Patent Grant
6902920
The invention concerns a method for producing a recombinant active heterodimeric AMV-RT by expressing one or several DNA sequences coding for the α- and/or β-subunit(s) of the AMV-RT in prokaryotic cells under certain growth and induction conditions.
The discovery of the reverse transcriptases in the seventies disproved the “central dogma” of molecular biology on the information transfer from DNA via RNA to protein as a unidirectional process (Termin H. and Mizutani S., 1970 Nature 226:1211-1213; Baltimore D., 1970, Nature 226:1209-1211). The enzymatic characterization of these RNA-dependent DNA polymerases is the basis for current understanding on the amplification cycle of RNA viruses and thus also on the development and spread of diseases that are caused by this type of virus (cancer, AIDS etc.).
However, reverse transcriptases are also a tool for molecular biologists for the synthesis, amplification and cloning of cDNAs (RT-PCR). This technology allows a simplified and accelerated examination of gene expression in eukaryotic cells. After isolating the total mRNA from cell extracts or tissues, the mRNA is translated back into cDNA by the reverse transcriptase and amplified by the subsequent PCR step to enable cloning and characterization. consequently it is not necessary to, on the one hand, elucidate the intron and exon structures of the genes but, on the other hand, it is also possible to examine gene expression in the cell during various life cycles or during the development of diseases (such as cancer).
Reverse transcriptases (RT) from three different retroviruses have hitherto been closely examined: The RT from Moloney Murine Leukemia Virus (M-MLV). This enzyme consists of a single subunit with a molecular weight of 78 kDa (Prasad V. R., 1993 reviewed in Reverse Transcriptase, Cold Spring Harbor, N.Y.: Cold Spring Harbor Laboratory Press, 135). In addition an RT from Human Immunodeficiency Virus (HIV) is known. This RT is a heterodimer that is composed of two subunits p66 and p51, the p51 subunit being formed by proteolytic cleavage of p66 (Le Grice S. F. J., 1993 reviewed in Reverse Transcriptase, Cold Spring Harbor, N.Y.: Cold Spring Harbor Laboratory Press, 163). In addition RTs are known from Avian Sarcoma-Leukosis Virus (ASLV). The RT obtainable from Avian Myeloblastosis Virus (AMV) also belongs to the ASLV family. This RT is also a heterodimer that is composed of an α-chain with a molecular weight of ca. 63 kDa and a β-chain with a molecular weight of ca. 95 kDa. In this case the α-chain is also formed by proteolytic processing of the β-chain (Golomb M. and Grandgenett D., 1979, J. Biol. Chem. 254: 1606-1613; Weiss R. et al., eds. 1984 , Molecular Biology of tumor viruses, 2nd edition: RNA tumor viruses 1/text. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.).
Whereas the M-MLV-RT is expressed in E. coli as a monomer and the HIV-RT as a heterodimer, it has so far not been possible to express the AMV-RT as an active or soluble heterodimer in E. coli or other prokaryotes to a satisfactory degree. Although according to WO 00/42199 certain RT variants are expressed in E. coli or preferably in eukaryotic insect cells, the desired RT that is obtained in this process mainly consists (ca. 90%) of an insoluble component.
In addition it is difficult to measure a recombinant AMV-RT in crude cell extracts of E. coli since, on the one hand, RNA templates are degraded by intrinsic E. coli RNases and, on the other hand, E. coli strains have a DNA polymerase which also has an RT activity in addition to the DNA polymerase activity (Ricchetti, M. and Huc, H., 1993, EMBO J. 12 (2), 387-396). Hence this intrinsic E. coli RT activity considerably interferes with the determination of the activity of the recombinant AMV-RT in crude E. coli extracts and in fractions from the purification.
Hence the object of the present invention is to provide a recombinant active heterodimeric AMV-RT in adequate amounts.
