J prepared from the cells using the procedure described in Example 2.

Restriction digests with ilaelll (purchased from Biolabs, digestion conditions as suggested by supplier), Pvull and Ball of the plasmids are analyzed on a 1.57 agarose gel in TBE buffcr. The restriction pattern and the predicted size of the newly formed junction frag- ment indicates that the plasmids are identical and contain all of the pBR322 sequences except for the Ball - Pvull fragment. These plasmids lack the Ball restriction site and are referred to as pBR322ABall. job) Cloning of a yeast 5.1 kb BamHI restriction fragment containing

PHO5 and PHO3 into pBR322aBall pJDB207,/PHO5,PHO3 (see fig. 1) contains a yeast 5.1 BamHI insert with the genes for regulated and constitutive yeast acid phosphatase (PHOS and PHO3). pJDB207/PHO5,PHO3 as well as plasmid pBR322ABall are di- gested with restriction endonuclease BamHI. After complete digestion the enzyme is inactivated for 2 min at 65°C. Both DNAs are precipi- tated by ethanol and resuspended in 10 mM Tris<HCl pH 8.0 at a con- centration of 0.2 mg/ml each. 0.5 pg of each of the two BamlHI-di- gested DNAs are combined and ligated in 20 ul of ligation buffer (as suggested by New England Biolabs), containing 300 U of T4 DNA - ligase, for 20 hrs at 15°C. 5 pl aliquots of the ligation mixture are added to 50 pl of calcium-treated E. coli HB1Ol cells and transformation is carried out as described in Example 4a. The trans- formed E. coli cells are tested for their resistance towards ampi-. cillin and tetracyclin. Eight amp, cet’ colonies are isolated and grown in 100 ml of LB medium containing 100 pg/ml of ampicillin,

Plasmid DNA is isolated from the cells (see Example 2). Restric- tiondigests with BamHI show that 4 plasmids contain a 5.1 kb insert besides the 3.7 kb vector fragment (pBR322ABall). Restriction di- gests with Sall (New England Biolabs) determine the orientation of the inserted 5.1 kb fragment: two plasmids have the insert orient- ed as shown in figure 4. One of them is referred to as p30. The direction of transcription of the PHO5, PHO3 genes in the 5.1 kb insert is anticlockwise as indicated in figure 4. ‘. Cr BAD Orutiinvre PZ!

Twine oe apts mcsmatannn

Example 5: Insertion of foreign DNA into p30 (see fig. 4) a) Isolation of a 3.9 kb EcoRI-Ball fragment of p30 (fragment A) 10 pg of p30 DNA are digested with restriction endonuclease Ball.

After extraction with phenol/chloroform, the DNA is precipitated with ethanol. The DNA is resuspended in 100 ul TE buffer. The restriction fragments are separated on a preparative 0.87 low melting agarose gel (Sigma). A 5.1 kb fragment, containing the vector part of p30 is cluted from the gel as described in Example 4a. The DNA is purified ’ by adsorbing the DNA on a DE52 (Whatman) ion exchange column in a low salt buffer (150 mM NaCl, 10 mM Tris HCl pH 8.0, 1 mM EDTA) and then eluting it with a high salt buffer solution (1.5 M NaCl, 10 nM

Tris*HCL pH 8.0 and 1 mM EDTA). The DNA is precipitated with ethanol and then further digested with EcoRI (Boehringer). The 3.9 kb EcoRI-

Ball restriction fragment is again separated on a preparative 0.8% low melting agarose gel, recovered as described in Example 4a and ethanol precipitated. This DNA fragment is called fragment A. b) Isolation of a 602 bp HaeIIl-EcoRI fragment of CG-pBR322/HLyc IFN-81 (fragment B)

E.coli strain HB-101 CC-pBR322/NLycIFN-8) (see Example 10E) is grown in 100 ml LB medium supplemented with 10 pg/ml tetracyclin and plasmid DNA is isolated as described in Example 2, Nine pg of HLycIFN-8/

DNA are completely digested with restriction endonuclease HaeIlIl. BE :

The restriction fragments are separated on a preparative 0.87 low : melting agarose gel. A 940 bp lHaelll fragment is cut out and eluted from the agarose gel as described in Example 4a. The DNA is purified on DE52 as described in Example 5a and then further digested with

EcoRI. The 602 bp EcoRI-HaeIlIl fragment is again separated on a preparative 0.8% low melting agarose gel, recovered as described in

Example 4a and ethanol precipitated. This DNA fragment is called fragment B.

J ORIGINAL J

(—

¢) Ligation of fragments A and B (gee fig. 5) 9, 6 26 6 j

The two restriction fragments can be ligated enzymatically via the

EcoRI sticky ends and the blunt ends of Ball and HaelIl respectively, thus creating a circular molecule with a unique EcoRI site and a

Ball-HaelIl junction which is cleavable with HaeIIl (but not with

Ball).

The ligation is carried out in a buffer system containing 60 mM

Tris+CL pH 7.5, 10 mM MgCl,, 10 mi DIT, 4 wh ATP, 300 units of

T4 DNA ligase for 16 hrs at 23°C at a DNA concentration of 20 gg/ml . of fragment A and 3 pg/ml of fragment B in a total volume of 10 gd. d) Transformation of E. coli IIB101 with the ligated fragments 2 pl aliquots of the ligation mixture (see Example 5c) are added to 50 pl of calcium-treated E. coli HBLOl cells (see Example 4a).

The mixtures are then plated on LB agar plates supplemented with 100 pg/ml ampicillin. The plates are incubated at 37°C for 16 hrs.

About 300 ampicillin resistant colonies of E. coli HUB101l are pre- pared. Plasmid DNA from eight ampicillin resistant colonies is . isolated, analysed and their structure is determined by comparing the mobility of the restriction {ragments obtained after cleavage with EcoRI and HaeIII with standard DNA [bacteriophage A DNA di- gested with HindIII (New England Biolabs), p30 plasmid DNA digested with HaelIl and EcoRI]. After verification of the structure of the junctions, 5 plasmids are obtained which have the correct structure.’

One of these plasmids containing the PHO5 promoter linked to the 8 -interferon polypeptide coding region (see fig. 5) is called p30LFN1(8,).

Example 6: Addition of replication origin and selective marker for yeast (see fig. 4) ’ a) lsolation of a 1.5 kb EcoRI fragment from plasmid Yrp7 and its ligation into plasmid p3OIFN1(8)) in order to facilitate the ligation reaction the 1.5 kb EcoRI restric- tion fragment is purified. Plasmid Yrp? (4 ) is cut with LcoRI,

BAD ORIGINAL 9 —

the two fragments obtained are separated on a 0.87 agarose gel and the 1.5 kb fragment containing a yeast autonomously replicating segment and the ycast TRPL gene is purified and isolated as describ- ed in Example 4a. Ligation is carried out (as suggested by New England

Biolabs) with 20 pg/ml of EcoRI cut p30LFNL(8)) and 10 pg/ml of the 1.5 kb

EcoRI restriction fragment from Yrp7; 100 units of T4 ligase are : used. b) Transformation of E. coli JA 194 with the ligated fragments

Plasmids containing the TRP1 yeast gene are directly selectable by transformation of the E. coli trpC mutant strain JA 194 (trpC, leuB,

BL). The E. coli trpC gene codas for the E. coli N-(5'-phosphoribosyl)- anthranilate isomerase. E. coli trpC mutants can be complemented by the yeast TRP1 gene (4). Transformation of E. coli strain JA 194 is carried out as described for E. coli 1IB101 (see Example 4a) except for the following modification: before plating the mixtures onto agar plates the cells are allowed to recover in 1 ml of LB medium at 37°C for 60 min; the cells are washed once with E. coli M9 minimal medium (30) and plated onto M9 minimal medium plates supplemented with vitamine B1 (1 je/ml) and L-lcucine (20 mg/ml). The plates are in-: cubated for 2 days at 37°C. Approximately 1000 tryptophan prototrophic

E. coli colonies are recovered. c) Isolation and characterization of hybrid plasmids IE .

Trp’ colonies are purified on LB plates supplemented with 100 pg/ml - ampicillin. Individual colonies are picked and plasmids are isolated as described in Example 2. Purified plasmids are analyzed by measur- ing the size of the restriction fragments generated after cleavage with EcoRI, HindIII, PstI and BglII (Biolabs). Two different types of plasmids are obtained which contain the 1.5 kb EcoRI restriction ‘ fragment in the two possible orientations (see fig. 4). They are : named p30LFN2(8)) and p30IFN2'(8)) as indicated in figure 4,

AD ORIGINAL J

—

Example 7: Transformation of Saccharomyces cerevisiae RH971 and induction of interfr;on production

Plasmids p3OTFN2(8,) and p30LFN2' (8) are each introduced into Saccharo- myces cerevisiae strain RH971 (a, trpl, leu, his4) in analogy as described by Hinnen et al.(1).0ne pg of plasmid DNA is added to 100 pl of a spheroplast suspension and the mixture is treated with polyethylene glycole as described (1). The spheroplasts are mixed with 10 ml regen- eration agar and plated onto yeast minimal medium plates without leucine. After incubation for 3 days at 30°C, about 1000 transformed cells are obtained.

One single yeast colony from the ycast transformation plates {named

Saccharomyces cerevisiae Ri971/p30IFN2(8,) and /p30LFN2'(8)) respectively] is picked into 10 ml of yeast minimal medium in a 100 ml

Erlenmeyer flask, and grown at 30°C at 200 rpm for 24 hrs to a density of about 2-3x10’ cells/ml. The cells are washed once with 20 ml of low-P, minimal medium. Three ml of the resuspended cells are used to inoculate 300 ml low-P, minimal medium and 300 ml normal minimal medium, respectively, in 1000 ml Erlenmeyer flasks. Incubation is at 30°C at 160 rpm. Induction of the PHOS5 promoter is followed by measuring the appearance of acid phosphatase activity in whole cells as described by Toh-e et al. (31). The cells are grown to about : 1-2x10’ cells/ml (26-30 hrs of incubation).

Example 8: Preparation of yeast cell extracts and determination of the interferon titer

Cells from the 300 ml culture medium (see Example 7) at a density of 1-2x10"/ml are collected by centrifugation in a Sorvall GSA rotor for 5 min at 8000 rpm at 4°C. The cells are washed once with 100 ml 1,0, resuspended in 6 ml ice cold lysis mix [0.1 M potassium phos- phate buffer pH 7.4, 12 (v/v) Triton X-100,0.0001M PMSF (Merck)] and transferred to a 30 ml corex tube. The suspension is centrifuged again for 5 min in a Sorvall SS-34 rotor at 8000 rpm at 4°C and re- suspended in 3 ml lysis mix at 0°C. Four g of glass beads (0.4 mm in diameter) are added to the cells and the suspension is shaken on a _ 'BAD OHlGI

_ _ — E— - 40 ~ 26266 !

Vortex Hier (Scientific lnstruments Inc., USA) at full speed for 10 sec and then cooled for 1 min in an ice bath. This shaking procedure is repeated 5 to 10 times until more than 907% of the cells are broken (check under light microscope). Cell debris and glass beads are removed from the solution by centrifugation for min at 8000 rpm at 4°C in a Sorvall HB-4 rotor. The supernatant is transferred to Eppendorf tubes, frozen in liquid nitrogen and stored at -60°C. Interferon activity is determined according to the procedure of Armstrong (32) using human cCcL-23 cells and vesicular 10 stomatitis virus (vsV) as the challenge virus. The results are summarized in Table 1.

Table 1

Interferon activity in Saccharomyces cerevisiae strain RHO71 after transformation with the recombinant plasmids p30IFN2(8)) and p3OIFN2' (81), respectively, and also with plasmid pJDB207/1FN2' (8) (see Example 9) a

Interferon activity expressed in units/ml yeast cell extract - plasmid p30IFN2(81) p30IFN2' (8]) | pipB207/1FN2" (8) — ee repressed conditions 0 I 100 |(normal phosphate content) dercpressed conditions 700 . 7000 , 50000 . -- (1ow phosphate content)

Example 9: Insertion of the interferon gene into the high copy number yeast 2p vector pJuB 207 (see Fig. 6) yeast Bp vector bo ee

Plasmid p30IFN2' (81) is digested with restriction endonucleases HindIII and BamHI according to the specifications of the supplier (Biolabs).

Two fragments are generated of the size of 4.0 kb and 2.0 kb. The 2.0 kb restriction fragment is separated and purified by low melting agarose gel electrophoresis as described under step 4a. a BAD ORGIES i

Plasmid pJDB207 (28) is digested with restriction endonucleases Hind

IIT and Banlll. Three fragments are generated. The 6.5 kb restriction fragment is separated as above. 0.3 ug of the 2.0 kb fragment (containing the PHOS promoter linked to the interferon protein coding region) is ligated for 15 hrs to the 6.5 kb vector fragment in a total volume of 20 ul using 300 U T4 DNA ligase under conditions described by the supplier (Biolabs). E. coli

IB101 cells are transformed and ampicillin resistant colonies gre selected. The plasmid DNA is isolated and the correct structure of the isolated plasmid DNA ig verified by restriction digestions using

HindIII and BamHI, with pIOIFN2'(8)) and pJDB207 digested with the same enzymes as molecular weight standards. The new plasmid obtained is called PIDB207/1FN2" (8!) plasmid pJDB207/IFN2' (81) is transformed into S.cerevisiae strainRH97] "in analogy as described (1) selecting for leucine prototrophic colonies, . | One single leucine prototrophic yeast colony [named Saccharomyces cerevisiae RUI7L/pJDB207/1FN2' (8))] is picked and grown as described in Example 7. The interferon titer is determined as described in

Example 8. The results are depicted in Table 1.

Ap ORIGINAL 9 ee

Example 10: Production of E.coli strains transformed with recombinant plasmids containing the coding regions for human lymphoblastoid interferons ‘ ee i

A. Isolation of poly(A) RNA enriched for HulkFN mRNA (figure 7) a) Induction of the Namalwa cells

Namalwa cells are grown in culture medium RPMI 1640 containing 10% fetal calf scrum at 37°C. When a cell density of 3.100 cells/ml is reached, the suspension is centrifuged at 800 x g for 10 minutes at : room temperature. The collected cells are resuspended in 200 ml of culture medium containing glutamine (0.027% by volume), penicillin (200 units/ml) and streptomycin (50 pg/ml). The cells are incubated for 90 minutes at 37°C with Newcastle discase virus (NDV 110) at a ratio of 190 wav 10° cells (HAU: haemagglutination units). By adding fresh culture medium the cell density is adjusted to 1.3-10° cells/ml . and the cell suspension is shaken at 34°C at 100 rpm. After 12 h, . 6+10° cells are harvested and resuspended in 50 ml phosphate-buffered saline ("PBS"; 1 1 PBS contains 80 g NaCl, 2 g KCl, 14.4 g Na, llEo, and 2 g KIL, PO,) Before harvesting the cells,a sample is removed and the interferon activity is determined according to the procedure of ’

Armstrong (32) using human CCL-23 cells and vesicular stomatitis virus (VsV) as the challenge virus. 4300 IFN units/ml are found. b) Disruption of the cells and deproteinization

The cell suspension (610° cells in 50 wml PBS) is added at room : temperature to 800 ml lysis buffer consisting of 0.05 M Tris-<liCl (pi 7.5), 0.1 ¥ NaCl, 5 md EDTA and 2% SDS (cryst. rescarch grade,

Serva). The lysate is digested with 0.2 mg/ml of preincubated (2 h at 37°C) protease (Protease P, type VI, Sigma) at room temperature for 1 hh while stirring the solution. The solution is deproteinized by extracting 3 times with 500 ml phenol satured with TNE and 5 times with 500 ml chloroform. 500 mg of nucleic acids are obtained as measured by absorbance at 260 nm. a0 ORIGINAL 9 . ln —

c) Removal of contaminating DNA and RNA 2 6 26 6

The slightly viscous dqueous solution obtained ag described above (step AD) is adjusted to 0.3 M NaCl and 1 g of oligo(dT) cellulose (type 7, p-L Biochemicals) is added. Alter stirring for 30 min at room temperature the suspension is centrifuged in 1 1 Sorvall bottles in a Sorvall RC-3 centrifuge at 4000 rpm for 10 min at room temperature and the oligo(dT) cellulose slurry is washed twice with 40 ml 2 x TNE containing 0.5% SDS. The bound poly(A) RNA is then eluted by five successive washes with 2.5 ml 1,0. The yield is 720 ug poly(A) RNA as determined by measuring the optical density. ‘the Supernatant RNA solution from the first adsorption is adsorbed a second time to 1 g i of oligo(dT) cellulose aud eluted as described above, yielding 320 pg

Poly(A) RNA. The eluates are pooled, adjusted to TNE and the poly(A)

RNA is precipitated with 67% ethanol at -20°C for 10 hours. The RNA is collected by centrifugation at 10 000 rpm in a Sorwall RC-5B centrifuge for 10 min at 0°C. The precipitate (1 mg) is redissolved in 1 ml of 1 mM EDTA.

The RNA is assayed for NulPN WRNA activity by injection into oocytes of Nenopus laevis as follows: : 50 nl of the RNA solution is injf.ted into cach of 20 oocytes. The oocytes are incubated in Barth medium ( 2 mM Tris, 88 mM NaCl, 1. mM oo

KCl, 0.33 mM Ca(N03) 2°H50, 0.41 mM cacL, 2110, 0.82 nf MgSO, <7H,0, : 2.4 mM NalCo,, 0.01 mg ml penicillin, 0.01 mg/ml streptomycin; the solution is adjusted to pH 7.6 with HC1) according to Gurdon (33),

Barth (34) and Colman et al. (35). The injected oocytes are incubated

Lor 42-48 hours and the incubation medium ig vtemovaed, centrifuged for 5 min in an Eppendorf centrifuge, and the supernatant is stored at —20°C or -80°C until it is used for assay. The IFN activity is assayed essentially according to Armstrong (32) except that vsy is used as

JO the challenge virus on Hep=-2-cells (Flow Laboratories), The oocyte extract has a specific activity of 600 1U interferon per yg RNA in- jected.

BAD ORIGINAL dP mamee r tomn

4) Lariching the poly(A) RNA for HulkN_mRNA 2 6 26 6

The poly(A) RNA is passed through a Chelex-100 column (200-400 mesh,

Bio-Rad) of 0.5 ml bed volume. The column is rinsed with L ml of 1 mM

LDTA. : .

The cluate (1 mg poly(A) RNA in 2 wl EDTA) is heated for 2 min at 100°C and subjected to centrifugation through a sucrose density gra- diene ( 6 14 ml sucrose solutions increasing in sucrose concen- tration from 5% to 23% (wv) and containing 50 mM Tris<lICl [pH 7.5], 0.2 M NaCl and 1 mM EDTA). The centrifugation is carried out in a .

TST 41 rotor (Kontron AG) at 35 000 rpm for 16 h at 5°C. 0.3 ml fractions are collected with an ISCO gradient collector. 2 volumes of cthanol are added to each fraction and the solution is allowed to stand for 10 h at -20°C. The precipitated mRNA is collected by centrifugation (Sorvall, HB-4 rotor at 0°C, 10 000 rpm for 10 min).

The precipitate of each fraction is redissolved in 25 ul of 1 mM EDTA and cach fraction is assayed for human IFN mRNA activity as described above (step Ac), except that only 10 oocytes are injected per RNA sample instead of 20. The results are given in table 2. ’ | RN AIEEE : ono omni: LE)

- 45 ~ 26266

Tabte 2:

HOlEN mRNA activity from fractions of sucrose-density gradient, ’ fraction No. ff LFN activity (units/ml) 1-18 = 19 162 20 162 21 162 22 162 23 not tested 24 729 25 not tested 26 405 27 not tested : 28 486 29 not tested 30 162 31 not tested : 32 162 33 not tested 34 54 35-40 mot tested Jo Co :

The fractions 23-29 are pooled and the poly(A) RNA is purified further as follows:

The poly(A) RUA solution is adjusted to 2 x TNE in 0.5% SDS and applied on a 200 ul oligo(dT) cellulose column. The column is washed vith 2 ml of 2 x TNE in 0.52% SDS and the poly(A) RNA is eluted by 5 washes with 0.5 ml H,0. The eluate is adjusted to TNE and the solu- tion is extracted twice with an equal volume of phenol (saturated in

THE) and twice with an cqual volume of chloroform. The poly(A) RMA ao ORIGINAL P is precipitated with 2 volumes of cthanol at -20°C for 10 hours and collected by centrifupation in a UB-4 rotor as described before.

