DK173671B1 - Process for the selection of transfixed vertebrate cells - Google Patents

Description

in DK 173671 B1

The general principle of using a host cell to produce a heterologous protein, i.e. a protein which is not normally produced by this cell is well known. However, there are many technical difficulties in obtaining reasonable amounts of the heterologous protein using vertebrate host cells, which is desirable because of their properties in handling the protein.

There have been a number of successful examples of incorporating genetic material encoding heterologous proteins into bacteria and obtaining expression thereof. Eg. Thus, human interferon, desdcetyl-thymosin α-1, somatostatin, and human growth hormone have been produced. Recently, it has been possible to utilize non-bacterial hosts, such as yeast cells (see, for example, EP of Fentilization Scroll No. 0 060 057) and vertebrate cell cultures (see EP Publication No. 0 073 656), as hosts. The use of 20 vertebrate cell cultures as hosts in the production of mammalian proteins is advantageous because such systems have additional capabilities for modification, glycosylation, addition of transport sequences and other subsequent processing of the resulting peptide time produced in the cell. While e.g. bacteria can be successfully transfected and expressed "α-thymosin", the polypeptide produced N-acetyl group is lacking in the "natural" α-thymosin found in the mammalian system.

30 ».

Generally, the gene-mirroring technique designed to enable Urt cells to produce heterologous proteins is the production of an "expression vector" which is a DNA sequence containing (1) a "promoter", i.e. a sequence of nucleotides that control and enable the expression of a coding sequence; (2) a sequence that provides mRNA with a ribosome binding site; (3) a "coding region", i.e., a sequence of nucleotides encoding the desired (4) a "termination sequence" which allows completion of the transcription when the entire code of the desired protein has been read; and (5) if the vector is not directly inserted into the genome, a "replicon" or replication origin, which allows the entire vector to be reproduced once in the cell.

O In the construction of vectors, the same promoter controls two coding sequences, one for a desired protein and the other for a secondary protein. The transcription termination is also shared by these sequences. However, the proteins are produced in a separate form because they are separated by a stop and start translation signal. 1

Generally, the genetic expression vectors take the form of plasmids, which are extra-chromosomal loops of double-stranded DNA. These are found in natural Form 3 DK 173671 B1 in bacteria, often in multiple copies. However, artificial plasmids (and these, of course, the most useful) can also be constructed by splicing the four necessary elements outlined above in proper order using appropriate "restriction enzymes". Restriction enzymes are nucleases whose catalytic activity is limited to lysing at a particular base sequence, each base sequence being characteristic of a particular restriction enzyme. By skillfully constructing the terminal ends of the elements outlined above (or fractions thereof), it is possible to have restriction enzymes spliced these elements together to form a final genetic expression vector.

Back then is to have the host cell incorporate the vector (transfection) and to grow the host cells in such a way as to produce synthesis of the desired polypeptide as a companion phenomenon 20 to normal growth.

Two typical problems are associated with the procedure outlined above. First, in addition to the four necessary elements outlined above, it is desirable to have a marker which will allow a direct selection for the cells that have actually received the genetic expression vector.

In the use of bacterial cells as hosts, frequently used markers are resistance to an antibiotic such as tetracycline or ampicillin. Only the 30 cells resistant to the agent will grow in cultures containing the antibiotic. Therefore, if the cell culture sought to be transfected is grown on a medium containing the antibiotic, only the actually transfected cells will appear as colonies. Since 4 DK 173671 B1 tans format in the οηκ Trek vein is quite low (about one cell per 10 is transfected under ideal conditions), this is practically a necessary prerequisite.

For vertebrate cells as hosts, the degree of transformation achieved is more efficient (about one cell per cell).

10 ^). However, a convenient selection remains important for obtaining the desired transfected cells. Selection is also made important because the cell division rate is about 50 times lower than in bacterial cells, i.e. whereas E. coli divides once for approx. every 20-30 minutes, human tissue culture cells divide only once every 12-24 hours.

One aspect of the present invention solves the problem of selecting for vertebrate cells that have taken up the genetic expression vector of the desired protein by utilizing the expression of the coding sequence for a secondary protein, such as a necessary enzyme for which the host cell is deficient. Eg. For example, dihydrofolate reductase (DHFR) 20 can be used as a marker using host cells deficient in DHFR.