The object is achieved by a method for producing an active heterodimeric AMV-RT in prokaryotic host cells wherein one or several DNA sequence(s) which code for the α and β subunit or chain of the AMV-RT, are cloned into expression plasmids, the expression plasmids are transformed in prokaryotic cells, the expression of the heterodimeric AMV-RT is induced and the recombinant heterodimeric AMV-RT is purified i.e. isolated from the cells. Suitable genes and DNA sequences are, among others, those which only code for one of the AMV-RT subunits. A portion or the expression product can subsequently be converted by certain measures, such as proteolytic cleavage of the β-chain, into the other subunit. The sequences SEQ ID NO:4 and SEQ ID NO:5 have proven to be particularly suitable for the method according to the invention which generates an active heterodimeric AMV-RT composed of the subunits SEQ ID NO:6 and SEQ ID NO:7.
The structural genes and DNA sequences coding for the subunits of the AMV-RT can either be cloned on different, separate expression plasmids or on one expression plasmid, optionally in the presence of so-called helper plasmids, and expressed in a suitable host cell. Suitable expression plasmids are for example pDS, pKK177-3 or pKKT5. The plasmid pKKT5 in which the respective structural genes are inserted under the control of the T5 promoter is preferred according to the invention. Other potential promoters, which are preferably IPTG-inducible promoters, are for example the lac, lac UV5 or tac promoter. Alternative helper plasmids such as the plasmid pUBS520 and suitable selection markers such as ampicillin or kanamycin are in principle known to a person skilled in the art.
The expression plasmids and optionally other helper plasmids are transformed into a suitable prokaryotic host cell. According to the invention it is preferable to use an E. coli strain such as E. coli K12 C600, DH5α, LE392, JM83, JM105, NM522, M15, RR1 Δ15, UT5600, TG1, A1200 or the strains E. coli B, BL21, HB101. The E. coli strain LE392 is particularly preferred according to the invention.
The expression of the heterodimeric AMV-RT can be induced by various measures. In particular certain growth and induction conditions have positive effects on the expression of active AMV-RT. A growth temperature in the range of 10° to 25° C. combined with a low inducer concentration has proven to be advantageous according to the invention. A growth temperature of about 15° C. and an inducer concentration between 0.1 and 0.5 mM, preferably of about 0.15 mM, have proven to be particularly suitable. IPTG (isopropyl-β-D-thiogalactopyranoside) or lactose are preferably used according to the invention as the inducer.
Furthermore it turned out that the soluble expression of AMV-RT in prokaryotic cells can be increased by the co-expression of helper genes. Potential helper genes are in particular the trpT gene which codes for the tryptophan tNRA. In addition chaperone genes are suitable for soluble expression such as the genes coding for GroEL and GroES, GrpE, ClpB, Dnak and DnaJ. The genes for one or several chaperones are then preferably located on a helper plasmid with an inducible promoter; the genes which code for the chaperones GroEL and GroES are under the control of a constitutive promoter on the expression plasmid on which the structural genes for the α and/or β chain are also located. However, it is particularly preferred according to the invention when the genes coding for GroEL and GroES are cloned on the expression plasmid which carries the genes for the α-and β-chain and the genes coding for Dnak, DnaJ, GrpE and ClpB are cloned on a helper plasmid.
In addition to methods that are generally known to a person skilled in the art, it is especially advantageous to use affinity chromatography materials such as metal ion chelating materials or cation exchangers to purify and isolate the recombinant heterodimeric AMV-RT from the cell extract. It is particularly advantageous for the purification of the AMV-RT for the expression products, i.e. the α- as well as the β-chain to be fused with peptide sequences that are able to reversibly bind to particular column materials such as cation exchangers, metal ion chelating materials such as nickel, copper or zinc nitriloacetic acid (NTA) resins. Peptide sequences that are suitable according to the invention can have from two to about 100 amino acids or amino acid derivatives. Peptide sequences which are composed of two to ten amino acids, e.g. arginine residues or histidine residues, have proven to be particularly suitable for the invention. In addition it has also proven to be particularly advantageous to use such peptide sequences comprising eight arginine or six histidine residues. In addition commercially available peptide sequences such as Strep-tag® (IBA GmbH, Göttingen/Germany) or GST-tag (Pharmacia, Uppsala/Sweden) are also suitable for the method according to the invention.