The poly(A) RNA Is dissolved in 100 pl of 0.5 mM EDTA. The yield is 40 pg as determined by measuring the optical density.

A portion of the poly(A) RNA is assayed for human IFN activity as described above by using 20 oocytes per assay. The poly(A) RNA pre- paration has a specific activity of 8100 IU interferon per pg RNA.

B. Preparation of double-stranded cDNA (figure 7)

Poly(A) RNA enriched for HulFN wRNA (sce step Ad) is used as a template to prepare double-stranded cDNA essentially as described by Efstra- tiadis et al. (36), Maniatis et al. (37) and Hoeijmakers et al. (38). a) First strand synthesis : 250 nl reaction mixture containing 40 mM Tris+lIC1 (pH 7.5), 30 mM NaCl, 5 mM MgCl,, 0.5 mM DTT (Calbiochem.), 1 mM dGTP, dCTP, dTTP (P-L Bio- chemicals) and 1 mM 32p_gatp (Amersham, specific activity 50 000 cpm snmole), 20 pg/ml oligo(dT) , |g (P-L Biochemicals), 40 pg/ml poly(A)

RNA and 100 units of avian myeloblastosis virus (AMV) reverse trans- criptase (Life Sciences, Inc., St ¢Petersburg, Florida) are incubated for 80 min at 37°C. The rcaction is toriinated by sadjusting the sdlu- tion to 10 mM EDTA and 0.1% SDS. The mixture is extracted once with ° 1 volume of phenol. The aqueous phase is reextracted with 1 volume of chloroform and applied on a 3 ml Sephadex G-50 (Pharmacia, fine) column. 0.1 ml fractions are collected. The radioactivity of cach friction is determined by weasuring the Cerenkov radiation. Radio- active fractions are pooled and the nucleic acids are precipitated with 2 volumes of cthanol at -20°C for 10 h. The sample is centri- fuged in a HB-4 rotor for 20 win at 10 000 rpm at 0°C. The precipitate is dissolved in 95 ul of H,0. 5 pl of 10N NaOH is added and the mixture is incubated at 25°C for 40 min. After neutralization with SM \ no ano ORIGINAL 9

26266 0 acetic acid, 50 pl I.,0 and 2 volumes of cthanol are added and the sample is stored at -20°C for 10 hrs. The precipitate is collected by centrifugation as described before and redissolved in 200 nl of 0.1 mid EDTA. The yield of single-stranded ¢DNA is 3.7 ng. The size of . the ¢bNA is 700-1500 nucleotides in length, as determined from its electrophoretic mobility in a 67% polyacrylamide gel in Tris-borate-

EDTA (108 g of Tris, 9.3 g of disodium EDTA, and 55 g of boric acid per one 1 solution of pil 8.3) containing 7 M urea relative to mavker DNAs of known length (39). . b) Sccond strand synthesis and 5, endonuclease digestion

The obtained ¢DNA solution is heated at 100°C for 90 scc, chilled and incubated in a 400 ul reaction mixture comprising 0.1 M potassium phosphate buffer (pH 6.9), 10 mM MgCl, 10 mM DTT (Calbiochem),

L mM dATP, 1 mM dCTP, 1 mM dTTP (P-L, Biochemicals), 1 mM 3i-acTP (Amersham, specific activity 94 000 cpm/nmole) and 165 units/ml of

E.coli DNA polymerase I (Biolabs, New England) for 8 h at 15°C. The reaction is terminated by adding EDTA and SDS to final concentrations of 10 mM and 0.1%, respectively. The mixture is extracted with phenol and chloroform, chromatographed over Sephadex G-50 (Pharmacia, fine, 2 ml bed volume) and ethanol precipitated as described above (step Ba).

The resulting DNA is treated in a 50 wl incubation mixture containing 0.25 M NaCl, 50 mM sodium acctate (pH 4.5) and 1 mM *ZnS0, with 6 Co . units of 5) endonuclease (P-L Biochemicals) at 37°C for 30 min. The = reaction is stopped with 0.1 7% SDS and 10 mM EDTA. The reaction mixture is deproteinized with 1 volume of phenol (saturated in 50 mM sodium acetate, pH 4.5) and chloroform. The aqueous phase is chromatographed on a 2 ml Sephadex G-50 (Pharmacia, fine) column in TNE. 100 ul fractions are collected and the Cerenkov radiation of each fraction is determined. The excluded fractions are pooled and the DNA is preci- pitated with 2 volumes of ethanol at -20°C for 10 h as described above.

The precipitate is centrifuged in a HB-4 rotor (see above) and the collected precipitate is dissolved in a 100 pl solution containing 10 wd Tris<BCl (pH 7.5) and 0.5 wm FDTA. 4 ng of DNA are obtained.

BAD ORIGINAL 9

. ~- 48 - 26266

The DNA is fractionated through a sucrose density gradient (5-23 %) in 50 mM Tris-HCl (pl 7.5) and 1 mM EDTA in a TST-60 rotor (Kontron AG). ]

Centrifugation is carried out at 55 000 rpm for 5 h at 15°C. The DNA, vhich sediments faster than a 800 [ase pair marker DNA, run in a parallel gradient, is pooled, adjusted to TNE and precipitated with 67 7 cthanol at -20°C for 10 hrs. 0.4 ng double-stranded cDNA are ob- tained.

C. Preparation of pBR 322 ~ linked ¢DNA (figure 7) a) Lrepacacion of dCHP-clonpated cDNA

The 3'-termini of 0.1 Ng of the obtained ds cDNA are provided with poly (dC) tails in a 10 Hl reaction volume containing 100 mM sodium cacodylate (pH 7.2), 2.5 mM CoCl,, 50 pg BSA (Calbiochem.) per ml,

L mM dCTP and 10 units of terminal deoxynucleotidyl transferase (P-L. Biochemicals) per M8 of ds cDNA. After incubation (20 min at 27°C), EDTA is added to 10 mM and the sample is stored at -20°C until use, . b) Preparation of Pst I cleaved, dCMP elongated PBR 322 10 pg of pBR 322 plasmid DNA is digested with 10 units of Pst I endo- nuclease (Biolabs) in a 100 M1 solution containing 50 mi NaCl, 6 mM

Tris-HCl (pH 7.5), 6 mM MaCl,, 6 mM 2-meticaptoethanel and 100 pgsml gelatine for 1 hat 37°C. The solution is extracted with 1 volume of = phenol and chloroform. The solution is adjusted to TNE and the

Lincarized DNA is precipitated with -2 volumes of ethanol at -20°C for 5 h. . 25 The lincarized plasmid DNA is elongated with dGMP in a 200 Ml reaction volume containing 100 mM sodium cacodylate (pH 7.2), 5 mM MgCl,, 20 mM Nall, ro, , 50 pig BSA per ml, 1 mM dGTP and 100 units of terminal deoxynucleotidyl transferase (P-L Biochemicals). After incubation for 20 min at 37°C, EDTA is added to 10 mM and the reaction mixture is frozen at -20°C until use. 1 , [BAD ORIGINAL 9

' 26266 c) Annealing of dGMP-clongated pBR 322 to dCMP-clonsated ds cDNA

A mixture of dCMP-elongated double-stranded cDNA (0.1 ng) and dGMP- tailed lincarized pBR 322 (0.5 ng) in 500 nl TNE buffer is incubated at 65°C for one hour, at 46°C for one hour, at 37°C for one hour and at 20°C for one hour. The solution containing the pBR 322-linked cDNA is put on ice and used immediately for transformation.

D. Transformation of E. coli UB 101 with the annealed hybrid plasmid

Calcium treated E. coli HB 101 is prepared for transformation by the method of Mandel et al. (29). 10 pl of the reaction mixture containing the annealed pBR 322 hybrid plasmid DNAs prepared as described above (step Cc) are added to a mixture containing 150 ul calcium-treated E. coli HB 101 in 10 mM

MgCl, 10 mM CaCl, and 10 mM Tris<HCl (pH 7.5) in a total volume of 200 pl.

The mixture is cooled in ice for 20 min, heated to 42°C for 1 min and incubated at 20°C for 10 min. 1 ml of tryptone medium (tryptone medium contains 10 g Bacto-Trypton (Difco); 1 g yeast extract (Difco); 1 g glucose; 8 g NaCl and 294 mg CaCl, 2 1,0 in 1 1 of distilled water) is added and the mixture id incubated for 30 min at 37°C by shaking at 300 rpm. The mixture is plated onto 2 agar plates .- CC (tfc Conkey agar, Difco; 0,6 ml/plate) supplemented with 10 pg/ml of tetracycline (Sigma). The plates are: incubated at 37°C for 12-17 hrs.

About 5600 tetracycline resistant colonies of transformed E.coli HB 10L are prepared.

E. Identification of clones containing HulFN cDNA a) Synthesis of a 13-mer oligodeoxynucleotide primer (figure 8)

Anolipodecoxynucleotide complementary to a stretch of 13 nucleotides which both HulFN-a, and HuIlFN-B mRNA share in common is chemically synthesized by the phosphotriester method (cf. Itakura ect al. (40), ano ORIGINAL J oT de ooij et al (41)). The individual steps of the synthesis are out- lined in figure 8. The starting materials indicated in line 1 of figure 8 (mono- and dideoxynucleotides carrying protective groups) are known from the literature. The protective groups are split off by the methods described by Itakura et al.: the deblocking of 5'-monomethoxy- trityl (M) or dimethoxytrityl (D) substituted hydroxyl groups is per- . formed with acetic acid (80%) at room temperature, and the f-cyano-— ethyl phosphate groups are cleaved with 0.1 N sodium hydroxide in dioxane-water (4:1) at room temperature. The condensation of the building blocks is accomplished by using triisopropylbenzenesulfonyl chloride as an activating agent to afford oligodeoxynucleotides up to the fully protected 13-mer primer represented in line 7 of figure 8.

The last step (complete removal of all protective groups) is achieved in the following manner:

A solution containing 64.6 mg of the fully protected 13-mer oligodeoxy- nucleotide in 3 ml dioxane and 1 ml acetonitrile is treated with 200 mg syn-p-nitrobenzaldoxime and 124 mg nin, n3 Nn -cetramethylguanidine and allowed to stand for 27 hours. 10 ml ammonia (25%) is added and - the solution is stored for 24 hours at 50°C. After the solvent has been evaporated in vacuo, the residue is dissolved in water, adjusted to pH 4 with acetic acid’ and the solution is extracted 20: oo times with chloroform. The aqueous solution is evaporated in vacuo . ST and the residue is dissolved in 1 ml acetic acid (80%). The solution is allowed to stand for 1 hour, diluted with 6 ml water, extracted 3 times with chloroform and lyophilized. The third part of the raw product obtained is purified by chromatography on DEAE-Sephadex A 25 (column size: 10+1.3 cm) through a 200 ml 0.2-1.2 M triethylammonium bicarbonate gradient. Elution of the main fraction occurs at a gradient concentration of 0.87 M. The main fraction, which consists of the pure product as indicated by a HPLC test, is evaporated 3 times with water, filtered through 10 ml Dowex 50 W (NH, -salt) and

JAD ORIGINAL 9D lyophilized. HPLC (permaphase AAX, column size 90:0.3 cm, 60°C, 2 ml/min gradient: A = 0.005 M KIL,PO,, B = 0.5 M KIL,PO,, 0.5 ¥ KCI, ptt 4.5; 202 A —— 100% B in 30 min): tp L1.8 min, b) Preparation of a 32p-1abeled human IFN-a and IFN-B specific cDNA probe (figure 9) ’ 40 pmol of the synthetic 13-mer oligodeoxynucleotide primer (cf. step

Ea) and 40 pmol of (y-22p]-arP (5700 Ciemmol ©, Amersham) are com- bined in 100 Hl of 50 mM Tris-HCl (pH 9.5), 10 mM MgCl, and 5 mM DIT. 50 units of T, polynucleotide kinase (P-L Biochemicals) are added and after 30 min at 37°C additional 20 units of the cuzyme are added, and incubation is continued for another 15 min at 37°C. The aqueous solution containing the 32) 1abeled primer is purified by phenol extraction. Further purification is accomplished by chromatography on a 4 ml Sephadex G-50 (Pharmacia, fine) column in 1 mM TriseHCl (pil 8.0). 0.1 ml fractions are collected. The radioactivity of each fraction is determined by measuring the Cerenkov radiation. A specific activity of 4+10° Cerenkov cpm per pmole of oligodeoxynucleotide is obtained. The 3p 1abeled primer (40 pmol) is lyophilized, resuspended in 91 nl of H,0 containing 14 ng of poly(A) RNA (from induced Namalwa culls, prepared as described in step A) and heated for 60 sec at 100°C. 9 pl of 4 M KCl is added and the mixture is incubated at 25°C for 60 minutes. 450 ul reverse transcriptase mix is added such that oo the reaction volume comprises 40 mM Tris<HCl (pH 8), 4 ml MgCl, ,

I mM DIT (Calbiochem, Inc.), 74 wM KCl, 1 mM each of dATP, dGTP, } dCTP, dTIP (P-L Biochemicals) and 90 units of avian myeloblastosis virus (AV) reverse transcriptase. The incubation is continued for 1 h at 37°C. The solution is extracted with 1 volume of phenol (saturated in TNE) and the nucleic acids are precipitated with 2 volumes of ethanol at -20°C for 10 h. The precipitate is collected by centrifugation (HB=4 rotor, 20 min, 10 000 rpm, 0°C) and dis- solved in 20 ul dye mix containing 90% (v/v) formamide (ilerck, pro analysis), 1 mM EDTA, 0.057 bromo .phenol blue and 0.05% xylene cyanol [BAD ORIGINAL I

- 52 - k 26266 blue. The sample is heated at 90°C for 2 min and applied on a 57 poly- acrylamide gel in Tris-borate-EDIA (ef. Peacock et al. (39). A single band is visible on the autoradiogram which migrates between the 267 bp and 435 bp 2p labeled marker PNA fragments obtained [vom the Hae LIL digest of the plasmid pBR 322. The 3p labeled cDNA fragment is extracted from the gel and purified as described by Mueller et al. (42). 20 000 Cerenkov cpm of the 32) labeled human IFN-a and IFN-$ specific cDNA probe are obtained. : c) Screening for colonies containing HULFN cDNA (figure 9) 1650 of the transformant colonies prepared as described above (step D) are transferred to nitrocellulose filters BA 85 (Schleicher & Schuell, 8 cm diameter). The cells are lysed and their DNA is denatured and fixed to the filters in situ, according to Grunstein and Hogness (20).

The filters bearing the colonies are prehybridized in 4 x SET (a solu- tion containing 0.15 M NaCl, 30 mM Tris<lCl (pH 8.0), 1 mM EDTA) 0.1% (w/v) Ficoll 400 (Pharmacia), 0.1% (w/v) polyvinylpyrrolidone (PVP-360,Sigma), 0.1% (v/v) BSA, 0.5% SDS, 50 pg/ml denatured calf- thymus DNA (prepared as follows: S mg calf-thymus DNA (type I, Sigma) is boiled for 10 min in 0.5 M NaOll to shear the DNA, neutralized with 5 M acetic acid and precipitated with 2 volumes of ethanol at -20°C. i

The precipitate is collected by centrifygation in a liB-4 rotor for '. 10 min at 0°C and redissolved in 500 pl 0.5 mM EDTA) at 65°C for 4h Co using 20 ml mixtures per filter ai, hybridized with 10° Cerenkov cpm of- Phe 3p 1abeled probe per nitrocellulose filter in 5 x SET, 0.02% (w/v) Ficoll, 0.01% polyvinylpyrrolidone, 0.02% (v/v) BSA, 0.27 SDS and 50 pred denatured calf-thymus DNA. The hybridization is per= formed at 65°C for 36 h.

The filters are rinsed once in chloroform, twice in SET,0.5%7 SDS at room temperature and twice in SET, 0.5% SDS for 1 h at 60°C and once with 30M Trizma base at room temperature for 1 h. The filters arc dried by blotting ou 3 MM¥-paper (Whatman), and an X-ray film :

- 53 ~ 26266 (Fuji) is exposed to the [ilters using a screen (Ilford intensifying screen) at -80°C for 72 h.

Mine positive colonies are identified on the autoradiogram and are used for further investigation.

Since the primary clones of transformed cells occasionally contain more than one species of recombinant DNA molecules, the hybrid plasmid DNAs are isolated from the 9 positively hybridizing clones and used to retransform E. coli HUB 101 as described before.

The hybrid plasmid DNA is isolated as follows: 1 colony is used to inoculate 10 ml of tryptone medium, supplemented with 10 Jesml of tetracycline as above in a 25 ml Erlenmeyer flask. The culture is shaken for 15-18 hrs at 37°C at 300 rpm. The cells are harvested by centrifugation (Sorvall, US-4 rotor, 10 min at 4000 rpm, 4°C).

About 0.1 g of cells are obtained and are resuspended in 1 ml 50 mM

Tris«liCl (pH 8.0). 0.25 ml of lysozyme solution (10 mg/ml in 50 mM

Trise«HCLl (pH 8.0) ,lysozyme is purchased from Sigma) , are added and after incubation at 0°C for 10 min, 0.15 ml of 0.5 M EDTA (pH 7.5) 1s added. After another 10 min at o°c, 60 ul of 27% Triton X-100 (Merck) is added. After 30 min at 0°C, the sample is centrifuged for 30 min at 15 000 rpm and 4°C in a Sorvall SA-600 rqtor. The super—*- natant is deproteinized with 1 volume of phenol (saturated in TNE).

The phases are separated by centrifugation (Sorvall HB-4 rotor) for

LO min at 5000 rpm at 4°C. The upper phase is extracted twice with

I volume of chlovoform. Pancreatic RNAse A (Sigma; 10 ma/ml in TNE, preheated 10 min at 85°C) is added to a final concentration of 25 pg/ml and the mixture is incubated for 40 min at 37°C. The solution is then adjusted to 1 M NaCl and 10% polyethylene glycol 6000 (Fluka, autoclaved for 20 min at 120°C) and incubated at -10°C for 2 hrs. The precipitate is collected in a Sorvall HB-4 rotor (20 min at 10 000 rpm, 0°C) and redissolved in 100 nl of TNE. The DNA solution 15 extracted with 1 volume of phenol and the DNA is precipitated with 2 volumes of ethanol at -80°C for 10 min. [BAD ORIGINAL IP

, 26266

The precipitate is collected by centrifugation in an ¥Eppendorf cen- trifuge and the DNA is redissolved in 20 pl of 10 mM Tris-UCl (pil 7.5) and 0.5 mM EDTA. 8-10 pg of hybrid plasmid DNA are recovered from a 10 ml culture.

E. coli HB 101 is transformed with cach of the nine isolated hybrid

DNAs and the transformed cells are plated on agar plates containing tetracycline, as described before (step D). From cach transformation, 3 tetracycline resistant clones are picked, 10 ml cultures are pre- pared and the hybrid DNAs are isovlated from the cultures as described before.

All the DNA samples before and after retransformation are analyzed by cleavage with Pst 1 endonuclease and electrophoresis through a 1% agarose gel in 50 mM Tris-acetate (pH 7.8) and 1 mM EDTA. All the samples display identical cleavage patterns before and after re- transformation.