Another problem in the production of polypeptides in a foreign host is the extraction of satisfactory amounts of protein. It would be desirable to have some mechanism for regulating, and preferably increasing, the production of the desired heterologous polypeptide. According to another side of the connection, a secondary coding sequence, which can be influenced by outside controlled parameters, is utilized to enable control of the expression ice by controlling these parameters. Furthermore, the provision of both sequences on a pol y-cistron per se allows selection of transformants with high expression levels of the primary sequence.

5 DK 173671 B1

It has been demonstrated that DHFR coding sequences can be introduced into, expressed in, and amplified in mammalian cells.

Genomic DNA from methotrexate-resistant Chinese hamster ovary (CHO) cells has been rapidly introduced into 5 mouse cells and results in transformants that are also resistant to methotrexate (1). The mechanism by which methotrexate (MTX) resistance develops in mouse cells is found to be threefold; by re-amplifying the DHFR coding sequence (2,3,4), 10 by decreasing the uptake of MTX (5,6) and by reducing the affinity of the DHFR produced for MTX (7).

It is found that amplification of the DHFR gene upon exposure to MTX can result in a concomitant amplification of a co-transfected gene sequence. It has also been shown that mouse fibroblasts transfected with both a plasmid containing hepatitis B DNA sequences and genomic DNA from a hamster cell line containing a mutant gene for MTX-resistant DHFR secrete increased amounts of hepatitis B surface antigen (HBsAg). in the medium when MTX is used to stimulate DHFR sequence amplification (B). Furthermore, mRNA encoding the E. coli protein XGPRT is amplified in the presence of MTX in CHO cells cotransfected with DHFR and XGPRT gene sequences under the control of independent promoters (9). Finally, increased expression of a sequence endogenous to the promoter has been demonstrated in a DHFR / SV40 plasmid combination in the presence of MTX (10).

The present invention is based on the recognition that in vertebrate cell hosts, where the genetic expression vector of a desired polypeptide contains a secondary genetic coding sequence under the control of the same promoter, this secondary sequence provides an appropriate selection marker for both Trans ferments in general ng for transformants exhibiting high expression levels For the primary sequence, while serving as a control device whereby the expression of a desired polypeptide can be regulated, is most frequently increased.

This is particularly important as the two proteins of the process of the invention are produced separately in mature form. Although both DNA coding sequences are controlled by the same transcription promoter 10 to form a linked mRNA, they are separated by a translation stop signal for the first and a translation start signal for the second, so that Lo independent proteins are generated.

Since a vertebrate host cell culture system is often advantageous because it is capable of glycosylation, phosphorylation and lipid association appropriate to animal systems (while bacterial hosts are not), it is important that marker systems and regulatory systems be provided in this context. .

Accordingly, the invention relates to a method for selecting vertebrate cells that has been transfected with an expression vector capable of expressing a desired protein, which method is characterized in that (a) the cells are treated with a vector containing code sequences the desired protein and a secondary protein, the presence of which is required for the growth of the host cells under selective culture conditions, both coding sequences operably linked to the same promoter sequence but separated by translation stop and start codons, and the (b) cells being cultured under the selective cultivation conditions.

DK 173671 B1 7. In the drawings, FIG. 1 the construction of an expression vector for 5 HBsAg, pE342.HS94.HBV; FIG. 2 the construction of an expression vector for DHFR, pE342.D22; and FIG. 3 the construction of expression vectors for DHFR and HBsAg, fiE342.HBV.D22 and pE342.HBV.E400.D22.

A. Definitions In this specification description, "plasmids" include both naturally occurring plasmids in bacteria and artificially constructed circular DNA fragments.

"Expression vector1" means a plasmid containing at least the four necessary elements as set forth above for expression of the heterologous peptide in a host cell culture.

"Heterologous protein" means a protein or peptide that is not normally produced by or required for the viability of the host organism.

"Desired protein" means a heterologous protein or peptide which the process of the invention is designed to produce.

"Secondary peptide" means the protein or peptide which is not the heterologous peptide desired as the primary product of the expression ice in the intestinal cell, but rather another heterologous peptide which, owing to its own properties or because of the properties of the sequence encoding it, it is able to "mark" transfection of the expression vector and / or regulate the expression of the primarily desired heterologous peptide.