The invention is further elucidated by the following examples:
The data bank sequence (MEDLINE ID 94366722, Baluda et al., 1994) was used to design oligonucleotide primers for the isolation of the β-chain (see SEQ ID NO:1 and 2). A EcoRI restriction endonuclease cleavage site was incorporated at the 5′ end and a PstI restriction cleavage site was incorporated at the 3′ end for the subsequent cloning into vectors. In addition a further 3′ primer was designed (see SEQ ID NO:3) which enables the isolation of the α-chain. Both chains were fished by means of PCR from a virus lysate (ATCC VR-265) by means of RT-PCR as well as from an E. coli clone (ATCC 31990) which carries the β-chain on a plasmid. The PCR mixtures were applied to a 1% agarose gel, the PCR fragments of ca. 1715 bp for the α-chain and ca. 2570 bp for the β-chain were isolated from the agarose gel (QIAEX II, Gel Extraction Kit, Qiagen/Germany), cleaved with the restriction endonucleases mentioned above and cloned into a vector fragment of pUC19 that had also been linearized with EcoRI and PstI and isolated. For this 1 μl (20 ng) vector fragment and 3 μl (100 ng) PCR fragment, 1 μl 10×ligase buffer (Maniatis et al., 1989 Molecular Cloning: A Laboratory Manual, second Edition, Cold Spring Harbor Laboratory Press NY (USA.), vol. 27), 1 μl T4 DNA ligase, 4 μl sterile H2Obidistilled were pipetted, carefully mixed and incubated overnight at 16° C. The cloned genes were subsequently examined by means of restriction analysis and sequencing. The sequences are shown in SEQ ID NO:4 (α-chain) and SEQ ID NO: 5 (β-chain).
Comparison with the data bank sequence (MEDLINE ID 94366722, Baluda, M. A., and Reddy, E. P., 1994, Oncogene 9:2761-2774) yielded a homology of 98.8% at the DNA level for the α-chain as well as for the β-chain. When the resulting amino acid sequences are compared, it becomes apparent that most substitutions at the DNA level are so-called silent mutations i.e. do not lead to amino acid substitutions. Only three base substitutions also resulted in amino acid substitutions but they are found reproducibly in each isolated PCR product. These are the substitutions Arg273Met, Arg304Gln and Asp495Glu. The amino acid sequences of both chains are shown in SEQ ID NO:6 (α-chain) and SEQ ID NO:7 (β-chain).
2.1. Construction of the Expression Plasmids pAMV-α and pAMV-β
In order to express the AMV-RT, the genes for both chains were cloned separately into expression vectors in such a manner that the structural genes were each inserted in the correct orientation under the control of the T5 promoter. For this the respective structural gene for the α-chain and the β-chain were cut out of the plasmid pUC19 by EcoRI and PstI, the restriction mixtures were separated by agarose gel electrophoresis and the 1715 bp fragment of the α-chain and the 2570 bp fragment of the β-chain were isolated from the agarose gel. The expression plasmid pKKT5, which is formed from pKK177-3 (Kopetzki et al., 1989, Mol. Gen. Genet. 216: 149-155) by replacing the tac promoter with the T5 promoter from pDS (Bujard et al., 1987, Methods Enzymol. 155: 416-433), was used for the expression. The EcoRI restriction endonuclease cleavage site in the sequence of the T5 promoter was removed by two point mutations. The resulting expression plasmid was cut with EcoRI and PstI for the insertion of the genes for the AMV-RT, the restriction mixture was separated by agarose gel electrophoresis and the resulting vector fragment of ca. 2500 bp was isolated from the agarose gel. The vector fragment obtained in this manner was separately ligated as described above with the genes for the α-chain and the β-chain described in example 1. The correct insertion of the genes was checked by restriction control and sequencing. The resulting plasmid pAMV-α and pAMV-β was firstly separately cotransformed with the helper plasmid pUBS520 for expression control in various E. coli strains. It is conceivable in this case that the α-chain and the β-chain could be separately expressed in order to obtain αα- and ββ-homodimers. The helper plasmid pUBS520 (Brinkmann et al., 1989, Gene 85: 109-114) carries inter alia the lacIq gene which codes for the lac repressor and the dnaY gene which codes for the rare tRNAArg in E. coli which recognizes the codons AGA and AGG (Garcia et al., 1986, Cell 45: 453-459). The kanamycin resistance gene from the transposon TN903 was used as a selection marker.