One of the recloned recombinant DNA molecules gives 2 bands, one with the mobility of Pst I-cleaved pBR 322, the other with a mobility corresponding to about 1000 bp. It is denoted CG-pER 322/ULycIFN-1"b,

Another recombinant DNA gives 3 bands, one with the mobility of Pst I- Co cleaved pBR 322, one with a mobility of about 600 bp aud one with a’ mobility of about 150 bp. The recombinant DNA molecule in this clone 1s designated CG-pBR 322/MLycIFN-8, . ‘ d. Characterization of the clones CG-pBR 322/HLycIFN-1'b and

CG-pBR_322/ILyc IFN-p,

The recombinant plasmid DNAs of the clones CG-pBR 322/HLycIFN-1'b and

CC-pBR 322/MLycIFN-B, are isolated from the cultures as described above (step Ec) and characterized by establishing the nucleotide sequence of the cDNA insért using the method described by ilaxam and

Gilbert (15). Basically, the following approach is used: ” ORIGINAL 9

- 55 ~ ' 26266

The isolated recombinant plasmid DNA is digested with various restric-— tion endonucleases. The enzymes are applied essentially as described by the supplier (New England Biolabs), except that BSA is replaced by selatin in the enzyme buffers. The solution containing the restricted

DNA is deproteinized with phenol (saturated with TNE). The DNA is pre- cipitated with ethanol, vedissolved in 50 mM Tris-HCl (pH 8.0) at a

DMA concentration of 50 ng/ml and incubated with 0.1 units of calf intestinal alkaline phosphatase (Boehringer) per pmole DNA 5' ends for 30 min at 37°C. The enzyme is Inactivated by heating the solution for 60 win at 65°C. The DNA is purified by DEAE-cellulose chromatography as described by Mueller et al.(42) and precipitated with ethanol. The

DNA is then 5'-terminally labeled with [y=>2p)-ATP O 5000 Ci/mmole, .

Amersham) and T4 polynucleotide Kinase (P-L Biochemicals) essentially as described by Maxam and Gilbert (15) except that the DNA is not de- natured before the Kinase rcaction. In general, the specific activities amount to 1-3-10° cpm/pmole 5'-ends.

The labeled DNA fragments are cleaved with a second restriction endonuclease and the products are separated by clectrophoresis through a 6%, 8% or 10% polyacrylamide gel in Tris-borate-EDTA buffer. The

DNA fragments ave extracted from the gel and purified as described by Mueller et al. (42). For the determination of the nucleotide. co sequences, the DNA fragments are chemically degraded and the products oo are separated by polyacrylamide gel electrophoresis as described by

Maxam and Gilbert (15), ’

In particular, the isolated plasmid DNAs of the clone CG-pBR 322/

HLycIFN-L"b are treated as follows. On the one hand, 5 1e of the plasmid DNA is digested with Bgl II, 5' terminally labeled, and cleaved with Pvu IL. The Pve IL-Bgl IL* (*indicates the labeled site) and Bgl IT- Pvu 11% DNA fragments are isolated on a 67 polyacrylamide 30gel. On the other hand, aso omen J)

i —— - a 0 tt 26266 yg of the plasmid is digested with Alu I, 5'-terminally labeled, and cleaved with Pst I. The Pst I —- Alu I¥ DNA fragment is isolated on a 8% polyacrylamide gel. The individual fragments are subsequently : degraded and sequenced according to Maxam and Gilbert. The 5 nucleotide sequence obtained is depicted in figure 10. A stretch of about 25-35 deoxyguanosine residues is preceding at the 5'-end of the ¢DNA insert. The nucleotide sequence shown is somewhat similar to that of 1FN-a (type F) cDNA described by Goeddel et al. [(43), cf. also :

Weissmann(44)], nevertheless displaying a lot of distinct deviations (point mutations) some of which are affecting the resulting amino acids (cf. fig.10).

The isolated plasmid DNA of the clone CG-pBR 322/MLycIFN-f, is treated in a similar manner. 5 ng of the plasmid is digested with

Pvu IT and S'-terminally labeled. One half of the mixture is cleaved with Pst I, and the rest with Bgl II. The Pst I-Pvu IT*® and Bgl II-

Pvu LI[* fragments are isolated by electrophoresis on a 6% polyacryl- amide gel and degraded as meutioned above. The nucleotide sequence (N-terminal sequence) is depicted in figure 1land reveals ‘ that the cDNA insert starts at nucleotide number 102 of the IFN-0, ; ¢DNA as described by Taniguchi et al. (45). Therefore, the cDNA insert has the capacity to code for human IFN-B, lacking 11 amino acids at the N-terminus. The cDNA insert is flanked at its S'tend by a stretch of about 20-25 deoxyguanosine residues and shows a point mutation at ~ - oo position 153, converting a C to a T residue without affecting the vesulting amino acid.

Ce. Tdentitication off clones containing recombinant DHA molecules cross-hybridizing to the inserts of CC-pBR 322/HLycIFN-1'b and ompR 322 Lye EN-0

The recombinant plasmid DNAs of the clones CG-pBR 322/ULycIFN-1"b and CC-pBR 322/MLyclEN=0, are isolated from the cultures as des-— cribed above (step Ec). The CG-pBR 322/1iLycIFN-1'b plasmid DNA (5 pe)

EO ~ aso ORIGINAL 9

- 57 = 26266 is digested with Bgl 11, 5° terminally labeled, and cleaved vith

Pvu LT. On the other hand, the isolated CG-pBR 322ALycIFN-B, plasmid

DNA (5 yi) is digested with Pvu IT, 5'-terminally labeled, and cleaved with Bgl II. The Pvu II-Bgl II* (351 bp) D¥A fragment (probe A) and the Pvu IL#-Bgi IT (368 bp) DNA fragment (probe B) are isolated from a 87 polyacrytamide gel as described above (step Ed) and used for in situ colony hybridization (see below). The restriction of the plasmid DNAs, the labeling, and the purification of the DNA fragments are accomplished in the same manner as described above (step Ed). 4000 of the transformant colonies prepared as described above (step D) dre transferred to nitrocellulose filters BA 85 (Schleicher & Schuell, 8 cm diameter). The cells are lysed and their DNA is denatured and fixed to the filters in situ, according to Grunstein and llogness QO). llybridizations to the probes A and B (both probes are mixed) are per- formed as described before (step Ec). 6 positive colonies are identi- fied by autoradiography, 3 of which, designated

E. coli HB 101 CC-plR 32241LyeIrN-4, | .

E. coli HB 101 CG-pBR 322LycIFN-5, and

E. coli HB 101 CG-pBR J22/lILycIFN-8)

I Co. Lo arc used for further investigation. The plasmid DNAs of these clones are isolated, retransformed, and re-isolated as described above _ (step Fe, Ed).

In order to establish the nature of the inserts of the recombinant

DNAs, the nucleotide sequences of the cDNA inserts (partial or complete) are established by using the general approach as described above (step Ed).

In particular, Sng of the isolated plasmid DNAs CG-pBR 322/1LycIPN-4, and CG~pBR 322/1Lycl¥N-8) are cach digested with Pvu II, 5'-terminally oxo ORIGINAL 9 rT 26266 labeled and cleaved with Pst I. The DNA fragments are fractionated un a 8% polyacrylamide gel and the Pst I-Pvu IL* (~ 120 Lp) from 8)

DNA and Pst I-Pvu II* (82 bp) from 4 DNA are isolated as usual, }

The isolated plasmid DNA CG-pBR 322/1LycIFN=-5, is treated as follows.

On the one hand, 5 ng of the plasmid DNA is digested with Hae III, 5'-terminally labeled and cleaved with Pst I. The Pst I-llae LIL* (57 bp) DNA fragment is isolated on a 10Z polyacrylamide gel. On the other hand, 5 yg of the plasmid is digested with EcoR I, 5'-terminally labeled and cleaved with Pst I. The Pst I-EcoR I* (235 bp) and EcoR I*-

Pst I (~700 bp) DNA fragments are isolated on a 8% polyacrylamide gel,

The various DNA fragments are subjected to sequence analysis according to Maxam and Gilbert (15), °

The nucleotide sequences of the ¢DNA ifuserts are depicted in figures 12-14.In figure 12,a partial nucleotide sequence of the cDNA insert of CG-pBR 322/ULycIFN-4, is shown. The insert is flanked at the 5 end by a stretch of 23 deoxyguanosine residues and comprises part of the IFN-a, (Le) cDNA described by Streuli et al. (46).1n the 3'-extra- cistronic region, there are some minor deviations (point mutations) and a stretch of additional 318 nucleotides. The nucleotide sequence of : the cDNA tusert of CG-pBR 322/1Lyc1FN-8 is depicted in figure 13.The insert is flanked at the 5' end by a stretch of 20-23 deoxyguanosine residues and is similar but not identical, to the IFN-a (type D) cDNA described by Goeddel et al. [(43); cf. also Mantei et al. (27)). Apart from differences in the cDNA : regions preceding and following the IFN coding sequence, the IFN gene con- tains at positions 28-30 a GCC triplet and at positions 409-411 a GCG triplet coding for alanine instead of GTC and GTG, respectively, coding for valine.

Finally, the nucleotide sequence of the cDNA insert of CG-pBR 322/11 Lyc

LEN-5, (sce figure l4)reveals a stretch of 17 deoxyguanosine residues at the 5' end. The nucleotide scquence is related to that of IFN-a (type B)

CDNA described by Goeddel et al. (43). However, there are additional nu- cleotides at the 5' end of the ¢DNA insert of HLycIFN-5,, point mutations, excisions and insertions in the extracistronic region and in the IFN coding sequence, especially at positions 22 and 361-372, are evident as well, ) ono ORIGINAL 9

F. Synthesis of human interferons by F. coli containing human IFN- specific recombinant DNA molecules

The 5 clones which have been shown to contain human IFN specific re- combinant DNA molecules, namely :

E. coli IB 101 CG-pBR 322/HLyclEN-1'b

FE. coli HB 101 CG-pBR 322/0Lyc1FN-4 ,

E.coli NB 101 CG-pBR 322/lLyclFN-5,,

E. coli HB 101 CG-pBR 322/1lLycTFN-8" , and

E. coli HB 101 CG-pBR 322/MLycIFN-,, ave tested for IFN activity, which, in cach case, is accomplished in the following manner:

Cultures of the corresponding E. coli clone (30 ml suspensions) are grown in tryptone medium to an optical density (0D) of about 1. The cells are harvested and resuspended in 0.5 ml of an aqueous solution containing 30 mM NaCl and 50 mM Tris-HCl (pH 8.0). Lysozyme (Sigma) is added to 1 mg/ml. After 30 min at 0°C, the suspensions are frozen (liquid nitrogen) and thawed (at 37°C) 5 times, and centrifuged for min at 20 000 rpm in a SS 34 Sorvall rotor at 4°C. The supernatants are assayed for IFN activity using the cythopathic bioassay according 20 to Armstrong (32) as described in step Ac. The following activities are found:

Source of extract Co IFN activity -. Ca

EL. coli HB 10l containing recombinant DNA (10/1) - N

CG-pBR 322/ILyc IFN-1"b 0;0

CG-pBR 322/1LycIrN-4, 030

CG-pBR 322/lLycLFN-5 10 000310 000

CC-phlt 322Ayelri-8" 1003100

CG-pBR 322/11LycliN-§, 0;0 possibly, clones exhibiting no measurable IFN activities contain re- combinant DMAs in which the HuLyIFN-cDNA insert is in an improper orientation in regard to the direction of transcription. Therefore, the recombinant DNA of one such clone (CG-pBR 322AILycIFN-1'b) con- - ORIGINAL PD ae taining a full length cDNA insert is reoriented as follows:

The plasmid DNA of the clone E. coli HB 101 CC-pBR 322ULyclFN-1'b is isolated as described above (step Ec) and cleaved with Pst I. 0.5 Ne of the cleaved DNA in 20 pt of a buffer mixture, containing 20 mM Tris-UCL (pH 7.8), 10 mM MgCl, , 10 mM DIT, 25 mM NaCl and 50 $87 ml pelatin, is treated with 0.2 units of 14 DNA ligase (Biolabs) and 0.5 md ATP for 2 h at 15°C. I coli IIB 101 is transformed with the cDNA mixture as described above (:/ep D). Transformed colonies are selected on Mc Conkey agar plates supplemented with tetracycline and, subscquently, replica-plated to nitrocellulose filters. 4 bacterial ot colonies hybridizing to the 32 Jabeled Pvu II-Bgl II* fragment (351 bp) of the recombinant DNA CG-pBR 322/HLycIFN-1'b (cf. step Ee) are designated E. coli HIB 101 CG-pBR 322/HLycIFN~-1"b, to -1'b,.

Extracts of the 4 clones are prepared and tested for IFN activity as + 15 described above. The following activities are found:

Source of cxtract IFN activity

EE. coli HB 101 containing recombinant DNA (IU/ml) © CG-pBR 322/HLycIFN-1'b, 0;0

CG-pLR 322/LyclFN-1"b, 0;0

CG-pBR 322/ULycTFN~-1"b 0;0

CC-pBR 322MLycIFN-1"b, 30;30 .

Hence, the plasmid CC-pBR 322/lLycIfN-1"b, contains a cDNA insert ; : capable of directing the synthesis of a polypeptide with IFN activity.

G. Construction of recombinant plasmids capable of producing high levels of polypeptides with LEN activity .

I. Construction of CG-pRR (AP)/LyIFN-a-1 recombinant plasmid

In order to improve the IFN specific protein yield of the clone

E. coli HB 101 CG-pBR 322/ULycIFN-1'b, the following construction is performed as indicated schematically in figure 15.

BAD ORIGINAL 9

. 61 pe 262066 a. Preparation of the cDNA ilngert | :

The recombinant plasmid DNA (150 yg) of the clone E. coli IB 101 }

CG-pBR 322/HLycIFN-1'b is cleaved with Pst I (Biolabs) using standard procedures (cf. step Ed). Following phenol extraction and ethanol precipitation, the excised iusert is isolated by mcans of sucrose density gradient centrifugation (5-23%) in 50 wM Tris-1iCl (pH 8.0) and 1 mM EDTA. The centrifugation is performed at 35 000 rpm in a TST 41 rotor (Kontron AG) at 15°C for 16 h. 0.3 ml fractions are collected with an ISCO gradient collector at 1 ml/min. The fractions con- taining the small fragment (i.e. the insert) are pooled. The DNA is precipitated with ethanol as usual, and the precipitate is collected by centrifugation in a HB-4 rotor (Sorvall) at 10 000 rpm at 0°C for 10 min. The precipitate is redissolved in 60 pl 10 mM Tris-HCI (pit 7.5) and 0.05 mM EDTA. 30 pg DNA are recovered as determinad by measuring the optical density.

The insert DNA (10 ng) is digested with Hae III (Biolabs) and the fragments are fractionated on a 2% agarpse gel in a solution con- taining 50 md Tris, 50 mM boric acid, 1 mM EDTA and 0.5 pg/ml ethidium bromide. The largest DNA fragments, Hae III-Pst I (869 bp) and

Hae III-Hae III (82 bp, %- figure 15,fragments 3 and 4 respectively), are each excised from the gel, squirted! through a, hin needle ‘with, ~ a syringe into Syl of 0.15 M NaCl, 50 mM Tris+lICl (pH 8.0), 1 mM a

EDTA, and el Bdpovernight by shaking. The eluate is passed through a 100 ul DE-52 (Watman) Pasteur-pipette column to adsorb the DNA.

The co lyme is Mashed with 2 ml of the same buffer and the DNA is clutcqvith 400 nl of a solution containing 1.5 M NaCl, 50 mM Tris (pH 8.0) and 1 mM EDTA. The DNA is precipitated with 2 volumes of ethanol at ~20°C overnight. The precipitate is collected by centri- fugation in an Eppendorf centrifuge.

The Hae TII-Hae ITI DNA fragment (82 bp) is redissolved and digested with Sau 3A (Biolabs). The enzyme is heat-inactivated at 65°C for 30 min. Lye of the llae I1T-Pst I DHA fragment (869 bp) is added, 50 ORIGINAL 9

L .

Jo 62 — . ‘ 26266 ) the sulution is adjusted to 10 mM HCL, 10 ml DIT and 0.5 wil ATP, and T4 DNA ligase (Biolabs) is added to 30 units/pl reaction volume. }

The solution is incubated for 10 h at 15°C. Follouing extraction with phenol and chloroform, the mixture is fractionated on a 27 agarose pel in Tris-borate-EDTA in the presence of ethidium bromide. The Sau 3A-Pst I DNA fragment (cf. figure 15,fragment 5) is extracted as described before, precipitated with cthanol and redissolved in 10 pl of a solution containing 10 mM TriseHCL (pH 7.5) and 0.05 mM EDTA. b. Preparation of the DNA fragment containing the f-lactamase regulatory region (ApPr) of pBR 322

The plasmid pBR 322 is cleaved with Pst I (cf. step Cb) and treated with 4 units/ml of the exonuclease Bal 31 (Bethesda Research Lab.) at 30°C for 4-10 min to remove the f-lactamase coding segment.

A chemical DNA linker of the formula 5'-ATGTGTGATCACACAT~3' . is synthesized using the method described above (step Ea). The linker is added to the Bal 31 treated pBR 322 DNA using conventional means.

The resulting hybrid molecule is cleaved with the restriction endo- nucleases Bel I (Biolabs) and EcoR I. The digestion products are fractionated on a 8% polyacrylamide gel in Tris-borate-EDTA as Co oo described before (step Ba). DNA fragments (ApPr DNA fragments), migrating between 184 bp and 234 bp marker DNAs, are isolated as described above (step Ia), and precipitated with cthanol as usual.

The precipitate is redissolved in a solution containing 10 md

Tris<HCLl (pH 7.5) and 0.05 mM EDTA. c. Ligation of the ApPr DNA fragment to the cDNA insert

The solutions containing the ApPr DNA fragments and the cDNA insert are pooled. The mixture is adjusted to 10 mM MgCl, , 10 mM DIT and 0.5 mM ATP, and incubated with 30 units ul T4 DNA ligase (Biolabs) oo ano ORIGINAL dD

’ - 63 ~ ' 26266 at 15°C for 12 h. Following extraction with phenol and chloroform, the mixture is fractionated on a 1% low melting agarose gel (Biorad),

The obtained ApPr-cDNA fragment is Juined to the large fragment of

PBR 322 cleaved with both Pst I (Biolabs) and EcoR I (Biolabs) in the following manner. The gel piece, containing the ApPr cDNA frag- ‘ment (about 20 p1) is mixed with the Pst I-EcoR I fragment of pBR 322, melted at 65°C for 2 min, cooled to 37°C, adjusted to 0.5 mM ATP, mM DIT and 10 mM MgCl, , and incubated with 30 units pul of T4 DNA ligase (Biolabs) for 12 h at 15°C. 10 One tenth volume of a solution containing 100 mM Tris-HCl (pl 7.5), 100 mM CaCl, and 100 mf MgCL,, is added, the solution is heated for 10 min at 65°C to inactivate the ligase and cooled to 37°C. The solution is then taken to transform ca?*-treated E. coli IIB 101 as described above (step Dp) and plated onto Mc Conkey agar plates supplemented with 10 pg/ml tetracycline. The transformed colonies are screened for IFN activity (cf. step F). The clone synthesizing the highest level of IFN activity is selected and designated E. coli HB 101 CG-pBR(AP)/LyIFN-a-1. An activity of 40000 (IU/ml) is found wlrich represents a 1300 fold stimulation compared to the original clone

E. coli HB 101 CG-pBR322/HLycIFN-1'b.

The recombinant plasmid DNA of the clone CG-pRBR (AP)/LyIEN-a-1 is, ., - Co isolated from the culture as described above (step 3c) and character— ized by establishing the nucleotide sequence of the cDNA insert (IFN gene) and the p~lactamase regulatory region. The result is summarized in figure 16, iI. Construction of the recombinant plasmid CG-pBR (AP) /LyIFN-a-3 —_— me Prasmid MumpoR AN /LyIPN-a-3

The IFN specific protein yields of the clone E. coli HB 101 CG-pBR 322/

HLycIFN-8] is improved as follows (cf. fig. 17): a. Preparation of the DNA fragment containing the P-lactamase regulatory region from CG-pBR (AP) /LyIFN-a-1

I” ORIGINAL 9

- 64 ~ 26266

CC-pBR (AP) /LyIFN-a-1 DNA (100 ug) is cleaved with Hind ITT (Biolabs) and Bgl IT (Biolabs). Following phenol extraction and ethanol pre- cipitation, the excised DNA fragment is isolated by means of sucrose density gradient centrifugation (5-237) in 50 mM Trise HCL (pH 8.0) and 1 mM EDTA. The centrifugation is performed at 58 000 rpm in a TST 60 rotor (Kontron AG) at 15°C for 4 hours. 0.2 ml fractions are collected as described before. The fractions containing the small fragment (lind III-Bgl II) are pooled and the DNA is precipitated with ethanol as usual. The precipitate is redissolved in 80 pl 10 mM Tris<lCl (pH 7.5) and 0.05 wM EDTA. 16 pg DNA are recovered as determined by measuring the optical density.