The peptide sequence may be either long or short in the range of about five amino acids to about 1000 amino acids. The conventional distinction between the words peptide and protein is not routinely followed in the description of the invention. If there is a distinction, 15 it will be stated.

"Primary sequence" is the nucleotic sequence encoding the desired peptide, and "secondary sequence" means a sequence of nucleotides encoding the secondary peptide.

"Transfection" of a host cell means that the expression vector has been recorded by the host cell in a traceable manner, whether or not any coding sequences are actually expressed. In the context of the invention, successful transfection will be recognized when there is any indication of the effect of this vector in the host cell. It is recognized that there are different levels of luck in this context. First, the coding sequence of the vector may or may not be expressed.

However, if the vector is properly constructed with the inclusion of promoter and terminator, then expression is highly likely. Second, if the plasmid representing the vector is taken up by the cell and expressed but not incorporated into the normal chromosomal cell material of the cell, the ability to express this plasmid will be lost after a few generations. On the other hand, if the vector is taken up into the chromosome, the expression ice remains stable through repeated replications of the host cell. There may also be an intermediate result. The precise details of the manner in which transfection can thus occur are not understood, but it is clear that there is an experimental continuum of results regarding the stability of expression ice over several generations of the host culture.

8. A Preferred Embodiment of the Desired Peptide In a preferred specific embodiment of the invention, the primary genetic sequence encodes hepatitis B surface antigen (HBsAg). This protein is derived from hepatitis B virus, the human hepatitis B infectious agent. This disease is characterized by impairment, ulcer damage, primary carcinoma and often death. The disease is quite widespread, especially in 20 many African and Asian countries where many people are chronic carriers With the ability to transmit the disease pandemically. The virus (HBV) consists of a DNA molecule surrounded by a nuclear capsid, which in turn is surrounded by a sheath. Proteins associated with this virus include the surface antigen (HBsAg), a core antigen and a DNA polymerase. HBsAg is known to produce antibodies in infected humans. HBsAg, which is found in the serum of infected individuals, consists of protein particles with an average diameter of approx.

30 22 nm and is therefore called "22 nm particles". Accordingly, it is believed that the HBsAg particle would be an effective basis for a vaccine.

A preferred embodiment of the secondary peptide

It has been recognized that environmental conditions are often effective in controlling the amount of certain enzymes produced by cells under certain growth conditions. In the preferred embodiment of the invention, advantage is taken of the sensitivity of certain cells to methotrexate (MTX), which is an inhibitor of dihydrofolate reductase (DHFR). DHFR is an enzyme required indirectly by synthesis reactions involving the transfer of single-carbon units. Lack of DHFR activity results in the inability of cells to grow, except in the presence of those compounds which otherwise require transfer of one-carbon units for their synthesis.

However, cells lacking DHFR will grow in the presence of a combination of glycine, thymidine and hyposan-thin.

Cells that normally produce DHFR are known to be inhibited by methotrexate. Most often, adding appropriate amounts of methotrexate to normal cells will result in cell death. However, some cells appear to survive the methotrexate treatment by producing increased amounts of DHFR, thus exceeding the capacity of the methotrexate to inhibit this enzyme (2,3,4). It has been previously shown that in such 25 cells there is an increased amount of m-RNA encoding the DHFR sequence. This is explained by assuming an increase in the amount of DNA in the genetic material encoding this m-RNA. Genetically, the addition of methotrexate apparently causes re-amplification of the 50 DHFR gene. Genetic sequences that are physically associated with the DHFR sequence, although not regulated by the same promoter, are also enhanced (1,8,9,10). As a result, it is possible to use the amplification of the DHFR gene that occurs with methotrexate treatment for DK 173671 B1

i J

simultaneously amplifying the gene for another protein, in this case the desired peptide.

Furthermore, if the host cells into which the secondary sequence of DHFR is introduced are inherently DHFR-deficient, DHFR also serves as an appropriate marker for selection of cells that are successfully transfected. If the DHFR sequence is effectively linked to the sequence of the desired peptide, this ability also serves as a marker for successful transfection - 10 with the desired sequence.

D. Applied Vector Construction Technique (Materials and Methods) ____

The vectors constructed in the examples listed in Section E are constructed by cleavage and ligation of isolated plasmids or DNA fragments.

Cleavage is carried out by treatment with restriction enzyme (s) in a suitable buffer. Generally, about 20 µg of plasmid or DNA fragments inherit about 1-5 units of enzyme in 200 µl of buffer-20 solution. (Suitable buffers for certain restriction enzymes are specified by the manufacturer). Incubation times of about 1 ti (ns at 37 ° C are convenient.

After incubations, protein is removed by extraction with phenol and chloroform and the nucleic acid is recovered from the aqueous fraction by precipitation with ethanol.

If blunt ends are required, the preparation is treated for 15 minutes at 15 ° C with 10 units of Polymerase I (Klenou /), phenol-chloroform extracted and ethanol precipitated.

Dry fragmentation of the cleaved fragments is carried out using a 6% polyacrylamide gel as described by Goeddel, D. et al., Nucleic Acids Res J3: 4057 (1980), which is considered to be incorporated in this specification by this reference.

For ligation, about equimolar amounts of the desired components, provided with appropriate tails at the ends to provide proper alignment, are treated with about the units of T4 DNA ligase per cell. 0.5 µg DNA.

10, E. DETAILED DESCRIPTION OF A PREFERRED EMBODIMENT In general, the expression vector calculated for the present invention is constructed by adaptation of the reflection technique. The starting material is a naturally occurring bacterial plasmid, if desired, modified in advance. In a preferred embodiment of the invention, a pML plasmid is used which is a modified pBR 322 plasmid Prepared according to Lusky, M. et al., Nature 239: 79 (1981), which is provided with a single promoter derived from the monkey virus 5V-40 and with the coding sequence for DHFR and for HBsAg.

In construction, the promoter (as well as a ribosome binding sequence) is placed above the coding sequence encoding a desired protein and a sequence encoding a secondary protein. A single transcriptional termination sequence is located below both. At the end of the top coding sequence, a translation stop signal is placed and a translation start signal begins the bottom sequence. Thus, the expression of the two coding sequences results in a single mRNA strand but two separate mature proteins.

In a particularly preferred embodiment, the sequence encoding the secondary peptide is below that encoding the desired peptide. Under these circumstances, procedures designed to select for the cells transformed with the secondary peptide will also select for particularly elevated production of the desired peptide.

F. Examples

The following examples serve to elucidate 10 of the invention.

EXAMPLE 1

Vector containing the HBpAq sequence, pE342.HS94.HBV FIG. Figure 1 shows the construction of the HBsAg plasmid.

i · · 1.

The 1986 bp EcoRl-BglI | fragment spanning the surface antigen gene was isolated from the HBV viral genome cloned with pBR3.22 as described by Liu et al., DNA 1: 213 (1982), def. incorporated in this description by this reference. This sequence was ligated between the Eco RI and Bam HI sites of pML, a pBR322 derivative lacking sequences that are inhibitory to its replication in monkey cells, as described by Lusky et al., Nature 293; 79 (1981) which is considered to be incorporated in this specification by this reference. In the single plasmid EcoRI-25 site of the resulting plasmid, the 342 bp original SV40 fragment obtained by HindIII-PvulI degradation was inserted into the viral genome, which had been modified to be bound by EcoRI restriction sites resulting in p342E (also referred to as pHBs by Levinson et al. in EP Publication No. 0 073 656, which is considered to be incorporated in this specification by this reference.

(Briefly, the origin of the monkey virus SV4D was isolated by degrading SV40 DNA with HindIII and converting the Hindi I I ends to EcoRI ends 5 by adding a converter (AGCTGAATTC). This DNA was cut with PvuII and After digestion with EcoRI, the 348 bp fragment spanning the original was isolated by polyacrylamide gel electrophoresis and electroelution and cloned into pBR322. The expression plasmid pHBs348-E was constructed by cloning the 1986 bp fragment from EcoRI and BglII Degradation of HBV (Animal Virus Genetics, (Ch. 5) Acad. Press. NY