The DNA fragment (Hind ILI-Bgl II) (4 pg) is cleaved with Sau 3A (Biolabs) and the digestion products are fractionated on a 6% poly- acrylamide gel in Tris-borate-EDTA as described before. The DNA frag- ments are stained in EtBr (0.5 pg/ml), the Hind ILI-Sau 3A DNA fragment (239 bp) is extracted and isolated as before. The DNA is precipitated with ethanol as usual. The precipitate is redissolved in 20 pi 10 mM Tris®

Cl (pH 7.5) and 0.05 mM EDTA. b. Preparation of the cDNA insert

The cDNA insert is excised from the racdombinant plasmid CG-pBR 322,

HLyclFN-8] as described above (section Ia). Ce .

The cDNA insert (2 pg) is digested with 2.5 units of Sau 3A (Biolabs) in 10 pg/ml FtBr and incubated at 37°C for 60 min. The digests are phenol extracted and the DNA is precipitated in ethanol as above. The

DNA fragments are fractionated on a 1.27 agarose gel in a solution containing 50 mM Tris, 50 mM boric acid, 1mM EDTA and 0.5 pg/ml ethidium bromide.

The second largest DNA fragment (Sau 3A-PstI: 693 bp) is extracted from the gel and purified as described in section Ta). The DNA is re- * ER i » . ono ORIGINAL 9

26266 i dissolved in 20 pl 10 mM Tris-HCl (pH 7.5) and 0.05 mM EDTA. ¢. Ligation of the liind TTT-Sau 3A DNA [ragment to the cDNA insert (Sau 3a-Tsen)

Equal amounts of both DNA fragments (~50 ng) are incubated in a solution containing 10 mM MiCl,, 10 mM DTT, 0.5 mM ATP and 30 units/pl

T4 DNA ligase (Biolabs) at 15°C for 3 hours. The mixture is incubated for 15 min. at 80°C and adjusted to 50 mM NaCl. The DNA mixture is digested with 0.5 units PstI (Biolabs) and 1 unit lind III (Biolabs) for 20 min, at 37°C. The DNA is phenol extracted, ethanol precipitated and redissolved in 20 pl 10 mM Tris-HCl (pl 7.5) and 0.05 mM EDTA.

One half of the resulting mixture is ligated to the large Hind III-Pstl

DNA fragment of the plasmid pBR 322 (~~100 ng) in 10 mM MgCl,, 10 mM

DTT, 0.5 mM ATP containing 30 units/ul of T4 DNA ligase (Biolabs) for 2h at 15°C.

One tenth volume of the solution is used to transform E. coli HB 101 as described in step D). The transformed colonies are used to test for

IFN activity as described earlier (cf. step F).

The clone synthesizing the highest level of IFN activity is selected and designated E. coli HB 101 CC-pBR (AP) /LyIFN-oa-3. . Doe ee no i * . SU

The IFN activity is determined as described above (step F). An activity of 70000 (IU/ml) is found which represents a 700 fold stimulation com- pared to the original clone E. coli HB 101 CG-pBR 322/MLycIFN-8,.

The recombinant plasmid DNA of the clone CG-pBR (AP)/LyIFN-a-3 is isolated from the culture as described above (step Cc) and charac- terized by establishing the nucleotide sequence of the cDNA insert (IFN gene) and the p-lactamase regulatory region. The result is summarized in figure 18. . ’

BAD ORIGINAL 9

OR

’ - 66 ~ : 26266

The construction protocol for the plasmid CG-pBR (AP) /LyIFN-a-3 can be used (or all a~IFN cDNA genes or appropriately cut chromosomal «-IFN genes in general.

For example, starting from the plasmid CG-pBR 322/ULycIFN-5 , the plasmid

CC-pBR (AP) /LyIFN-a-2 is obtained in an identical manuer as described for the plasmid CG-pBR(AP)/LylFN-a-3. This new plasmid contains the

DNA insert of CG-pBR 322/1LycliN-5, and the f-lactamase regulatory region from CG-pBR (AP)/LyIFN-a-1. A clone designated E. coli HB 101

CG-pBR (AP) /LyIFN-a-2 is sclected as described above. An IFN activity of 50000 (IU/ml) is found which represents a 5 fold stimulation com- pared to the original E. coli HB 101 CG-pBR 322 ULycIFN-5, . The nucleo- tide sequence of the cDNA insert and the f-lactamase regulatory region of the plasmid CC-pBR (AP) LylFN-a~2 is established as described above and depicted in fig. 19. 1II. Deposition of prepared microorganisms

Micro-organisms and recombinant DNA molecules prepared as described in

Example 10 are exemplified by cultures depusited in the culture collec- tion of the Agricultural Research Culture Collection (NRRL) on September 14, 1981 and are assigned the following accession numbers:

Co tan

E. coli UB 101 CG-pBR 322/HLyclFN-f : NRRL B-12528 Ce Co

E. coli HB 101 CG-pBR 322/HLycIFN-4 : NRRL B-12529 Co

E. coli HB 101 CG-pBR 322/1LyclFN-1'b: NRRL B-12530

E._coli WB 101 CG-pBR 322/NLycIFN-5 : NRRL B-12531 :

E. coli UB 101 CG-pBR 322/ULyclFN-8 1 RRRL B-12532 a0 OIG 9

Example 11: The lymphoblastoid IFN-1'b and IFN-5, coding sequences of the E. coli plasmids CG-pBR322/HLycIFN-1'b and 5, (cf. Example 10) can be subcloned in plasmid p30 (cf. Example 4) in an analogous manner as described for IFN-8) in Examples 5 and 6. A partial digestion with restriction endonuclease Haelll is required. The yeast hybrid plasmids obtained in this way are p30IFN2(1'b), p30IFN2'(1l'b), p30LFN2(5,) and p30IFN2'(5,).

These obtained hybrid plasmids can be used to transform Saccharomyces cerevisiae RH971 as described in Example 7. The following colonies containing a hybrid plasmid with a lymphoblastoid IFN cDNA insert can be selected:

S. cerevisiae RH97L/p30IFN2(1'Db)

S. cerevisiae RH971/p30IFN2'(1'b)

S. cerevisiae RH971/p30LFN2(5,) «

S. cerevisiae RH971/p30IFN2' (5,)

Example 12: In an analogous manner as described in Example 9, the following yeast hybrid plasmids can be obtained starting from the plasmids p30IFN2(1'b), =(5,), -(8)) and p30IFN2'(1'b), =(5.), respectively: pJDB207/1FN2(1'b), pJDB207/IFN2' (1'b), pJDB207/TFN2(5,), pJDB207/TFN2' (5,) and pJDB207/1FN2(8.) oe

These hybrid plasmids can be transformed into S. cerevisiae strain

RHI71 selecting for leucine prototrophic colonies. The following colonies containing a hybrid plasmid with a lymphoblastoid IFN cDNA insert can be obtained:

S. cerevisiae RH971/pJDB207/IFN2(1'b)

S. cerevisiae RH971/pJDB207/IFN2'(1'b)

S. cerevisiae RH971/pJDB207/IFN2(5,)

S. cerevisiae RH971/pJDB207/IFN2'(5 )

S. cerevisiae RH$71/pJDB207/1FN2(8))

Example 13: Construction of an expression plasmid containing the PHOS promoter and I'lI05 transcription termination signals (see fig. 20) a) Elimination of the EcoRI restriction site in plasmid p30:

The scheme outlined in fig. 20-22 requires elimination of the unique

EcoRI restriction site in plasmid p30. Five mg of p30 DNA (cf.

Example 4) are digested to completion with restriction endonuclease

EcoRI (Boehringer). In order to fill in the resulting sticky ends, - lug of EcoRI digested p30 in 50 ml of 50 mM NaCl, 10 mM TriseHCl pl 7.5, 10 mM MgCl, , 1 mM DTT, 0.25 mM dATP and 0.25 mM dTTP is incubated for 30 min 37°C with 1 unit of DNA polymerase (Klenow large fragment, BRL). The DNA recovered from ethanol precipitation is ligated as usual and used for transformation of competent E. coli

HB101l cells as described in Example 4. Clones that are resistant to . R

EcoRI digest are referred to as p30/EcoRI . b) Isolation of a 0.37 kb Sau3A-PstI PHOS transcription termination fragment: .

The PHOS transcript has been mapped by S1 nuclease mapping (48). The signals for transcription termination have been shown to be located in a 0,37 kb Sau3A-PstI fragment of the PHO5 gene. The nucleotide sequence of the Sau3A-Pstl fragment is given in fig. 21. .

Five ug of pJDB207,/PHUS, PHO3 DNA (cf. Example 2) are digested to comple tion with restriction endonucleases Sau3A and PstI. The restriction fragments are separated on a vertical 1.5% low melting agarose gel in

TBE buffer. The 0.37 kb Sau3A-PstI fragment is localized by ethidiumbromide staining and a gel block as small as possible is cut out containing this DNA fragment. ¢) Cloning of the Sau3A-PstI PHOS fragment in M13mp9:

M13mp9 phage DNA is a useful cloning vector with a cluster of unique restriction sites (49). Five ug #. M13mp9 DNA are digested to completion with restriction endunucleases BamHI and PstI. The larger

’ - bY ~ 26266 7.2 kb DNA fragment is separated from a very small fragment (8 bp) on a 0.8% low melting agarose gel. The gel block containing the large

DNA fragment is cut out of the gel. Gel blocks with the 0.37 kb

Sau3A-PstI fragment of pJDB207/PHO5,PHO3 (cf. Example 13b) and the 7.2 kb BamHI-PstI fragment of M13mp9 are liquefied at 65°C, mixed in about equimolar amounts and diluted with H,0 to lower the agarose concentration to 0.3%. Ligation is carried out in a 200 ul solution containing 60 mM Tris.HC1 pH 7.5, 10 mM MgCl, 10 mM DTT, 1 mM ATP and 600 units of T4 DNA ligase (Biolabs). Transduction of competent cells of the strain E. coli JM101 (Ca++) is done according to the manual '"M13 cloning and DNA sequencing system" published by New England Biolabs.

Phages from a number of white plaques are grown and analyzed for the size of their DNA insert by cleavage with restriction endonucleases

EcoRI and PstI.

A M13mp9 derived clone containing the Sau3A-Pstl PHOS transcription termination fragment is isolated and referred to as M13mp9 PHOS (Sau3A-Pstl). d) Cloning of the PHOS transcription termination fragment in p30,/EcoRLt

The original PHOS transcription termination fragment cloned in phage

M13mp9 (M13mp9,PHOS5(SaulA-Pstl)) is recloned as a HaeIlI-HindIII fragment in plasmid p30 EcoRI cleaved with Ball and IindII1: no

M13mp9/PHO5 (Sau3A-PstI) DNA is cleaved to completion with restriction - endonucleases HaeIIl and HindIII. The resulting two DNA fragments are separated on a 1.5% vertical low melting agarose gel in TBE buffer.

The 0.39 kb fragment is isolated in a gel block cut out of the gel. p30 EcoRI DNA is digested with Ball and HindIII. The large 3.98 kb fragment is separated on a 0.87 low melting agarose gel in TBE buffer and isolated by cutting a gel block containing the DNA fragment.

Gel blocks with the 0.39 kb HaeIII-HindIII PHOS transcription temination fragment and the 3.98 kb Ball-HindIII fragment of p30/

EcoRI" are melted at 65°C and mixed in about equimolar amounts. Liga- (peaam— Co

BAD O° A —

tion and transformation of competent E, coli HBIOL cells are as described in Example 4. DNA of transformed cells is analysed by cleavage with Ball and HaelIl. A clone containing the PHOS transcrip- tion temnination fragment is further analyzed and referred to as p31 (see Figure 20). .

Expression plasmid p31 contains the PHOS promoter region with part of the signal sequence of P05 and adjacent to it a DNA fragment with the PHO5 transcription termination signals, Foreign coding sequences to be expressed in this vector may conveniently be inserted between promoter and transcription termination sequences.

Example 14:Insertion of lymphoblastoid interferon-3, DNA into : plasmid p31 (see Figure 22) a) Isolation of HaelII-Hpal fr ,ments of plasmid CG-pBR322/HLycIFN-5,

E. coli strain HB1O1 CG-pBR322/HLycIFN-5, (see Example loE) is grown in 100 ml LB medium supplemented with 10 ug/ml tetracyclin and the plasmid DNA is isolated as described in Example 2. Ten mug of

CG-pBR322/HLycIFN-5, DNA are completely digested with restriction endonucleases PstI and Hpal. The restriction fragments are separated on a preparative 0.87 low melting agarose gel. The PstI-Hpal fragment of about 860 bp containing the IFN-5, coding sequence is cut out of the gel and eluted from the agarose gel as described in Example” ha and purified by DE52 ion exchange chromatography as detailed in

Example 5a.

The Pstl-Hpal fragment contains 3 Haelll sites: at position 41, 65 and 146 (from the ATG) in the IFN-5, coding sequence. Partial HaeIll digestion leads to three HaeIII~Hpal fragments of 699 bp, 780 bp and 804 bp, respectively. Haelll digestion is carefully adjusted to obtain about equal amounts of all three fragments. The mixture of fragments is phenol extracted, ethanol precipitated and resuspended in 10 mM Tris pH8 at a concentration of 0.1 mg/ml.

Ino ano ORIGINAL J b) Preparation of Ball cleaved, dephosphoryiated plasmid p31

Six jig of p31 DNA (cE. Example 13d) are completely digested with re- striction endonuclease Ball (BRL). After phenol extraction and ethanol precipitation the DNA is redissolved in 100,ul of 50 mM Tris pH 8.0 and passed through a 50 ul bed of equilibrated Chelex 100 (BioRAD) in a siliconized Pasteur pipet. The flow through and 450 ml of subsequent wash are combined. 0.4 units of calf intestine alkaline phosphatase (Boehringer) are added. After 1 h of incubation at 37°C the enzyme is inactivated at 65°C for 1.5 hrs. The NaCl concentration in the incubation mixture is adjusted to 150 mM. The linearized dephosphory-— lated p31 DNA is purified by DE52 ion exchange chromatography (see

Example 5a). After ethanol precipitation the DNA is resuspended in 10 mM Tris pH 8 at a concentration of 0.3 mg/ml. ¢) Ligation of linearized, dephosphorylated p31 DNA to the

HaeIII-Hpal fragments of IFN-5, DNA 0.6 ug of dephosphorylated p31 vector DNA cleaved with Ball is ligated to 0.5 mug of partial HaeIlI-Hpal fragments of IFN-5, DNA (see .

Example 14a). Ligation is carried out in 10 ml of 60 mM Tris pH 7.5, 10 mM MgCl,, 10 mM DTT, 4 mM ATP and 300 units of T4 DNA ligase (Biolabs) overnight at room temperature. A 1 ul aliquot of the ligation mixture is added to 50 ul of calcium treated transformation competent E. coli HB1Ol cells. The trans formatibn protocoll is as : described in Example 4a. Co

Transformed, amp" colonies are grown individually in LB medium con- taining 100 pg/ml ampicillin. Plasmid DNA is prepared according to the method of Holmes et al., (50) and analysed by digestion with re- striction endonuclease BstEII (one unique site in the PHO5 promoter). 20 clones containing the IFN-5, insert are further analysed by

BstEII-EcoRI double digests to determine the orientation and size of the insert. Among 8 clones with the insert in the right orient- ation all 3 expected insert sizés are found. The size corresponds .

BAD ORIGINAL J

PE

26266 to the three HaeIlIl, ,-Hpal fragments created by partial HaeIIl digest of the IFN-5, gene (cf. Example 14a). The clones’ are referred to as p31/IFL(5,), p3L/LEZ(5,) and p3L/TE3(5,) with IFN-5, inserts of 804 bp (HaeI1l, -Hpal insert), 780 bp (HaelIl,-lpal) and 699 bp (Nae [11 ,-Hpal), respectively.

Example 15: Insertion of lymphoblastoid interferon—1'b DNA into plasmid p3l (see Fig. 22) a) Isolation of HaeIII-Rsal fragments of plasmid CG-pBR322/MLycIFN-1'b:

Ten ug of CG-pBR322/MLycIFN-1'b DNA (see Example 10E) are digested with restriction endonucleases Pstl and Rsal. The restriction frag- ments are separated on a 0.8% low melting agarose gel. A PstI-Rsal fragment of about 870 bp is isolated from the gel and purified as described above (Example l4a).

The PstI-Rsal fragment contains three HaeIll sites: at positions 13, 37 and 118 from the ATG of the IFN-1'b coding sequence. Partial :

HaelIl digestion leads to three HaelII-Rsal fragments of 735 bp, 816 bp and 840 bp, respectively. The mixture of fragments is phenol extracted, ethanol precipitated and resuspended in 10 mM Tris pH 8.0 at a concentration of 0.1 mg/ml. b) Ligation of linearized, dephosphorylated p31°DNA to HaeIII-Rsal fragments of IFN-1'b DNA - ; 0.6 jug of dephosphorylated p31 vector DNA cleaved with Ball (see

Example 14b) is ligated to 0.5 ug of partial Haelll-Rsal fragments of IFN-1'b DNA. The ligation procedure, the transformation of compe- tent E, coli HB 101 cells with the ligation mixture and the selection of the transformed amph colonies is carried out as described in Example l4c. Plasmid DNA is prepared according to the method of Holmes et al. (50) and analysed by digestion with restric- tion endonulcease BstEII.

So aw ORIGINAL 9

- 73 - Lo 266 26 7 clones containing the IFN~i'b insert are analysed by BstEI1-Pvull double digests. Two clones are shown to contain the Haelll,-Rsal fragment (816 bp) in the right orientation. This construction is referred to as p31/IF2(1'b).

Example 16: Insertion of lymphoblastoid interferon-8! DNA into plasmid p31 (see figure 23) a) Isolation of a 1.46 kb SalI-EcoRI fragment of plasmid p30TFN1(81)

Five pg of pIOLENL(8}) DNA (see Example 5d) is digested with re- striction endonucleases Sall and EcoRI. A 1.46 kb Sall-EcoRI fragment, containing the PHO5 promoter linked to the protein coding region of IFN-8; is separated on a 0.87 low melting agarose gel,

The DNA band is localized by ethidium bromide staining and cut out of the gel. ’ b) Isolation of a 3.5 kb Sall-EcoRI fragment of plasmid p31

Five mug of p31 DNA (see Example 13d) are completely digested . with restriction endonucleases Sall and EcoRI. The 3.5 kb vector fragment containing the PHOS transcription termination sequence is’ separated on a 0.8% low melting agarose gel and the DNA band is cut out,

Co . Ceo ¢) Ligation of a 1.46 kb Sall"EcoRL fragnent of ,3orNi(8}) to a 3:5 kb

Sall-EcoRI fragment of p31

Gel blocks with 0.67 ug of the 3.5 kb Sall-EcoRI fragment of p31 and 0.5 pg of the 1.46 kb Sall-EcoRI fragment of p30IFN1(8)) are ligated in 240 ml as described in Example 4a at 15°C overnight. 10 ul of the ligation mixture are used to transform competent

E. coli 1IB101 cells.

Transformed, amp colonies are grown individually in LB medium con- taining 100 ug/ml ampicillin. Plasmid DNA is prepared according to the method of Holmes et al. (50) and analysed by digestion with restric-— tion endonuclease BstEIT (one unique site in the THOS promoter).

- 74 = t 26266

A number of clones containing the LFN-8) insert. are analysed by BstEII~-

Pvull double digests. They all contain the 1.46 kb SalIl-fcoRI fragment,

The identical clones are referred to as PIL/IF(8]).

Example 17: Subdloning of gene constructions in the high copy number yeast vector pJDB207

The constructions described in Examples 14-16 contain the PHOS promoter, different interferon coding regions and the PHOS trans- cription termination signals in a tandem array, all inserted in a

PBR322 derived vector. For expression in yeast the whole insert is subcloned as such in yeast vectosd pJDB207 (28) allowing selection for leucine prototrophic colonies (cf. Example 9 and fig. 6). 2 pg each of p31/1F(8}) DNA, P3L/IF1(5)) DNA, p31/IF2(5 ) DNA,

P3L/IF3(S,)) DNA and p31/1F2(1'b) DNA are digested with restriction endonucleases Sall and HindIII. The restriction fragments are separated on a preparative 0.8% low melting agarose gel. The small fragment (~2 kb in size) of each digest is cut out of the gel. ° 10 pug of pIDB207 DNA are digested with restriction endonucleases

Sall and HindIII. The large 6.2 kb fragment is isolated from a preparative 0.87 low melting agarose gel. Gel blocks containing, the DNA fragments are liquified at 65°C and diluted with Ho'to lower the agarose concentration to about 0.3%. - :

Each of the 2 kb SalI-HindIII fragments of the plasmids p3L/IF(8)), p3L/IFL(S,), P3L/IF2(5,), P31/IF3(5,) and p31/1F2(1'b) is mixed with an equimolar amount of the 6.2 kb HindIII-Sall fragment of pJDB207.

Ligations are carried out in 100 ul for 4 hrs at 15°C. 10 jl of each ligation mixture are used to transform competent E. coli HB101 cells as described in Example 4a. Several amp’ colonies from each experiment are grown individually in LB medium containing 100 ug/ml of ampicillin. The plasmid DNA is analysed for the size of the insert

: 90266

A number of clones containing the IFN-8] inscrt. are analysed by BstEII-

Pvull double digests. They all contain the 1.46 kb Sall-ZcoRI fragment.

The identical clones are referred to as p3L/TIF(8)).

Example 17: Subcloning of gene constructions in the high copy number yeast vector pJDB207

The constructions described in Examples 14-16 contain the PHOS promoter, different interferon coding regions and the PHOS trans- cription termination signals in a tandem array, all inserted in a

PBR322 derived vector. For expression in yeast the whole insert is subcloned as such in yeast vectol pJDB207 (28) allowing selection for leucine prototrophic colonies (cf. Example 9 and fig. 6). 2 jg each of p31/IF(8)) DNA, p31/IF1(5,) DNA, p3L/TF2(5,) DNA, p31/IF3(5,) DNA and p31/IF2(1'b) DNA are digested with restriction endonucleases Sall and HindIII. The restriction fragments are separated on a preparative 0.87 low melting agarose gel. The small fragment (~2 kb in size) of each digest is cut out of the gel. ° 10 pug of pJDB207 DNA are digested with restriction endonucleases

Sall and HindIII. The large 6.2 kb fragment is isolated from a preparative 0.87 low melting agarose gel. Gel blocks containing. the DNA fragments are liquified at 65°C and diluted with 0 to oo lower the agarose concentration to about 0,3%.

Each of the 2 kb SalI-HindIII fragments of the plasmids p3L/IF(8)), p3L/IF1(5,), p3L/IF2(5,), p3L/IF3(5,) and p31/IF2(1'b) is mixed with an equimolar amount of the 6.2 kb HindIII-Sall fragment of pJDB207.

Ligations are carried out in 100 mul for 4 hrs at 15°C. 10 ml of each ligation mixture are used to transform competent E. coli HB101 cells as described in Example 4a. Several amp® colonies from each experiment are grown individually in LB medium containing 100 ug/ml of ampicillin. The plasmid DNA is analysed for the size of the insert by cleavage with restriction endonucleases ind{II and Sall. The resulting clones with the correct inserts are named pINB207/IF (4), pJDB207/1F1(5,), pJDB207/1F2(5,) (cf. fig. 27), pJDB207/TF3(5,), and pJDB207/1IF2(1'b) (cf. fig. 27).

Example 18: Transformation of Saccharomyces cerevisiae AH220 and induction of interferon production:

Plasmids pJDB207/IF(8}), pJDB207/IF1(5,), pJDB207/IF2(5,), pJDB207/1F3(5,) and pJDB207/1F2(1'b) are each introduced into

Saccharomyces cerevisiae strain AH220 (a, trpl, leu2-3, leu2-112, his3, pho5, pho3) using the transformation protocoll described by

Hinnen et al. (1). Transformed yeast cells are selected on yeast minimal medium plates deficient in leucine. Single transformed yeast colonies are picked and grown as described in Example 7. The different yeast colonies are referred to as

Saccharomyces cerevisiae AH220/pJDB207/IF(8)),

Saccharomyces cerevisiae AH220/pJDB207/1F1(5,),

Saccharomyces cerevisiae AH220/pJDB207/1F2(5,), .

Saccharomyces cerevisiae AH220/pJDB207/IF3(5,) and

Saccharomyces cerevisiae AH220/pJDB207/1F2(1'b)

Example 19: Preparation of yeast cell extracts and determination of the interferon titer: f IE hen Co

Cell extracts are prepared and interferon activity is determined as a described in Example 8.

The results are summarized in Table 3. ’ oo - vo]

— 76 a

Interferon activity in Saccharomyces cerevisiae strain AHz20 after transformation with the following recombinant plasmids:

Plasmids Interferon activity units/ml yeast cell extract pJDB207/1F (8!) 1 + 107 pJIDB207/TF1(5,) 7. 10° pIDB207,/1F2(5 ) 5 10° pJDB207/1F3(5,) 3. 103 pJDB207,/1F2(1'b) 4 + 10°

Example 20: Expression of hepatitis B virus surface (HBVs) antigen under the control of the yeast PlIO5 promoter a) Construction of a fusion between the PHO5 promoter and the HBVs protein coding region ’ 4 . CL EE 5 pg DNA of plasmid pHBV130 (51) is digested with restriction - endonuclease Aval as recommended by the supplier (New England Biolabs).

A fragment of 1336 base pairs is obtained which contains the entire protein coding region of HBVs, including 27 base pairs of a potential pre~-HBVs sequence (see Fig. 2 in ref. 51: the Aval fragment spans the

DNA segment from the Xho site until an Aval site 62 base pairs beyond the BamHI site located between the surface coding region and the core coding region). The 1336 base pair fragment is purified by soft agarose electrophoresis (0.87 agarose gel) as described in Example 4a.

BAD ORIGINAL 9

5g of plasmid pBR322,PHUS Bam-Sal (see fig. 1) is cut with restric- tion endonucleases Sall and Aval (position 1424 of pBR322) and the resulting 3.9 kb vector fragment coniaining pBR322 sequences together with the PHOS Bam-Sal segment is purified by soft agarose electrophoresis as described above. 1 ug of the 1336 base pair fragment is ligated to 3 ug of the 3.9 kb vector fragment in 50 pul of 60 mM Tris-HCl pH 7.5, 10 mM MgCl, , mM DTT, 1 mM ATP and 600 units of T4 DNA ligase (Biolabs) at 15°C for 4 hours. Transformation of E. coli HB1Ol to ampicillin resistance 10 and plasmid isolation is carried out as described in Example 4a. ’

The correct structure of the plasmid is verified by restriction analysis. The new plasmid thus constructed is called pBR322/PHO5/HBVs (see fig. 24), b) Adjustment of the PIIO5 promoter to the exact HBVs protein coding region

The fusion described creates a DNA sequence arrangement as depicted in fig. 25. The sequence data are taken from fig. 3 (PHOS) and from

Pasek et al. (52; HBVs). The mRNA initiation site is determined by conventional S1 mapping (48) using the BamliI-Sall fragment of pBR322/PHO5 Bam-Sal (fig. 1). In-order to eliminate the PHOS protein coding region present in pBR322/PHO5/UBVs 5 ug of the plasmid is digested with restriction endonuclease KpnI (conditions specified by supplier, New England Biolabs) which produces a linearized plasmid. 4 mg of linearized plasmid is digested with 1 unit of exonuclease Bal3l (Bethesda Reasearch Laboratory) at 30°C for 45 sec in 12 mM CaCl,, 12 nM

MgC1,, 300 mM NaCl, 20 mM Tris, 1 mM EDTA pH 8.1 in a total volume of 100 mul. The reaction is stopped by phenol extraction as described above. After ethanol precipitation the DNA is resuspended in 50 mul TE. - ‘ wo [BAD ORIGINAL 9

- 78 - 9 6 20 6 1 jug of DNA is recircularized by ligation with T4 DNA ligase in a volume of 20 mul (conditionc see Example 4a). After transformation of

E. coli UBLOl to ampicillin resistance (see Example 4a) plasmid

DNA is isolated as described and individual plasmid preparations are analysed by restriction analysis with the following enzymes: Haelll,

Pstl, BstEII and Hhal. This analysis allows the determination of the presence of the Hhal site (6 base pairs before the start of the HBVs protein coding region) and gives a measure for the size of the deletion. The DNA sequence in the junction area is determined using the method of Maxam and Gilbert (15) (radioactive labelling at the BstEII site at position-374, see fig. 3). The endpoints of the deletion generated in one of the plasmids are indicated in fig. 24,

This plasmid is called pBR322/PHO5/HBVsAl4. ¢) Transfer of the PHO5-IBVs fusion to the yeast plasmid pJDB207 (see fig. 26) 5 mg DNA of plasmid pBR322,/PHO5/HBVsAL4 is digested with restriction endonuclease BamHI (New England Biolabs, conditions as described by supplier). A 1.8 kb BamHI fragment is prepared by soft agarose gel electrophoresis (0.87 agarose) as described in Example 4a. 2 pg of the yeast vector pJDB207 is digested with the same enzyme. 1 ug of digested pJDB207 and 1 ug of the 1.8 kb BamHI restriction fragment is ligated in a total volume of 20 ul using the conditions described oo in Example 20a. Transformation of E. coli HB1O1 to ampicillin’ RE : resistance and isolation of plasmid DNA is carried out as described above (Example 4a). Individual plasmids are analyzed by BamHI restriction analysis and the orientation of the inserted Bamll fragment is determined by HindIIL/BstEII double digestion. Fig. 26 outlines the construction. The plasmid obtained as indicated in fig. 26 is called pJDB207/PHO5/1BVsAl4. ’

The plasmid is transformed into yeast strain AH220 as described in

Example 7. Transformed yeast cells are selected, incubated in liquid medium and grown under derepressing conditions as described in

Example 7. A single transformed yeast colony is referred to as

Saccharomyces cerevisiae AH220/pJDB207/HBVsalé4. ano ORIGINAL J

Preparation of cell extract is donc as described in Example B and the amount of liBVs protein produced is determined using the Abbotr radioimmun assay (51). Under the assumption that the HBVs antigen produced by yeast reacts similarly to the antigen in human seruu, about 2 pg HBVs antigen per ml yeast extract are found under derepres- sing conditions. Under repressing conditions the titer is below 0.001 mg ml. d) Transfer of the PHOS5-1IBVs fusion to a yeast plasmid containing a PHOS transcription termination sequence 5 mg of DNA of plasmid pJDB207/1F (8) (Example 17) is digested with

BamHI as described above and the 6.9 kb vector part is isolated by soft agarose electrophoresis (0.8% agarose) as described in Example 4a. 5g of pBR322 /PHO5/HBVsAl4 is digested with BamHI as described above and the 1.8 kb BamHI fragment is isolated by soft agarose gel electrophoresis. 1 ug of the 6.9 kb vector fragment is ligated with lug of the 1.8 kb BamHI fragment in a total volume of 20 ul using the conditions described in Example 20a. Transformation of E. coli

HB1Ol to ampicillin resistance and plasmid isolation is done as - described in Example 4a. Plasmid analysis is performed by restriction endonuclease digestions. The plasmid obtained as indicated in fig. 26 is called pJDB207./PHO5/1BVsAl4t. Transformation of yeast strain

AH220 and selection of the transformed cells is done as described .. } in Example 7. A single transformed yeast colony i$ referred to as

Saccharomyces cerevisiae AH220/pJDB207/HBVsAl4t. ;

Example 21:

Hepatitis B virus (HBV) DNA sequences excised from plasmids pBR322-Pst1 dG:HBV-KpnI dC; pBR322-PstI dG: HBV-BamHI dC; pBR322-pPstl dG:HBV-pg11I dC; pBR322-Pstl 4G:HBV-EcoRI dC; pBR322-BamHI:: HBV-BamHI; pBR322-EcoRI: HBV-EcORI; pBR322-Pstl dG:l1BV-Kpnl dC, aso ORIGINAL D

.—- 80 =

PBR322-PstI' dG:pHBV1l4~PstI dC; 2 6 26 6 ((pBR322-EcoRI HindIIL: Lac promoter sequence) -HindIII:HBV114-Hhal

HindIII linkers)-BamHI,

PUR2-EcoRI: HBV1l4-Hhal EcoRI linkers,

PpUR2-EcoRI: HBV11l4-Hhal EcoRI linkers;

PBR322-Pst1 dG:pHBV1l4-Aval dC, and : pBR322-Pstl dG:pHBV114-Taq dC. as described in European Patent Application 13828 tan be inserted (preferably after appropriately adapting the termini) into plasmids p30 or p31 according to Examples 5, 14 or 15 and subsequently into plasmid pJDB207 according to Fxample 17. Transformation of

S. cerevisiae is performed according to Example 18. Expression of polypeptides displaying HBV antigenicity is determined according to . Example 20c.

Example 22: Deletion of 3' nontranslated DNA sequences in plasmids pJDB207/1F2(5,) and pJDB207/1F2(1'b) (see figures 27 and 28)

The construction of the plasmids pJDB207/1F2(5,) and pJDB207/IF2(1'b) (cf. Example 17) resulted in a relative long 3' nontranslated region of about 440 bp and 480 bp, respectively. To shorten this region of the -constructs the DNA is digested with exonuclease. Bal3l from an unigye - yo

Smal site in the middle of this region. Xho linkers are introduced and the DNA is circularized by ligation. : " pJDB207/1F2(1'b) 20 pg each of the plasmid DNAs are digested with restriction endo- nuclease Smal. After extraction with phenol/chloroform, the DNA is pre- cipitated with ethanol. The DNA is resuspended in 10 mM Tris pH 8.0 at a concentration of 0.5 mg/ml. 10 re of the Smal cleaved DNAs are each digested with 2 U of endonuclease Bal3l (BRL) in 100 pl of 20 mM Tris pH 8.0, 100 mM NaCl, 12 mM MgCl,, 12 mM CaCl, and 1 mM

EDTA. Aliquots of 3 pg of DNA are withdrawn after 90, 120 and 150 seconds of incubation at 30°C and are immediately mixed with 50 ul of phenol and 60 pl TNE. After extraction with phenol/chloroform and ethanol precipitation, the DNA is resuspended in 10 mM Tris pH 8.0 at a concentration of 100 pg/ml. Tc analyse the exonucleolytic digestion of Bal3l, aliquots of 0.7 pug of DNA from each time point are digested with HindIII/EcoRI for pJDB207/1F2(5,) derived samples or with

Pvull/HindII1 for pJDB207/1F2(l'b)-derived samples. For further experiments the DNAs from the 90 second time points are used. b) Addition of Xhol linkers to the Bal3l treated DNAs 2.2 pug each of plasmid DNA PpJIDB207/1F2(5) and pJDB207/1F2(1'b), after 90 sec. of Bal3l digestion (see Example 22a) are incubated for 1 hour at 37°C with 2.8 U of Klenow DNA polymerase (large fragment of polymerase I, BRL) in 35 pl of 60 mM Tris pH 7.5, 10 mM MgCl, and 0.1 mM dNTP's.

Three pg of Xhol linkers (5'-CCTCCAGG-3', Collaborative Research) are kinased in 50 pl of 6 mM Tris pH 7.5, 10 mM MgCl, , 4 mM DTT, 0.5 mM

ATP and 35 U of T4 polynucleotide kinase (Boehringer) for 30 min at 3i°c. 0.67 y8 of kinased Xhol linkers and 0.4 pI: of Bal3l treated blunt end

DNA of plasmid pJDB207/1F2(5,) or pJDB207/1IF2 (1b) are ligated ‘over night at room temperature in 25 pl of 60 mM Tris pH 7.5, 10 mM MgCl, 5 mM DIT, 3.5 mM ATP and 450 U of T4 DNA ligase. The ligated DNA is separated from excess linkers by isopropanol precipitation in the presence of 10 mM EDTA, 0.3 M sodiumacetate pH 6.0 and 0.54 volumes of isopropanol. After 35 min. of incubation at room temperature the DNA is sedimented by centrifugation. The pellets are dried at the air and resuspended in 17 pl of 6 mM Tris pH 7.9, 150 mM NaCl, 6 mM MgCl, and 6 mM mercaptoethanol. The Xhol linkers ligated to the DNA are cleaved with XhoI, the DNA is precipitated with isopropanol as described before and circularized by ligation. After 6 hours of ligation at 15°C in 50 pl of 60 mM Tris pH 7,5, 10 mM MgCl, , 10 mM DTT, 1 mM ATP and vo (BAD ORIGINAL 9

600 U of T& DNA ligase 10 pl of each ligation mixture are added to 100 pl of calcium-i reated, transformation competent E. ccli HBLO1 cells (see Example 4a). 72 transformed, amp colonies containing plasmids with an IFN-5, insert are grown individually in LB medium containing 100 mg/l ampicillin.

Plasmid DNA is analysed by HaellI digestion. The restriction pattern allows to judge the approximate size of the deletion introduced by

Balll. Two clones are further analysed and assayed for interferon activity. They are referred to as pJDB207/1F2(5,)A72 and pJDB207/1F2 (5,)082.

The nucleotide sequence on either side of the new junction (Xhol linker) between the 3' nontranslated region of the IFN-5, gene and the PHOS transcription termination region is given in fig. 28.

In an analogous manner 60 amp colonies containing plasmids with an

IFN-1'b insert are grown individually in LB medium containing 100 mg/l ampicillin. Plasmid DNA is analysed as described above. One clone is selected and assayed for IFN activity. It is referred to as pJDB207/1F2(1'b)A.

Example 23: Construction of recombinant plasmids, qottaining pattable

IFN-5., 8, and -1'b cDNA inserts which can be used for direct : expression of mature lymphoblastoid IFN (cf. figures 29 and 30) a) Preparation of the cDNA inserts ~The cDNA inserts are excised from the recombinant plasmids

CG-pBR322/HLycIFN-8,, CG-pBR322/HLycIFN-1'b, CG-pBR322/HLycIFN-5, by digestion of each of 150 pg of plasmid DNA with PstI (Biolabs) using the procedure as suggested by the supplier. Following phenol extraction and ethanol precipitation, the excised inserts are isolated by means of sucrose density gradient centrifugation (5-237) in 50 mM Tris-HCl (pH 8.0) and 1 mM EDTA. The centrifugation qo is performed at 35 000 rpm in a TST 41 rotor (Kontron AG) at 15°C

Cee BAD ORIGINAL

- . ‘ 26266 for 16 h. 0.3 ml fractions are collected with an ISCO gradient : collector at 1 ml/min. The fractions containing the small fragment (i.e. the insert) are pooled. The DNA is precipitated with ethanol as usual, and the precipitate is collected by centrifugation in a

HB-4 rotor (Sorvall) at 10 000 rpm at 0°C for 10 min. The precipitate is redissolved in 60 pl 10 mM Tris-HCL (pH 7.5) and 0.05 mM EDTA.

About 30 ug DNA are recovered as determined by measuring the optical density. 2 pg of each of the cDNA inserts are digested with 2.5 units of Sau3A (Biolabs) in 10 pg/ml EtBr and incubated at 37°C for 60 min. The digests are phenol extracted and’ the DNA is precipitated in ethanol as above. The DNA fragments are fractionated on a 1.2 Z agarose gel in a solution containing 50 mM Tris, 50 mM boric acid, 1 mM EDTA and . 0.5 pg/ml ethidium bromide.