(1980)) (spanning the gene encoding HBsAg) into the plasmid pML (Lusky et al., Nature 293: 79, 1981) at the EcoRI and BamHI sites. (pML is a derivative of pBR322 which has a deletion that removes the sequence which is inhibitory for plasmid replicate ion in monkey cells). The resulting plasmid (pRI-Bgl) was then linearized with EcoRI, and the 34B bp fragment representing the SV40 origin region was introduced into the EcoRI site of pRI-Bgl. The origin fragment can be inserted in any orientation. Since this fragment encodes both the early and late SV40 promoters in addition to the origin of replication, HBV genes could be expressed under the control of one or the other promoter depending on this orientation (pHBS348-E represents HBs expressed under the control of the early promoter) . pE342 is modified by partial degradation with EcoRI, filling in the digested site using Klenow DNA polymerase I and comparing the plasmid again, removing the EcoRI site prior to SV40 origin in pE342. The resulting plasmid, designated pE342 ARIτ, is digested with Eco RI, filled in using Klenow 35 DNA polymerase I and digested with BamHI. Following electrophoresis on acrylamide gel, the fragment of DK 173671 B1 is electroeluted for approx. 35Q bp, abstracted with phenol-chloroform and trap 3 with ethanol as above). The 5 'untranslated leader region of HBsAg was removed by treatment with EcoRI and with Xba, and the analog 150 bp 5 EcoRI-Xba fragment from a hepatitis expression plasmid pHS94 (Liu et al., Above) was inserted into its site to form pE342.HS94.HBV.

(As described by Liu et al., PH594 contains the translational start codon of the authentic HBsAg gene, but lacks all 5 'untranslated messenger sequences. The expression levels of both the authentic EcoRI-BgIII and the pHS94-derived equivalent under the control of the early SV40 promoter as described above is equivalent and is interchangeable without affecting the behavior of the plasmid).

EXAMPLE 2

Vector containing the DHFR sequence pE342.D22

A plasmid bearing DHFR as the only express sequence is pE348.D22, the construction of which is shown in FIG. 2nd

The 20,600 bp PstI insert in the DHFR cDNA plasmid DHFR-11 (Nunberg et al., Cell 1: 935, 1980) was treated with the exonuclease Bal31 to remove the poly-G: C region up to the Pst I gene. sites, were digested with BglII, and the resulting fragments of about 660 bp 25 were isolated from the gels. The Bal31-BglII-digested cDNA was ligated into a pBR322 plasmid derivative containing a BglII stcd. (After digestion of pBR322 with Hind III, the plasmid fragment was loaded using Klenow DNA polymerase in the presence of the four deoxynucleotide triphosphates and subdivided with BglII). The reciting plasmid, pDHFR-D22, has an Eco RI site located 29 bp above the 29 bp junction site between pBR322 and the 5 'end of the DHFR cDNA. The EcoRI-BglII fragment enclosing the coding sequences of the cONA insert was then excised from 5 pDHFR-D22 and ligated to EcoRI-BamHI digested pE342.HBV (Example 1) to generate the DHFR expression plasmid pE342.D22.

EXAMPLE 3

Vectors containing both DHFR and HBsAq sequences Two such vectors, pE342.HBV.D22 containing a polycistron, wherein the DHFR gene is below the HBsAg gene, and pE342.HBV.E400.D22 were constructed (Figs. 3) in which the genes encoding DHFR and HBsAg are not polycistronic.

A. pE342.HBV.D22 was constructed by ligation of

EcoRI-Taql fragment from cloned HBV DNA (Liu et al., Supra) to EcoRI-Clal digested pE342.D22.

B. This plasmid was further modified by associating an additional early SV40 promoter between the 20 BglII site and the Cla I site of the DHFR insert in pE342.HBV.D22 to form pE342.HBV.E400.D22.

Viral HBV DNA contains a single Taq1 site 20 bp outside the BglII site used to create the EcoRI-BglII fragment enclosing the surface antigen gene. Thus, EcoRI and Taql degradation of cloned viral HBV DNA results in a fragment of ca. 2,000 bp spanning the surface antigen gene and containing a single BglII site (1985 bp from the EcoRI site (Liu et al., Supra)). (The Taql and Clal ends of DNA fragments formed by degradation are cohesive and will align together).