The second largest DNA fragment (Sau 3A-Pstl) from each of the 3 digests is excised from the gel, squirted through a thin needle with a syringe into 5 ml of 0.15 M NaCl, 50 mM Tris<HCl (pH 8.07%, 1 mM EDTA, and eluted overnight by shaking. The eluate is passed . through a 100 pl DE~-52 (Whatman) Pasteur-pipette column to adsorb the DNA. The column is washed with 2 ml of the same buffer and the DNA is eluted with 400 ul of a solution containing 1.3 M NaCl, 50 mM Tris IR (pH 8.0) and 1 mM EDTA. The DNA is precipitated with 2 volumes Bh of ethanol at -20°C overnight. The precipitates are collected by centrifugation in an Eppendorf centrifuge and redissolved in 10 pl 10 mM TriseHCl (pH 7.8), 0,05 mM EDTA. b) Preparation of the acceptor plasmid DNA-fragment 1. Digestion of the plasmid pBR322 by EcoRI and 5, 10 pg of plasmid DNA pBR322 are digested with 15 units of EcoRI (Biolabs) for 60 min. at 37°C under conditions described by the supplier. Following phenol extraction and ethanol precipitation, the DNA is redissolved in 25 pl H,0 and the staggered ends are aso ORIGINAL 9 ] ee ean ment aR A EE 3 OTHE Ly OS ARTE NL RR EIRSR P a Cee removed Ly treatment of the DNA with 20 units of endonuclease 51 (P-L Biochemicals) in 350 pl of a solution containing 250 mM NaCl, . 30 mM sodium acetate (pH 4.5), 1 mM ZnsS0, at 30°C for 30 min. The reaction is stopped by adding EDTA (pH 7.5) at 10 mM. The DNA is extracted with phenol, concentrated by ethanol precipitation and redissolved in 50 pl of 10 mM TriseiiCl (pH 7.8), 0.05 mM EDTA. 2. Ligation of a chemically synthesized DNA linker to pBR322 digested with EcoRI and Sy

Two oligodeoxynucleotides of the formulae 5'-AATTCTATGIGT-3' and 5'—GATCACACATAGAATT-3' are synthesized using the procedure described in Example 10 Ea.

The synthetic oligodeoxynucleotides are phosphorylated at their 5'-ends by incubating 80 pmoles of both oligodeoxynucleotides with uCi [y-22p)-ATP (6700 Ciemmol }, Amersham) in a 80 pl reaction volume, containing 0.1 mM rATP (Sigma), 50 mM Tris+HC1l (pH 9.5), 10 oM MgCl, , 5 mM DTT and 20 units of T, polynucleotide Kinase . (P-L Biochemicals) for 30 min. at 37°C. The reaction is stopped by freezing the mixture at -80°C. : 20 The resulting radioactively phosphorylated linker of the formula (3%p)-aarrcratcrer nC Co

TTAAGATACACACTAG-[ "P] : . : is subsequently incubated with 6 pg of pBR322 cleaved with EcoRI and 5, (see above) in a 200 pl reaction volume containing 0.5 mM rATP (Sigma) 0.1 mM DIT (Calbiochem), 20 mM TriseHCl (pH 7.8), 10 mM MgCl, and 4e10° units of T, DNA ligase (Biolabs) by incubating the mixture for 2 h at 15°C. ,

The solution is deproteinized by phenol extraction and the DNA is concentrated by ethanol precipitation. The DNA is redissolved in 100 pl 10 mM TriseHCl (pH 7.5), 0.05 mM EDTA and centrifuged through a sucrose gradient (5 - 23 Z) in 50 mM Tris+HCl (pH 8.0),

BAD ORIGINA- }

Ce ae ee ia Aen EU RE

: ~ 85 - . 26206 ! ’ 4 1 mM IDTa. The centrifugation is performed at 60 000 rpm in a TST ) 60 rotor (Kontron AG) at 15°C for 2.5 h. 0.2 ml fractions are ; collected with an ISCO gradient collector at 1 ml/min. The fractions i containing the (32pj-1abeled plasmid DNA (fractions 11 - 14 out ; of 22 fractions) are pooled, the DNA is precipitated with ethanol and digested with 12 units of Bcll (Biolabs) as recommended by the supplier. After phenol extraction and ethanol precipitation, the digested DNA is treated with 10 units of PstI (Biolabs) as recommended by the supplier. The phenol extracted digest is then centrifuged through a (5 = 23 %) sucrose density gradient for 15 h at 35 000 rpm at 15°C in a TST 41 rotor (Kontron AG). 0.3 ml fractions are collected (see above) and the fractions containing the large [32p)-1abeled Bell - PstI DNA fragment (fractions 27 - 31 out of 42 fractions) are pooled and concentrated by ethanol precipitation.

The DNA is redissolved in 20 pl TriseHCl pH 7.8, 0.05 mM EDTA. c. Ligation of the acceptor plasmid fragment to the cDNA inserts 2 pl of the acceptor plasmid DNA fragment (~100 ng) (see above) are incubated with each of 5 Pl of cDNA inserts (~~20 ng) (see above) in a reaction volume containing 20 mM TriseHCl (pH 7.8), 10 mM MgCl,, 0.1 mM rATP, 0.1 mM DTT and 400 units of T, DNA ligase in 10 pl for 3 h at 15°C. : . PE A 5 pl of the reaction mixtures are then added to a mixture containing 150 pl calcium-treated E. coli HB 101 in 10 mM MgCl, , 10 mM CaCl, and 10 mM TriseHCl (pH 7.5) in a total volume of 200 pl. : 25 The mixture is cooled in ice for 20 min. heated to 42°C for 1 min. and incubated at 20°C for 10 min. 1 ml of tryptone medium (tryptone medium contains 10 g Bacto-Trypton (Difco); 1 g yeast extract (Difco); 1 g glucose; 8 g NaCl and 294 mg CaCl, +2 H,0 in 1 1 of distilled water) is added and the mixture is incubated for 30 min. at 37°C by shaking at 300 rpm. The mixture is plated ontc 2 agar plates (Mc Conkey agar, Difco; 0,6 ml/plate) supplemented with ' wo ORIGINAL 9

10 pg ml or tetracycline (Sigma). The plates are incubated at 37°C for 12-17 hn, Approximately 1000 colonies are obtained per transfor- mation mixture. 4 colonies are picked from each transformation mixture for further analysis. d) Restriction analysis of the hybrid plasmids

In order to further analyse the potential hybrid Plasmids, the plasmid DNA ig isolated From 12 colonies (4 from each of the 3 ligation mixtures, sce above),

The hybrid plasmid DNA is isolated as follows: |] colony is used to inoculate 10 ml of tryptone medium, supplemented with 10 pg ml of tetracycline ag above in a 25 pi Erlenmeyer flask. The culture ig shaken for 15 - 18 h at 37°C at 300 rpm. The cells are harvested by centrifugation (Sorvall, Hs-4 rotor, 10 min. at 4000 rpm, 4°C),

About 0.1 8 of cells are obtained and are resuspended in 1 ml 50 mM

TriseHC1 (pH 8.0). 0.25 ml of lysozyme solution (10 mg/ml in 50 mM

Tris-HCl (pH 8.0), lysozyme is puchased from Sigma) are added and after incubation at 0°C for 10 min., 0.15 pi of 0.5 M EDTA (pH 7.5) is added. After another 10 min, at 0°C, 60 Pl of 2 7 Triton X-100 . (Merck) is added. After 30 min. at 0°C, the sample ig centrifuged for 30 min. at 15000 rpm and 4°C in a Sorvall SA-600 rotor. The super- natant is deproteinized with } volume of. Phenol (saturated. inTNE).

The phases are separated by centrifugation (Sorvall up-j4 rotor) for _ 10 min. at 5000 rpm at 4°C. The upper phase ig extracted twice with 1 volume of chloroform. Pancreatic RNAse A (Sigma; 10 mg/ml in TNE, preheated 10 min, at 85°C) is added to a final concentration of 25 pg/ml and the mixture jig incubated for 40 min. at 37°C. The solution is then adjusted to 1 M NaCl and 10 2 Polyethylene glycol 6000 (Fluka, autoclaved for 20 min, at 120°C) and incubated at -10°C for 2 nh, The precipitate is collected in a Sorvall HB~4 rotor (20 min. at 10 000 rpm, 0°C) and redissolved in 100 Ml of TNE. The

DNA solution ig ¢xtracted with 1 volume of phenol and the DNA is

Precipitated with 2 volumes of ethanol at -80°C for 10 min.

NE.

Lao ORG" a

10 pg/ml of tetracycline (Sigma). The plates are incubated at 3’°c for 12-17 h. Approximately 1000 colonies are obtained per transfor- mation mixture. 4 colonies are picked from cach transformation mixture for further analysis. d) Restriction analysis of the hybrid plasmids

In order to further analyse the potential hybrid plasmids, the plasmid DNA is isolated from 12 colonies (4 from each of the 3 ligation mixtures, sce above).

The hybrid plasmid DNA is isolated as follows: 1 colony is used to inoculate 10 ml of tryptone medium, supplemented with 10 pg ml of tetracycline as above in a 25 ml Erlenmeyer flask. The culture is shaken for 15 - 18 h at 37°C at 300 rpm. The cells are harvested by centrifugation (Sorvall, MS-4 rotor, 10 min. at 4000 rpm, 4°C).

About 0.1 g of cells are obtained and are resuspended in 1 ml 50 mM

Trise<HCl (pH 8.0). 0.25 ml of lysozyme solution (10 mg/ml in 50 mM

Tris~HC1 (pH 8.0), lysozyme is puchased from Sigma) are added and after incubation at 0°C for 10 min., 0.15 ml of 0.5 M EDTA (pH 7.5) is added. After another 10 min. at 0°C, 60 pl of 2 7 Triton X-100. (Merck) is added. After 30 min. at 0°C, the sample is centrifuged for 30 min. at 15000 rpm and 4°C in a Sorvall SA-600 rotor. The super-— natant is deproteinized with 1 volume of phenol (saturated. in: TNE). ‘

The phases are separated by centrifugation (Sorvall HB-4 rotor) for So 10 min. at 5000 rpm at 4°C. The upper phase is extracted twice with

I volume of chloroform. Pancreatic RNAse A (Sigma; 10 mg/ml in TNE, preheated 10 min. at 85°C) is added to a final concentration of . 25 pg/ml and the mixture is incubated for 40 min. at 37°C. The solution is then adjusted to 1 M NaCl and 10 Z polyethylene glycol 6000 (Fluka, autoclaved for 20 min. at 120°C) and incubated at =10°C for 2 h. The precipitate is collected in a Sorvall HB-4 rotor (20 min. at 10 000 rpm, 0°C) and redissolved in 100 pl of TNE. The

DNA solution is extracted with 1 volume of phenol and the DNA ig precipitated with 2 volumes of ethanol at -80°C for 10 min. . Co Ce {| > oo laa ORIGINAL 9

26266 1

The precipitate is collected by centrifugdtion in an Eppendorf centrifuge and the DNA is redissolved in 20 pl of 10 nM Tris-HCl (pH 7.5) and 0.5 mM EDTA. 8 —- 10 pg of hybrid plasmid DNA are recovered from 10 ml culture.

All of the plasmid DNAs are analysed by the following double digests: 0.5 pg of each DNA is digested with HindIII (Biolabs) and Pvull (Biolabs), UindIIl and PstI (Biolabs), HindIlIl and BamH, (Biolabs),

EcoRI (Biolabs) and Pstl using standard protocols, and fractionated on a 1.5 % agarose gel according to size in 40 nM Tris-acetate (pH 7.8), 1 mM EDTA containing 0.5 pg/ml EtBr.

The hybrid plasmids having the desired restriction enzyme pattern are selected. The result is summarized in figures 29 and 30.

Plasmid DNA, containing the insert derived from pBR322/MLycIFN-8) or pBR3I22/HLycIFN-5, or pBR322/HLycIFN-1'b are denoted

CG-pBR3I22HLycl¥N(a-3)-252 and CC-pBR3I22/HLyciFN(a-2)-261 and

CG-pBR3I22/1LycIFN(a-1)-258, respectively. ]

In order to further confirm the structure at the junction point between the linker and the start of the coding sequence of the IFN cDNAs, the nucleotide sequence is determined at this area. In particular, 2 pg of the isolated plasmid DNA CG-pBR322/HLy cTFN(a-1)-258 : is digested with EcoRI, 5'-terminally labeled and cleaved with Pstl: -

The DNA fragments are fractionated on a 6 Z polyacrylamid gel and the

EcoRI*-Pstl (9.4 bp) DNA [ragment is extracted as described above.

The DNA fragment is subjected to sequence analysis according to Maxam and Gilbert (15).

The structure at the junction point between the linker and the start of the coding sequence of the IFN cDNAs in plasmids

CG-pBR322 HLycIFN(a-2)-261 and CG-pBR322/HLycIFN(a-3)-252 is confirmed analogously.

Joao omic: A !

y

In the plasmids CC-pBR322/HLyclFN(a-1)-258, (a-2)-261 and (a-3)--252 the IFN coding sequences are preceded by the following nulceotide segment containing an EcoRI restriction site.

EcoRI SaulA ~NGAATTCTATGTGTGATC. .... ~NCTTAACATACACACTAG. .... >

IFN gene

Example 24: Deletion of the PHOS signal sequence in the expression plasmid p3l (see figure 31)

Expression plasmid p3l contains the PHO5 promoter sequence including the mRNA start sites, the translation start codon ATG of acid phos- phatase and additional 40 nucleotides coding for part of the acid phosphatase sigual sequence. In this construction, the nucleotides for the signal sequence and the ATG are eliminated by Bal3l digestion.

EcoRl linkers are introduced to allow joining of the PHO5 promoter to appropriate mature coding sequences (e.g. interferon genes). a) Bal3l digestion of Ball cleaved plasmid p30 pg of p30 DNA (see Example 4b) are digested with restriction epdo- nuclease Ball, resulting in 2 fragments of 3.7 and 5.1 kb. After : 70 extraction with phenol/thloroform, the DNA is precipitated with : ethanol. The DNA is resuspended in 10 mM Tris pH 8.0 at a concentration of 0.5 pg/ml. 9 pg of Ball cleaved p30 DNA are digested with 2 U of exonuclease Bal3l (BRL ) in 100 pl of 20 mM Tris pH 8.0, 100 mM

NaCl, 12 mM MgCl,, 12 wd CaCl, and 1 mM EDTA. Aliquots of 2 yig DNA each are withdrawn after 15 sec., 30 sec., 45 sec. and 60 sec. of incubation at 30°C and are immediately mixed with 50 pl phenol and 60 pl TNE. After extraction with phenol/chloroform and ethanol preci- pitation, the DNA is resuspended in 10 mM Tris pH 8.0 at a concentra- leap ORIGINAL 9 tion of 100 pg/ml. To analyse the extent of exonucleolytic cleavage by Bal3l 0.5 jig of DNA from each time point are digested with endo- nuclease Bamlll and analysed on a 1.5 Z agarose gel in Tris-borate buffer pH 8.3. On the average 70 bp are removed from each end of the fragment after 45 sec. of Bal3l digestion. For further experiments

DNA from the 45 second time point is used. b) Addition of EcoRl linkers to the Bal3l treated DNA

Two As 60 units of EcoRI linkers (5'-GGAATICC-3', BRL) are resuspended in 250 pl of 10 mM Tris pH 8, 1 wM EDTA. Two pg of EcoRI linkers are kinased in 75 pl of 60 mM Tris pl 7.5, 10 mM MgCl, , 15 mM DDT, 10 i

ATP and 33 U of T4 polynucleotide kinase (Boehringer). After 1 h at 37°C the mixture is allowed to cool to room temperature and is then stored at -20°C.

The anncaled, double stranded EcoRI linkers are ligated with their blunt ends to the Bal3l treated DNA fragments. Half a microgram of .

Bal3l treated DNA (see Example 24a) is incubated for 16 hours at room temperature with a SOfold excess of kinased EcoRI linkers in 20 ni of 60 wM Tris pH 7.5, 10 mM MgCl,, 10 mM DIT, 4 wM ATP and 600 U of T4

DNA ligase (Biolabs). After inactivation of the T4 DNA ligase (10 min at 65°C) the excess of EcoRl linkers is cleaved by 50 U of EcoRI (Boehringer) in a volume of 50 il. The, DNA is extracted with phenol/ chloroform, precipitated by ethanol and resuspended ‘in 10 mM Tris, EE 1 mM EDTA. ’

Restriction endonuclease EcoRI not only cleaves the terminally added

EcoRI linkers of both Ball fragments (3.7 kb and 5.1 kb) but also at an internal EcoRI site in the 5.1 kb fragment giving rise to a 3.9 kb and a 1.2 kb fragment. The 3.7 kb and 3.9 kb fragments are separated from the 1.2 kb fragment on a 0.8 7 low melting agarose gel (Sigma) in 90 mM Tris-HCl pH 8.3, 90 mM boric acid and 2.5 mM EDTA. The DNA bands are stained with ethidium bromide and visualized under long wave

Uv light at 366 nm. The two large DNA fragments of 3.7 kb aud 3.9 kb en

BAD ORIGINAL Yl are not separated. They are cut out of the gel 2 0.4 gel block and are extracted as described in Example 4a.

The linear fragments terminating in EcoRI sticky ends are circularized by ligation. About 0.25 pg of fragments are ligated in 100 pl of 60 mM

Tris pH 7.5, 10 mM MgCl, , 10 mM DIT, 1 mM ATP and 600 U T4 DNA ligase for 4 hours at 15°C.

Ten pl aliquots of the ligation mixture are added to 100 ul of calcium : treated, transformation competent E.coli HB101 cells (see Example 4a). 35 transformed, amp" colonies are grown individually in LB medium ’ containing 100 pg/ml of ampicillin. Plasmid DNA is prepared according to the method of Holmes et.al. (50) and is analysed by EcoRI BamHI double digestion. c) Nucleotide sequence analysis to determine the position of the

EcoRI linker addition

Most of the 35 clones will differ from each other in the position of

EcoRI linker addition in the PHOS promoter region depending on the ] : degree of Bal3l digestion of the individual DNA molecules. For the nucleotide sequence analysis plasmid DNA is digested with EcoRI. After extraction with phenol chloroform the restricted DNA is precipitated

With ethanol. The DNA is dephosphorylated and 5'-terminally labeled as oC described in Example 10Ed. The labeled DNA fragments are cleaved with a _ second restriction endonuclease, BamHI. The products are separated on a 0.8 Z low melting agarose gel. The 0.5-0.6 kb 5'-labeled EcoRI-BamHI «fragment is isolated from low melting agarose as described in Example 4a. For the determination of the nucleotide sequence adjacent to the

EcoRI linker the different DNA fragments are chemically degraded and the products are separated by polyacrylamide gel electrophoresis as described by Maxam and Gilbert (15). _ ORIGINAL 9 ia

The different clones and the position of the corresponding last nucleo- tide of the PHO5 sequence (then followed by an EcoRI linker) is listed in

Tab. 4 (see also figure32).

Table 4 . clone position of, last nucleotide of the PHOS sequence pE +25 =

PG +16 pe +15 pd +12 pY -4 pR -10 pP -16 pV -18 pL -21 pN C22 pC 24 pH -27 : pS -28 pk : -29 pl Co -38 ne pM -50 oo - RE po -53 - pF - om -59 - pm -67 pk -73 pi -81 ph -137 d) Isolation of a 0.53 kb BamHI-EcoRl fragment containing the

PHOS/R promoter:

Plasmid pR contains the PHO5/R promoter on a 534 bp BamHI-EcoRIl fragment. According to the numbering in fig. 3a, the fragment covers

PHOS promoter sequences from nucleotide - 541 (BamllI site) to nucleotide —- 10. An EcoRI linker, ligated to nucleotide - 10 (see

Example 24b) contributes two G-residues upon EcoRI cleavage.

Plasmid pR is digested with restriction endonucleases BamHI and

EcoRL. The 0.53 kb BamlI-EcoRI fragment is separated on a 0.8% low melting agarose gel and isolated as described in Example 4a. The nucleotide sequence is given in fig. 33. . i

In an analogous manner plasmid pY is digested and a 0.53 kb

BamllI-EcoRl fragment is isolated containing the PHOS5/Y promoter.

The nucleotide sequence is given in fig. 34. ’ ; 15 e) Replacement of the Sall-EcoRL fragment in plasmid p31 by a - Sall-EcoRI fragment of the new constructions =. a

Five mg of plasmid p3l (cf. Example 13d) are digested with restriction .~ : endonuclease SalI. The restricted DNA is precipitated with ethanol and resuspended in 50 ul of 100 mM Tris pH 7.5, 50 mM NaCl, 5 mM MgCl, - «20 The DNA is digested with EcoRl’ to completion. The restriction fragments are separated on a 0.87 low melting agarose gel in Tris-borate-EDTA buffer pH 8.3. A 3.5 kb DNA fragment is isolated in a small gel block containing the DNA band.

- 03 -

Five pg each of clones pR and pY (cf. Table 4 and fig. 32) are digested with Sall and EcoRI in the same way :s described above. The 0.8 kb DNA fragments are isolated in small blocks of low melting agarose gel. 0.67 pg of the 3.5 kb Sall-EcoRl fragment of vector p3l is ligated to . . 5 0.34 pg of the 0.8 kb SalI-EcoRI fragment of plasmid pR or pY, respectively. Appropriate gel blocks, containing the DNA fragments are mixed and melted at 65°C. The liquified gel is diluted three times.

Ligation is performed in a total volume of 240 pl of 60 mM Tris pH 7.5, 10 mM MgCl,, 10 mM DIT, 1 mM ATP with 750 U of T4 DNA ligase (Biolabs) 10 overnight at 15°C. A 2 pl aliquot of each of the ligations is added to 100 pt of calcium treated, transformation competent E. coli HB1O1l cells (see Example 4a). "8 transformed, amp” colonies each are grown individually in LB medium containing 100 pg/ml ampicillin. Plasmid DNA is analysed by restriction 15 analysis. The clones of each group are identical. One clone each is further used and referred to as p3L/R or p3l/Y, respectively (Figs 31).

Example 25: Insertion of lymphoblastoid interferonw-3 DNA into : plasmid p3L/R or p3L¥ (cf. figure 35)

This construction joins the PHOS promoter region to the gene coding 20 for mature interferon-a-3. Neither of the signal sequences of PHOS" - or interferon is present but there is an EcoRI linker introduced at the site of the junction. ~~ n i zil \aa ORIGINAL SH re"

- ’ . J to. . i - 94 - . oo 26266

Plasmid p3l/R (sce Example 24e) contains the P05 promoter sequence which is terminated 9 nucleotides before the AIG of the acid phosphatase gene by an rcokf linker 57 —GUAAYILL=3". Lhe iymphoblastoid interferoma-3 gene in plasmid CG-pBR322/HLycIFN (a=3)-252 (see

Example 23) is specifically adapted for the junction to the EcoRI linker in the PHO5 promoter. The additional nucleotides at the 5' end of the coding sequence of mature interferon-a-3 provide an EcoRI restriction site and an ATG, necessary for the translation of the inter- feron gene (cf. figure 29). Essentially the same construction is also done with plasmid p3LY (see Example 24e). a) Preparation of EcoRI cleaved, dephosphorylated plasmid pIL/R

Five pg of plasmid p31 R are digested with restriction endonuclease

EcoRI (Boehringer) to completion. After extraction with phenol/chloro- form, the DNA is precipitated with ethanol and resuspended in 100 pl of 50 mM Tris pH 8.0. Passage of the DNA through Chelex 100 (BioRAD), dephosphorylation by calf intestine alkaline phosphatase (Boehringer) and purification of the dephosphorylated DNA by DE52 ion exchangt chromatography is as described in Example 14b. The DNA is resuspended : in 10 mM Tris-HCl pH 7.5, 1 mM EDTA at a concentration of 0.2 mg/ml. b) Isolation of a 0.6 kb EcoRl fragment of plasmid

CG-pBR322/HLycIFN(a-3)-252 containing the IFN-a-3 coding sequence

Ten pg of plasmid CG-pBRI22/HLycIFN(a—3)-252 are digested with restric- tion endonuclease EcoRI. The dfpest results in 2 fragments of 3.8 kb . l8AD ORIGINAL

Fro . Cee

\ > 6266 and 0.6 kb. The 0.6 kb fragment contains the we 2 coding region. The fragment is isolated on a 0.6 7 low melting agarose gel in

Tris-borate~EDTA buffer. The gel piece containing the 0.6 kb DNA fragment is cut out of the gel and used for ligation. ¢) Ligation of linearized, dephos horylated p3L'R DNA and the 0.6 kb nny Sm p0Ty ated pIL/R DNA and the 0.6 kb : EcoRI fragment of IFN-a-3 DNA 1.5 J of dephosphorylated p3L/R vector DNA cleaved with EcoRI is ligated to 0.19 pg of the 0.6 kb EcoRI fragment of IFN-a-3. The latter fragment is contained in a small block of low melting agarose gel whiéh is melted at 65°C. The liquidified gel is diluted two times.

Ligation is performed in a total volume of 220 pl 60 mM Tris pH 7.5, 10 mM MgCl, , 10 mM DTT, 1 mM ATP with 800 U of T4 DNA ligase (Biolabs) overnight at 15°C. A 10 pl aliques of the ligation mixture is added to 100 yl of calcium treated, transformation competent E. coli HB10l cells (see Example 4a). 6 transformed, —— colonies are grown individually in LB medium, containing 100 Me ml ampicillin. Plasmid DNA is prepared according to the method of Holmes et al. (50) and is analysed by BglII /BstEII double digests to determine the orientation and the size of the insert.

One of these clones is referred to as p31R/IF(a-3). .

The same construction is done with paLY (See Example 24e). Plasmids from 6 transformed, amp" colonies are analysed. 2 clones have the right orientation of the insert. One.of.them is referred to as p31Y/IF(a-3).

Example 26: Insertion of lymphoblastoid interferon-t-2 DMA into plasmid

P3LR (see figure 36)

This construction joins the PHOS promoter to the mature interferon-a-2 coding region. a) Isolation of a 3.9 kb HindIII-EcoRI fragment of vector p3LR

Ten pg of vector p3L/R are digested with HindIII to completion. The buffer is ddjusted with 0.1 volume of 1 M Tris pH 7.5. The HindJI1-

TR v ono ORIGINAL 9 cleaved p3L/R DNA is then digested with EcoRI. The 3.9 kl HindlII-

EcoRI fragment is isolated from a 0.8 I low melting agarose gel in a gel block cut out of the gel, b) Isolation of a 0.9 kb Xbal -HindIII fragment of pJDB207/1F2(5,)

Five ug of plasmid pJDB207/1F2(5 ) (cf. Example 17) are digested with

HindIII to completion. The buffer is adjusted with 0.1 volume of 1 M Tris pH 7.9. The HindIII-cleaved plasmid is then digested with

XbaI, The 0.9 kb Xbal-HindIII fragment contains part of the interferon- a-2 coding sequence and the downstream PHO5 transcription termination signals. The 0.9 kb fragment is separated on a 0.87 low melting agarose gel and cut out of the gel. c¢) Isolation of a 252 bp EcoRI-Xbal fragment of plasmid

CG-pBR322/HLycIFN(a=2)-261 containing part of the IFN-a-2 coding sequence

Ten pg of plasmid CG-pBR322/HLycIFN(a-2)-261 (see Example 23) are . digested with Xbal in 100 pl of 6 mM Tris pH 7.9, 150 mM NaCl, 6 -mM

MgCl, and 6 mM mercaptoethanol. After complete digestion with Xbal the linearized plasmid DNA is partially digested with 3 U of EcoRI (Boehringer). After 20 min at 37°C the digestion is stopped by freezing at -70°C. The DNA fragments are analysed on a 1.5 % agarose gel in

Tris-borate-EDTA buffer pH 8.3. The 252 bp EcoRI-Xbal fragment contains the 5' part of the mature interferon-a-2 coding sequence (up to the

Xbal site) with the specific linker for the junction with the PHOS promoter. The 252 bp fragment 1¥ 185lated in a small gel block from a 0.8 Z low melting agarose gel, d) Ligation of DNA fragments

Three DNA fragments described in Examples 26 a-c, having appropriate sticky ends are ligated in one reaction: 0.67 pg of the 3.9 kb HindIII-EcoR{ fragment of vector p3L/R, 0.16 pI: of the 0.9 kb XbaI-HindIII fragment of pJDB207/1F2(5,) and about 70 ng ano ORIGINAL 9

- 97 - ° 26266 of the 250 bp EcoRI-Xbal fragment of CG~pBR3I22HLyclFN(a-2)-261 are ligated. All three DNA fragments acre contained in small gel blocks of low melting agarose. The three pieces ot agarose gel are pooled, melted at 65°C and diluted three times. The ligation is done in a total volume of 450 pl of 60 mM Tris pil 7.5, 10 mM MgCl,, 10 mM DIT, 1 mM ATP with 1200 U of T4 DNA ligase at 15°C for 16 hours. A 10 pl aliquot of the ligation mixture is added to 100 pl of calcium treated, transformation competent E. coli HB1O1l cells (see Example 4a). : 12 transformed, amp™ colonies are grown individually in LB medium containing 100 pg/ml of ampicillin. Plasmid DNA is prepared according to the method of Holmes et al. (50) and is analysed by BamHI HindIIL . double digestion. All clones show. an identical digestion pattern.

One of them is referred to as p31R/IF(a-2),

Instead of the 0.9 kb Xbal-HindIII fragment of pJDB207/1F2(5,) also a 0.5 kb Xbal-HindIIl fragment of plasmid pJDB207/1F2(5,)472 or pJDB207/1F2(5 )AB2 (see example 22) can be used for ligation. Alsp, instead of the 3.9 kb HindIII-EcoRI fragment of vector p31/R the ligation is carried out with the 3:9 kb HindIII-EcoRI fragment of vector p3lY.

Conditions for ligation of the DNA fragments and the transformation of E.coli HBIOl are the same as described above. 12 transformed, amp colonies “0} @ich ligation are grown individually in LB medium containing 100 pg/ml ampicillin. Plasmid DNA is analysed by BamHI HindIII double digestion. The resulting clones are referred to as p3lR/IF(a-2)A72, p31R/1F (a2) 282, p31Y/IF (a2), p31Y/IF(a-2)A72 and p31Y/IF(a-2)A82.

BAD ORIGINAL 9

- y8 - \ 26266

Example 27: Insertion of lymphoblastoid AN into plasmid p3L/R (see tigure 37) TT a) Isolation of a 3.9 kb HindLII-EcoRI fragment of vector p3l/R:

Ten pg of vector p31l/R are digested with HindIII and EcoRI as de- scribed in Example 26a. The resulting 0.4 kb and 3.9 kb fragments are separated on a preparative 0.8% low melting agarose gel. The 3.9kb

Hind I[11-EcoRLl fragment is eluted from the gel as described in Example 4a. The DNA is purified by DE52 (Whatman) ion exchange chromatography (see Example 5a), precipitated with ethanol and resuspended in 10 mM

Tris pit 8.0, 1 mM EDTA at a concentration of 0.1 mg/ml. b) lsolation of 0.9 kb Pvull-HindI1I fragment of pJDB207/1F2(1'b) :

Five mg of plasmid pJDB207./1F2(1'b) (cf. Example 17) are digested with

Pvull and MindI1I. The resulting fragments of 0.9 kb and 7.3 kb are separated on a preparative 0.87 low melting agarose gel. The 0.9 kb fragment is eluted from the gel and purified as described in Example 27a. The DNA is resuspended in 10 mM Tris pH 8.0, 1 mM EDTA at a concentration of 0.05 mg/ml. . ¢) Isolation of a 286 bp EcoRI-Pvull fragment of plasmid

CG-pBR322/HLycIFN(a-1)-258 containing part of the IFN-a-1 coding sequence

Ten pg of plasmid CG-pBR322/ULyclFN(a-1)-258 (see Example 23) are digested with Pvull and EcoRI. A 286 bp restriction fragment is separated from a 4.2 kb fragment on a preparative 0.87 low melting aparose gel. The 286 bp EcoRL-Pvull fragment is eluted from the gel and purified as described in Example 27a. The DNA is resuspended in 10 mM Tris pH 8.0, 1 ntl EDTA at a concentration of 0.03 mg/ml. d) Ligation of DNA fragments 0.5 pg of the 3.9 kb HindI[1I-EcoRI fragment of vector p31/R, 0.25 pe of the 0.9 kb PvulI-HindIII fragment of pJDB207/1F2(1'b) and 0.1 Je of the 286 bp EcoRl-Pvull fragment of plasmid .

CG-pBR322/HLycIFN(a-1)-258 are ligated for 16 hrs at 15°C in 20 pl

BAD ORIGINAL 9

~~ 26266 wr’ of 60 mM Tris pH 7.5, 10 mn MpCl,, 10 mM DTT, 4 mM ATP and 600 U of

DNA ligase. A 3 pl aliquot of the ligation mixture is added te 100 pl of calcium treated, transformation competent E. coli NB1O1 cells (sce Example 4a). 12 transformed, amp’ colonies are grown individually in LB medium with 100 pg/ml of ampicillin. Plasmid DNA is analysed by BamHl Hind1lII double digestion. One clone giving rise to a 1.4 kb Bamlll fragment and a 390 bp BamllI-Nind11l fragment, is further analysed and referred to as pJIR/IF(a~1).

The 3.9 kb HindllI-EcoRI fragment of vector p3l/R can be replaced by the HindII1-EcoRI fragment of vector p3l/Y. Also, instead of the 0.9 kb Pvull-HindI1I fragment of pIDB207/1F2(1'b) , a 0.45 kb Pvull-

HindIII fragment of pJDB207/IF2(1'b)A can be used for the ligation.

The resulting clones are analysed as described above. The clones are referred to as p31R/1IF(a-1)a, p31Y/IF(a-1) and p31Y¥/IF(a-1)A.

Example 28: Subcloning of gene constructions in the high copy number yeast vector pJDB207 (see fig. 38) :

The constructions described in Example 25-27 contain the PHOS promoter, different interferon coding regions and the PHOS transcription termi- nation signals in a tandem array, all inserted in a pBR322 derived vector. Sall-HindIII fragments containing the whole array are ligated into the 6.2 kb SalI-HindIII fragment of pJDB207 as described in

Example 17. 2 pg of plasmid p31R/IF (@=3) are digested with restriction endonucleases Sall and HindIII. The restriction fragments are sepa- rated on a preparative 0.8% low melting agarose gel. The small frag- ment (1.8 kb in size) is cut out of the gel. ’

Plasmid pJUB207 is digested with restriction endonucleases Sall and

HindlIl and the large 6.2 kb fragment is isolated as described in

Example 17. \ean ORIGINAL >

Ligation of the DNA fragments and transformation of competent E. coli

HB101 cells is carried out as described in Example 17. 8 amp colonies are grown individually in LB medium containing 100 jig/ml of ampicillin.

The plasmid DNA is analysed for the size of the insert by cleavages with restriction endonucleases HindIII and Sall. One clone having the correct insert is selected and referred to as pJDB207R/IF(a-3).

In an analogous manner, starting from the plasmids p31Y/IF (a-3), pIIR/LF (2), p31R/1F(a-2)A72, p3IRVIF@-2)882, p31Y/1F (0-2), p31Y/IF(a-2)A72, p31Y/1F(a-2)082, p31R/1F(a-1), p3IR/IF(er1) 8, p31Y/LF(a-1) and p31Y/1F(a-1)4, the following clones with the correct insert are obtained: pJDB207Y/IF («-3), pJDB207R/IF (a-2), pJDB207R/IF(a-2)872, pJDB207R/1F(a-2)082, pIDB207Y/IF («=2), pJDB207Y/IF(a=2)A72, : . pJDB207Y/1F (a-2)482, pJDB207R/LF (a1), pJLB207R/LIF(a-1)a, pJDB207Y/IF(a-1), and pJDBR207Y/IF (=1) 3,

Example 29: Transformation of Saccharomyces cerevisiae All220 and induction of interferon production

Plasmids pJDB207/1F2(5,)A72, pIDB207/1F2(5 )A82, pJDB207/1F2(1'b)A, pJDB207R/IF (3), pJDB207Y/IF (a3), pJDB207R/IF (a~2), pJDB207R/IF (a=2)472, pJDB207R/1F (a—2) 82, pJDB207Y/1F (u-2), pJDB207Y/1F (a-2)A72, pJDB207Y /1F («—-2)A82, pJDB207R/1IF(a-1), pJDB207R/IF («—1)A, pJDB207[/1F(a-1) and pJDB207([/IF(a-1)A (see Examples 22 and 28, respectively) are each introduced into Saccharomyces cerevisiae strain AN220 (a, trpl, leu2-3, leu2-112, his3, phos, phol) using the transformation protocol described by

Geen laAD ORIGINAL 2

R266

Hinnen et al. (1). Transformed yeast cells are selected on yeast minimal medium plates deficient in leucine. Single transformed yeast colonies are picked and grown as described in Example 7. The different yeast colonies are referred to as

Saccharomyces cerevisiae A1220,/pJDB207/1F2(5,)A72,

Saccharomyces cerevisiae AN220/pIDB207/1F2(5,)A82,

Saccharomyces cercvisiae AH220/pJDB207/1F2(1'b)A,

Saccharomyces cerevisiae Al220/pJDB207R/1LF(a-3),

Saccharomyces cerevisiae AN220/pIDB207Y/1F (a-3),

Saccharomyces cerevisiae AH220/pJDB207R/1F(a-2),

Saccharomyces cerevisiae Al1220/pJDB207R/1IF (a~2) 472,

Saccharomyces cerevisiae AN220, pJDB207R/IF(a-2) 282,

Saccharomyces cerevisiae AN220/pJDB207Y/ LF (a=2),

Saccharomyces cerevisiae AN220/pJDB207Y/1F(a-2)272,

Saccharomyces cerevisiae Al220/pJDB207Y/1F(a-2)082,

Saccharomyces cerevisiae Al1220/pJDB207R/IF (a-1),

Saccharomyces cerevisiae Al220/pIDB207R/IF (a-1) 4,

Saccharomyces cerevisiae AH220/pJDB207Y/IF(a-1) and

Saccharomyces cerevisiae AI220/pJDB207Y/TF(a=1)A. :

Example 30: Transformation of Saccharomyces cerevisiae strain GRF18 and induction of interferon production

Analogous to the process described in Example 29, Saccharomyces cerevisiae strain GRF18 (a, his3-11, his3-15, leu2-3, leu2-112, can") is transformed with the plasmids listed in Example 29. The different yeast colonies are referred to as

Saccharomyces cerevisiae CRI'18/pJDB207/1F2(5,)a72,

Saccharomyces cerevisiae GRF18,/pJDB207/1F2(5 )482,

Saccharomyces cerevisiae GRF18,/pJDB207/1F2(1'b)4,

Saccharomyces cerevisiae GRF18/pJDB207R/1F (a3),

Saccharomyces cerevisiae GRF18/pJDB207Y/1F(a-3),

Saccharomyces cerevisiae GRF18/pJDB207R/1F (a2),

Saccharomyces cerevisiae CRF13/pJDB207R/1F(a-2)472, aio ORIGINAL J

- 102 ~ 26266

Saccharomyces cerevisiae CRrY¥18 /pIDB207R/1F(a-2)482,

Saccharomyces cerevisiae GRF18 pJDB207Y/1F(a~2),

Saccharomyces cerevigiae GRri1e./pJPB207Y/1F(a-2)a72,

Saccharomyces cerevisiae GRF18 /pJDB207Y/1F(a-2)082,

Saccharomyces cerevisiae GRF18,/pJDB207R/1F(a-1),

Saccharomyces cerevisiae GRF18 /pJDB207R/1F(a-1)A,

Saccharomyces cerevisiae GRF18/pJDB207Y/IF(a-1) and

Saccharomyces cerevisiae GRF18,/pJDB207Y/1F(a-1)a.

Example 31: Preparation of yeast cell extracts and determination of the interferon titer

Cells from 50 ml of culture medium at a cell density of 1-2x10"/ml are collected by centrifugation in a Sorval GSA rotor for 10 min. at 3000 rpm. The cells are resuspended in 2 ml of 0.1 M KH,ro, pH 7.4, and centrifuged at room temperature for 5 min. at 3000 rpm. The sedimented cells are resuspended in 1.2 ml ice cold lysis mix {0.1 M potassium phosphate buffer pH 7.4, 1% (v/v) Triton X-100,0.1 mM PMSF (Merck)]l and transferred to a 30 ml Corex tube. 1.6 g of glass beads (0.4 mm in diameter) are added to the cells and the suspension is shaken on a

Vortex Mixer (Scientific Instruments Inc., USA) at full speed for sec. and then cooled for 1 min. in an ice bath. This shaking procedure is repeated 5 to 10 times until more than 90% of the cells are broken (check under light microscope). ’

NL Co RAE BAD ORIGINAL 9

Cell debris and glass beads are removed from the solution by ceuntri- fugation for 10 min. at 8000 rpm at 4°C in a Sorvall HB-4 rotor.