17 DK 173671 B1

The Clal site is restored; thus, pE342.HBV.D22 contains both. BglII and Clal site located immediately in front of the DHFR coding sequences.

An SV40 origin bound at restriction sites 5 cohesive with the BglII and Clal sites of pE342.HBV.D22 was constructed by digesting SV40 DNA with Hpall, filling as described above, and subdividing with HindIII. A 440 bp fragment spanning the original was isolated. This was ligated by a tripartite ligation to the 4,000 bp pBR322 fragment created by HindIII and BamHI degradation and to the 1 986 bp fragment spanning the surface antigen gene created by degradation of the cloned viral HBV DNA with EcoRI , filling with Klenow DNA polymerase I, degradation with BglII and isolation on an acrylamide gel. Ligation of all three fragments can only be achieved by joining the filled Hpall site with the EcoRI site, the two Hind 111 sites with each other, and the BglII site with the BamHI site. Thus, when the resulting plasmid is restriction treated with Clal and BamHI, a 470 bp fragment containing the SV40 origin is obtained. This fragment is inserted into the ClaI and BglII sites of pE342.HBV.b22 (Section A) to form pE342.HBV.E400.D22 (Fig. 3).

EXAMPLE 4

Transfection of host cells

The host cells used herein are vertebrate cells cultured in tissue culture. These cells, as known in the art, can be kept as permanent cell lines made by successive row transfers from isolated normal cells. These cell lines are maintained either on a solid support in liquid medium or by growth in suspensions containing supportive nutrients.

In the preferred embodiment, CHO cells which are deficient in DHFR activity are used.

These cells are prepared and propagated as described by Urlaub and Chasin, Proc. Natl. Acad. Sci. (USA) 5 ΎΤ_'Λ216 (1980), which is considered to be incorporated in this specification by this reference.

The cells are transfected with the 5 mg desired vector as prepared above using the method described by Graham and Van Der Eb, Virology 10 52: 456 (1978), which is considered to be incorporated in this specification by this reference.

The method ensures the effect of a collection of plasmids on a particular host cell, thereby increasing the probability that if a plasmid is absorbed by a cell, additional plasmids will also be absorbed. Accordingly, it is convenient to introduce both the primary and the secondary coding sequences using separate vectors for each as well as using a single vector containing both sequences.

EXAMPLE 5 Growth of transfected cells and expression of peptides The CHO cells subjected to transfection as above were first cultured for 2 days in non-selective medium, then the cells were transferred to medium lacking glycine, hypoxanthine and thymidine, thus selection for cells capable of expressing the plasmid DHFR. After about 1-2 weeks, individual colonies were isolated with cloning rings.

19 DK 173671 B1

Cells were plated in 60 or 100 mm tissue culture dishes at about 0.5 x 10 ° cells / dish. After two days of growth, the growth medium was changed. HBsAg was tested 24 hours later at RIA (Ausria II, Abbott).

Five cells were counted and HBsAg production standardized, on a per-cell basis. 10-20 random colonies were analyzed in this way for each vector used.

In an example of the practice of the invention, the following results were obtained: DK 173671 B1

2 G

.

O O -H O lA O

o <r iA 1 — t Λ o o cn in CO »- (

Ό I O Ή O 1Λ CO

V. o lA

U O <D lA rH

i ~ l O

ω ό o o »<o o

Li LA

NO El O <—I O O CO

O O O CM lA

r-10 0 "CD.-C

σ> c c>

H

»ΟΌ

C C O

O ID H

• ri I O rH O O f- '-U · <-! O IA r-i J * -H Ό U T> 0)

0 H

Ch C O

ion 1> —I I O Ή O O Ν 'D> O O LA Ή c r *.

to co o? o i — i o σ o X - o o

<-I

· * -> ^ 0) Li

-u ro • Η -H

> c

ri O.

i-} I-I

island

0) X

_____- u_ Λ Bi Lt Li

ID Dl DJ CM

C Ή Ή »