The supernatant is transferred to Eppendorf tubes, frozen in liquid nitrogen and stored at -60°C. Interferon activity is determined according to the procedure of Armstrong (32) using human CCL-23 cells and vesicular stomatitis virus (VSV) as the challenge virus. The results are summarized in Table 5.

The interferon activity in S. cerevisiae strains All220 and GRF18 after transformation with a recombinant plasmid is generally identical.

Table 6 shows a comparison of the interferon activities of both strains after transformation with the plasmids listed as examples. ea ORIGINAL P

Table 5: 9 A 26 QA

Interteron activity in S. cerevisiae strain AH 220 after transformation with the following recombinant plasmids:

Interferon activity expressed in plasmids Example | units/ml yeast cell]units/1l yeast cell extract culture py 7 8 ’ pJDB207/1F(8) 17 1-10 2-10 pJDB207R/IF(a-3) 28 1-108 2.10% pJDB207Y/1F (—3) 28 ’ 1-10° 2.10° 5 7

PpJOB207/1F1(5,) 17 7¢10 1.4410 7 « PIDB207/1F2(5)) 17 54107 1.010 4 pJDB207/1F3(5,) 17 3.103 6.3210" 5 7 pJDB207/1F2(5,)a72 22 5+10 1.10 7 pJDB207/1F2(5)482 22 5.10° 1.10 7 8 pJDB207R/IF (2) 28 1+10 2-10 g pJDB207R/IF(a-2)A72 28 1-10’ 2:10 pJDB20TR/LF(a-2) 582 28 1-10 2.10% 8 pJDB207Y/1F{(a~-2) 28 1-107 2-10 7 8 pJDB207Y/IF(a-2)A72 28 1-10 2-10 7 8 pJOB207Y/1F(a-2)A82 28 1-10 2-10 — 3 | % + pJDB207/1F2(1'D) 17 4410 8+10 4 pJDB207/IF2(1'b) a 22 1-10° 2:10 4 6 pJDB207R/IF (a-1) 28 5¢10 1-10 pJDB207R/IF (a—-1)A 28 4010" 8+10° pJDB207Y/LF(a-1) 28 5.10” 1010 pJDB207Y/IF(a—1)A 28 410" 8:10" - Co BAD ORIGINAL 9

- 105 ~ .

Table 6: Comparison of interferon activity in S. cerevisiae strains

Ali220 and CRF18 after transformation with the following recombinant plasmids:

Co ————————————————————— plasmids Interferon activity (units/l yeast cell culture

AH220 CRF18 . 8 8 pJDB207/1F (8) 2010 210 pJDB207R/1F (a-3) 2010° 2010°

I — — . 7 6 pJDB207/1F2(5;) 1+10 9e10 8 8 pJDB207R/IF (a-2) 2¢10 2410 ’ pJDB207R/ LF (-2) £82 2010° 2010°

Example 32: Production of interferon-a-2 by a recombinant strain of the yeast Saccharomyces cerevisiae on a 300 1 scale

HEE

Saccharomyces cerevisiae strain CRF18/pJDB207R/IF (a-2)A82 carries a plasmid which includes a leucine marker allowing selective maintenance of the plasmid in the host organism, a structural gene for luman interferon-a-2 and the acid phosphatase PHOS5 promoter which allows expression of the 1FN-a-2 gene in media with limiting amounts of inorganic phosphate.

The strain is maintained on agar slant cultures prepared with a de- fined medium lacking the amino acid leucine to ensure retention of the plasmid. Freshly inoculated slants are incubated for 24 hours at 30°C.

The surface culture of one slant is resuspended in 3 ml pre-culture medium which is then transferred to the first shake flask pre-culture.

The 500 ml flask has a single baffle and contains 100 ml pre-culture medium having the following composition (values in g/1):

Yeast extract (Difco), 10.03 L-asparagine, 6.6; KH, Fo, , 1.0;

Mg$0, + 71,0, 1.0; L-histidine, 0.02 and D-glucose (monohydrate), 33.0.

BAD ORIGINAL, & \

~- 106 - al 26266

The medium which has been prepared using deionised water, has a pll value of approximately 6.0. The glucose is sterilised separately. This first pre-culture is incubated for 24 hours at 30°C on an orbital shaker with 5 cm throw at a speed of 250 rev/min.

The first pre-culture flask provides the inoculum for the second pre-culture flasks. These flasks receive an inoculum level of 17% v/v. The medium and incubation conditions are identical with those for the First pre-culture. The culture broths from 36 such flasks are combined to provide a 1% v/v. inoculum for the main production fermenter.

The production fermenter has a total volume of approximately 500 1, contains 4 baffles and a single six-bladed disc turbine agitator with a diameter of 230 mm. The agitation rate is 450 rev/min, the overpressure 0.3 bar and the aeration rate is 1 vol/vol /min.

The fermenter contains 300 | of a medium with the following composi- tion (values in g-/1): L-asparagine, 2.0; L-histidine, 0.02;

KIPO,, 0.03; MgSO, ¢7H,0, 0.5; NaCl, 0.1; CaCl,*2H,0, 0.1; KCl, 1.0;

D-glucose (monohydrate), 20.0; vitamin solution, 5 ml/l and trace element solution, 5 ml/l. The medium is adjusted to pH7.2 using NaOH before sterilisation. The glucose, vitamins and trace elements are sterilised separately and added to the medium. The stock solutions for vitamins and trace elements have the following compositions (in g/1): Vitamins - biotin, 0.0002; calcium-D-pantothenat, 0.04; folic acid, 0.0002; nicotinic acid, 0.04; p-aminobenzoic acid, 0.02; pyridoxine hydrochloride, 0.04; riboflavin, 0.02; thiamine hydrochloride, 0.04; and inositol, 0.2 in 1 1 of deionised water; trace elements — boric acid, 0.05; CuS0,*51,0, 0.004; KI, 0.01;

FeCl,6l,0, 0.02; MnS0, 4H, 0, 0.04; Na, MoO, +210, 0.02 and

ZnSO, *7H,0, 0.04 in 1 1 of deionised water. The fermentation temperature is 30°C. The pH value falls to a value of about 4.0-4.2 but can be controlled if desired at an intermediate value using sodium hydroxide. After fermenting for about 18 hours the maximum lao ORIGINAL J §

~ 107 - 26266 yield of interferon is reached [as determined reco Of

Armstreng (32)]). The optical density, which reaches about 2.0 units, and the acid phosphatase activity are useful indications of the progress of the fermentation. The fermentation broth may be cooled to 10°C if required prior to harvesting of the yeast cells.

Example 33: Isolation and purification of HLyIFN-a-2 a. Preparation of the polypeptide solution for the monoclonal antibody column

A total volume of 600 1 of culture broth having a pH of 4.1 is cooled to 10°C. The cells are separated using an Alfa-Laval BRPX-207 de-sludger centrifuge. The clear supernatant contains no IFN-activity.

Residual supernatant liquor entrained with the cells is displaced by washing with 20 1 Lysis Buffer A [100 mM Kit, po, , 500 mM NaCl, 0.17 v/v Triton x-100® and 0.1 mM PMSF adjusted with KOH to pH 7.5].

The contents of the centrifuge bowl (7 1) are ejected with complete desludging and the de-sludger washed once with 5 1 Lysis Buffer A.

The cell mass obtained is diluted with Buffer A to 60 1 and has a pH value of 7.3. The suspension is cooled to 5-10°C and passed through a pund®mit (type KD5) at a feed rate of 100 1h.

The mill is equipped with polyurethane agitator discs and 4200 ml glass beads of 0.5-0.75 mm diameter and is operated at 1625 rev/min.

The ruptured cell suspension (pH ~~ 7.3) is centrifuged as described previously: The supernatant (75 1) is concentrated to 3 1 by ultra- filtration. An aliquot (300 ml) of this polypeptide solution is passed through a HIP100 Hollow filter cartridge using an Amicon DC-2

Hollow Fibre System. A further 2 1 of buffer system B [30 mM

Tris-Cl, 500 mM NaCl, adjusted to pll 8.5] is applied to the filter.

The combined filtrate and washings (2 1) are concentrated to 100 ml by means of a HIPLO Hollow filter cartridge. The concentrate is adsorbed onto a column of DEAE-Trisacryl © DEAE (LKB Ltd.). The column is washed and then eluted with Buffer C (200 mf NaCl, 25 mM

Tris-HCl at pit 8.5). The eluate has an interferon activity of

J ORIGINAL JP be i or 26266 1.4 x 10° 10 my polypeptide when assayed according to the method of

Armstrong (32). The eluate is stored at -20°C prior to further puri- fication on the monoclonal antibody column. b) Purification of human LyIFN-o-2 on a monoclonal antibody column

The monoclonal autibody column NK2 (purchased from CELLTECH U.K.) (bed volume 20 ml), is equilibrated with 20 mH Na-phosphate, 154 mit NaCl, pll 7.4 and portions of the above polypeptide solution are applied onto the column at room temperature with a flow rate of 50 ml/h. The first fractions containing the nonadsorbed polypeptides and 100 ml of PBS washings are discarded. Further non specific bound polypeptides are eluted with 110 ml of PBS containing additional 0.5 M NaCl and 0.27

Triton X 10d®, The column is washed with 200 ml of 20 mM Na-phosphate, } 0.3 M NaCl, pH 7.4, whereupon the specifically adsorbed polypeptides are eluted with 50 ml of Buffer D (0.1 M citric acid, 0.3 M NaCl, pH 2). This solution is adjusted to pH 6.3 with 2N NaOH and concen- trated at 4°C with the aid of an immersible-CX | molecular separator

Millipore. The concentrate is applied onto a Sephadex c-2® fine column (2.6x34 cm, 200 ml bed volume) equilibrated with 0.025 M histidineelCl at pH 6.3. The column is eluted with the same oo histidineeliCl buffer at 4°C and with a flow rate of 42 ml. 20 fractions are collected of each 10.5 ml. Polypeptide containing fractions are detected by their optical absorption at 280 nm.

Fractions 7 and 8 contain the polypeptide with IFN activity as localised by the assay according to Armstrong (32). The active fractions containing LyIFN-a-2 are stored at -20°C until further use.

The IFN activity of the fractions is 1.8010° 1U/mg polypeptide (32). ’

By lyophilizing the above fractious from 1 ml solution 20-40 ug of polypeptide are obtained. .

SDS polyacrylamide gel electrophoresis (cf. (53)) reveals a molecular weight for the obtained LyIFN-o-2 of about 18 kDaltous. -

Ce a0 ORIGINAL I) (Hi

- 109 -- 0 Q . fooz' 1206 6

Example 34: Secretion of interferon by transformed yeast cells into the culture med ium

In order to determine the effect of a N-terminal protein signal sequence on protein secretion, yeast strain S. cerevisiae GRF18/pJDB207/LF(8)) (containing a hybrid signal sequence, see example 17) and yeast strain

S. cerevisiae CRF18/pJDB207R/LF(a-3) (without signal sequence) are grown as described in Example 28. The amount of the produced interferon present in the culture medium as well as the amount of interferon present in cell extracts. (prepared as described in Example 31) is determined and the results are given in table 7.

Table 7: Comparison of interferon secretion of transformed

S. cerevisiae GRFI8 strains into the culture medium: ——

S. cerevisiae strain Interferon activity (units/l yeast cell culture) cell extract culture medium 9 4

JRF18,/pJDB207R/LF (a~3) 1.5 « 10 <3 +10 8 7

SRF18,/pJDB207./LF (81) 2 + 10 2 «10 } ee ’ =" .

BAD ORIGINAL 9

Refevences 0 . «6266

L. A. Hinnen ct al., "Transformation of yeast", Proc. Natl. Acad.

Sci. USA 75, 1929 (1978) 2. J.D. Beggs, "Transformation of yeast by a replicating hybrid plasmid”, Nature 275, 104 (1978) 3. J. licks et al., "Properties ,'f yeast transformation', Cold Spring

Hlarbor, Symp. Quant. Biol. 43, 1305 (1979) 4. K.Struhl et al., "High-frequency transformation of ycast; auton- omous replication of hybrid DNA molecules", Proc. Natl. Acad. Sci.

USA 76, 1035 (1979) 5. R.A. Hitzeman et al., "Expression of a human gene for interferon in yeast", Nature 293, 717 (1981) 6. J.D. Beggs et al., "Abnormal expression of chromosomal rabbit f-globin gene in Saccharomyces cerevisiae", Nature 283, 835 (1980) 7. R. Axel, "The use of cukaryotic promoter sequences in the produc- tion of proteinaceous materials", PCT patent application 81/02425. 8. J.A. Carbon et al., "DNA capable of replication and stable mitotic maintenance in a host cukaryote, method of preparation thereof and eukaryotic cell containing same", European patent application 48081 (The regents of the University of California) 9. D.T. Stinchcomb et al., "Eukaryotic autonomously replicating segment", European patent application 45573 (The board of trustees of Leland Stanford Junior University) 10. "Plasmidvektoren, Hybridplasmide und ihre Verwendung zur llerstel- lung von Protcinen”, German Offenlegungsschrift 2923 297 (Insti- tut Pasteur)

BAD ORIGINAL J

11. "Procédé dec production de protéines par expression des gtnes correspondants dans des microorganismes et vecteurs susceptibles d'étre mis en ocuvre dans de tels procédés', Treuch Patent appli- cation 2 458 585 (Institut Pasteur) 12. M. Aigle et al., "Nouveaux plasmides hybrides et microorganismes les contenant", European patent application 11562 (Agence natio- nale de valorisation de la recherche) 13. A. Schurr et al.; J. Gen. Microbiol. 65, 291 (1971) and A. Toh-e et al., Mol. Gen. Genet. 162, 139 (1978) 14. P. Mildner et al., Biochim. Biophys. Acta 429, 274 (1976) 15. A.M. Maxam et al. in "Methods in Enzymology", vol. 65, p.499,

New York 1980 16. A. Hinnen et al., "Vectors for cloning in yeast", Curr. Top.

Microbiol. Immunol. 96, 101 (1982) 17. A. Hinnen et al. in "Eukaryotic Gene Regulation", vol. 14, p. 43,

New York 1979 ' * 18. "Genetic Engineering" (ed. A.M. Chakrabarty), West Palm Beach 1978 19. J. Lodder, "The Yeasts", Amsterdam 1971 20. M. Grunstein et al., Proc. Natl. Acad. Sci. USA 72, 3961 (1979) 21. A. Jimenez et al., Nature 287, 869 (1980) . 22. T. Staehelin et al., J. Biol. Chem. 256, 9750 (1981) 23. M.V. Olson et al., J. Mol. Biol. 132, 387 (1979) 24. WN, Meyer, FEBS Lett. 90, 344 (1978) 25. B. Hohn et al. in "Genetic Engineering", vol. 2, p. 169,

New York 1980 26. B. Hohn in "Methods in Enzymology", vol. 68, p. 299, New York 1979 27. N. Mantei et al., Gene 10, 1 (1980) 28. J.D. Beggs in "GCeuctic Engineering", vol. 2, p. 175, New York 1981

29. M. Mandel et al., J. Mol. Biol. 53, 159 (1970) 30. J.H. Miller, "Experiments in Moleculur Geneties', Cold Spring

Harbor 1972 31. A. Toh-e et al., J.Bacteriol. 113, 727 (1973) 32. J.A. Armstrong, Appl. Microbiol. 21, 723 (1971) 33. J.B. Gurdon, J. Embryol. Exp. Morph. 20, 401-414 (1968) ; 34. Barth, J. Embryol. Exp. Morph. 7, 210-222 (1959) 35. A, Colman, Cell 17, 517 (1979) 36. A. Efstratiadis et al., Cell 4, 367-378 (1975) 37. T. Maniatis et al., Cell 8, 163-182 (1976) 38. J.H.J. loeijmakers et al., Gene 8, 391-417 (1980) 39. A.C. Peacock et al., Biochemistry 6, 1818 (1967) 40. K. Itakura et al., J. Am. Chem. Soc. 97, 7327 (1975) 41. J.F.M. de Rooij et al., Recl. Trav. Chim. Pays-Bas 98, 537-548 (1979) 42. W. Mueller et al., J. Mol. Biol. 124, 343 (1978) 43. D.V. Goeddel et al., Nature 290, 20 (1981) 44, C. Weissmann, European patent application No.32134 (Biogen N.V.) 45, T.Taniguchi et al., Gene 10, 11 (1980) 46. M. Streuli et al., Science 209, 1343 (1980) 47. Perlman et al., Proc. Natl. Acad. Sci. USA 79, 781 (1982) 48. A.J. Berk et al., Cell 12, 721-732 (1977) 49, J. Messing, In the 3rd Clevsland Symposium on Macromolecules:

Recombinant DNA (ed. A. Walton), Elsevier, Amsterdam 1981, pp. 143-153 © 50. D.S. Holmes et al., Anal.Biochem. 114, 193-197 (1981)

51. N.M.

Gough et al., J.

Mol.

Biol. 162, 43-67 (1982) 52. pasek ei al., Nature 282, 575-579 (1979). 53. U.K.

Laemmli, Nature 227, 680-685 (1970). ‘

: Appendix 9 6 26 6 = oP

Symbols used in figures 10-14 of the accompanying drawings have the following meanings: amino acid exchange * and 2 nucleotide exchange “To sequence not present in prior art v polyadenylation sites in prior art deletion of a nucleotide ' insertion of a nucleotide .

In the respective figures the indicated symbols are referring to the closest prior art references as mentioned in Example 10. °°

In the other figures of the accompanying drawings, the symbol used have the following meanings: ,

26266 +

Fszoraz HBVs gene =X deletion of a restriction site human Iymphoblastold ————— pBR 322 sequences . nmuenl® interferon gens yeast chromosomal ONA errr yeast chromosomal ONA : orem go vived from P03, PHOS region derived from IRPI region —Y—

PIII yeast 2ueplasnid DNA deletion in pBR 322 yeast chromosomal DNA _ 4 restriction site derived from LEU 2 region

R amp ampicillin resistance gene em direction of transcription

R ——e tet tetracycline resistance gene : 4 = linker DNA stretch

Claims (14)

~- 116 ~ + . CLAIMS

1. A hybrid vector comprising the yeast PHO3 ‘OY PHOS5 promoter and a yeast or a non-yeast polypeptide coding region which is controlled by said promoter. ’ rs

2 . A hybrid vector according to claim 9, wherein the yeast acid phosphatase promoter is the PHOJ promoter.

3 . A hybrid vector according to claim 9, wherein the yeast acid phosphatase promoter is the PHOS promoter.

’ 4. A hybrid vector according to claim 9 wherein the yeast or non-yeast polypeptide coding region codes for an enzyme, . a hormone, a polypeptide with immunomodulatory, anti-viral or anti- , cancer properties, an antibody, a viral antigen, a vaccine or a clotting factor.

: 5. A hybrid vector according to claim 12 wherein the yeast or non- yeast polypeptide coding region codes for a human interferon.

6.A hybrid vector according to claim 12 wherein the yeast or non- veast polypeptide coding region codes for a hepatitis B virus antigen. a

7.- A hybrid vector according to claim 9 wherein the transcripts coding for the yeast or non-yeast polypeptide end in a DNA sequence containing transcription termination signals of a yeast gene.

8 . A hybrid vector according to claim 16 containing transcription termination signals of the PHOS gene. )

9 . A hybrid vector according to claim 9 wherein the yeast or non- yeast polypeptide coding region is preceded by a8 signal sequence.

10. A hybrid vector according to claim 9 wherein the yeast or non- yeast polypeptide coding region is joined to the acid phosphatase promoter region between the major acid phosphatase mRNA start and the ATG of the acid phosphatase coding region.

11. The hybrid vector pJDB207R/IF (0-2) according to claim 9.

12. The hybrid vector pJDB207./PHO5/HBVsAl4 according to claim 9.

13. The hybrid vector pJDB207/PHO5/HBVsAl4t according to claim 9.

14. The hybrid vector pJDB207R/IF (a-2)A82 according to claim 1.

I ' . 9 A > 6 4-140%0/1-3/+/C 266 — YEAST HYBRID VECTORS AND THEIR USE FOR THE PRODUCTION OF POLYPEPTIDES Abstract of the disclosure There are provided new hybrid vectors comprising the yeast PHO3 or PHO5 promoter and a yeast or non-yeast polypeptide coding region which is controlled by said promoter.

The hybrid vectors are useful for the production of polypeptides. i