O ri H LA O O O O

• H (DO) fA il <t CM r-i

+ J O O Oi ^ IA iA

0 * I 0) CAO

Cl- Li- o CO -C r-i C O \ ro co U ti- 3 v— ro \ -3 ·

ON

uo cm

X CM

• Oh

CM »o- O DA CM CO UJ N <t CL Q Id <r + · * CM CN CM 0 =>

CM OO CM CD CD

Q 0C Q OO or

Li · · · · ·

O CM CM CM CM CM

-U <J- <t <fr <i

ri IA IA tA YES IA

ro ui uj uj lu uj> α cl α. α cl 21 DK 173671 B1

The production of surface antigen in several of the highest-expressing cell lines has been controlled for more than 20 passages and is stable. The cells expressing surface antigen remain indefinitely attached to the substrate and will continue to secrete the large amounts of surface antigen as long as the medium is replenished.

It is clear that the polycistronic gene construct results in isolation of the cells producing 10 the highest levels of HBsAg. 100¾ of the colonies transformed with pE342.H8V.D22 produced over 500 ng / 10 6 cells / day, while 92% of them remained. transformed with the non-polycistronic plasmid pE342.HBV.E400.D22 produced less than that amount.

Only cells from the polycistronic transfection exhibited production levels greater than 1500 ng / 10 6 cells / day.

EXAMPLE 6

Treatment with Methotrexet I 1 20 The surface antigen-expressing cell lines are inhibited by methotrexate (MTX), a specific inhibitor of DHFR at concentrations greater than 10 nM. In accordance with previous studies on the effects of MTX on tissue culture cells, clones that are resistant to higher concentrations (50 nm) of MTX occasionally occur at a frequency of about 10 ~ * \ However, these clones no longer produce surface antigen despite the amplification of HBV sequences in the MTX-resistant clones. Thus, the KBV gene is amplified even though the expression ice fades in this case. This suggests that further production of surface antigen may be fatal to the cell.

EXAMPLE 7

Extraction of the desired peptide

The surface antigen produced is in the form of a particle analogous to the 22 nm particle observed in the serum of patients infected with the virus. This form of antigen has been shown to be highly immunogenic. When the cells are grown in medium lacking calf serum or other supplements, about 10% of the protein contained in the medium is surface antigen and 10 -this protein can be isolated by well known methods in the art. The surface antigen migrates on SDS-polyacrylamide gels together with the protein derived from the 22 nm particle.

REFERENCES

1. Wigler, M. aK, Proc. Natl. Acad. Sci. 77: 3567 (1980) 2. Schimke, Robert T. et al., Science 202: 1051 (1978) 3. Biedler, J.L. et al., Cancer Res. 32: 153 (1972) 4. Chang, S.E. et al., Cell 7: 391 (1976) 5. Fischer, G.A., Biochem Pharmacol. 11: 1233 (1962) 20 6. Sirotnak, F.M. et al / L, Cancer Res. 28:75 (1968) 7. Flintoff, W.F. et al., Somat. Cell. Genet. 2: 245 (1976) 8. Christman, J. et al., Proc. Natl. Acad. Sci. 79: 1815 (1982) 9. Ringold, Gordon et al., J. Molec and Appl. Gen. 1: 165 (1981) 10. Kaufman, R.F. et al., J. Molec. Biol. 159: 601 (1982) 11. Perucho, Manuel et al., Cell 22: 309 (1980)

Claims (4)

23 DK 173671 B1 Patent claims: A method for selecting vertebrate cells that have been transfected with an expression vector capable of expressing a desired protein, characterized in that (a) the cells are treated with a vector containing coding sequences for both the desired protein and a secondary protein whose presence is required for the growth of the host cells under selective culture conditions, both coding sequences operably linked to the same promoter sequence but separated by translation stop and start codons, and (b) the cells being cultured under the selective culture conditions. . Process according to claim 1, characterized in that the secondary protein is DHFR. Process according to claim 3, characterized in that the selective culture conditions comprise a medium lacking glycine, hypoxanthine and thymidine. A method according to any one of the preceding claims for selection of vertebrate cells producing high levels of a desired heterologous protein, characterized in that (a) the cells are treated with a vector containing the coding sequences of a secondary protein whose presence serves as the selection marker for the transfected cells, below the coding sequence for the desired protein, both coding sequences operably linked to the same promoter sequence but separated by translation stop and start signals, and (b) the cells are grown under selective cultivation conditions. 5