2.2 Separate Transformation of the Expression Plasmids pAMV-α and pAMV-β in E. coli
Competent cells of various E. coli strains were prepared according to the method of Hanahan (J. Mol. Biol. 1983, vol. 166, 557). 200 μl of E. coli LE392 cells prepared in this manner were admixed with 20 ng isolated expression plasmid pAMV-α DNA or pAMV-β DNA and 40 ng helper plasmid DNA. After 30 min incubation on ice a heat shock (90 sec. at 42° C.) was carried out. Subsequently the cells were transferred to 1 ml LB medium and incubated for 1 hour at 37° C. in LB medium for the phenotypic expression. Aliquots of this transformation mixture were plated out on LB plates containing ampicillin and kanamycin as selection markers and incubated for 15 hours at 37° C.
2.3 Expression of the Gene for the α-chain in E. coli
In order to express the gene which codes for the α-chain of the AMV-RT, plasmid-containing clones were inoculated in 3 ml LBampkan medium and incubated at 30° C. in a shaker. At an optical density of 0.5 (measured at 550 nm, OD550) the cells were induced with 0.5 mM IPTG and incubated for 4 h at 30° C. in a shaker. Subsequently the optical density of the individual expression clones was determined, an aliquot which corresponded to an OD550 of 5.0/ml was removed and the cells were centrifuged (10 min, 6000 rpm, 4° C.). The cell pellet was resuspended in 400 μl TE buffer (50 mM TRIS/50 mM EDTA, pH 8.0), the cells were disrupted by ultrasound and the soluble protein fraction was separated from the insoluble protein fraction by centrifugation (10 min, 14000 rpm, 4° C.). Application buffer containing SDS and β-mercaptoethanol was added to all fractions and the proteins were denatured by boiling (5 min 100° C.). Afterwards 10 μl of each was analysed by means of an analytical SDS gel (10%) (Laemmli U. K., 1970, Nature 227: 555-557).
Analysis of the SDS gel shows a clear overexpression of the α-chain. A strongly overexpressed additional band is seen at ca. 63 kDa which is not observed with the non-induced control clones or the induced control cloneswhich do not contain plasmid. A small portion of the overexpressed α-chain appears in the soluble protein fraction whereas the major amount is formed as an insoluble expressed protein.
2.4 Expression of the Gene of the β-chain in E. coli
In order to express the gene which codes for the β-chain of the AMV-RT, 3 ml LBampkan medium was inoculated with plasmid-containing clones and incubated at 30° C. in a shaker. At an OD550 nm of 0.5 the cells were induced with 0.5 mM IPTG and incubated for 4 h at 30° C. in a shaker. Subsequently the optical density of the individual expression clones was determined, an aliquot which corresponded to an OD550 of 5.0/ml was removed and the cells were centrifuged (10 min, 6000 rpm, 4° C.). The cell pellet was resuspended in 400 μl TE buffer (50 mM TRIS/50 mM EDTA, pH 8.0), the cells were disrupted by ultrasound and the soluble protein fraction was separated from the insoluble protein fraction by centrifugation (10 min, 14000 rpm, 4° C.). Application buffer containing SDS and β-mercaptoethanol was added to all fractions and the proteins were denatured by boiling (5 min 100° C.). Afterwards 10 μl of each was analysed by means of an analytical SDS gel (8%) (Laemmli U. K., 1970, Nature 227: 555-557).
Analysis of the SDS gel shows a clear overexpression of the β-chain. A strongly overexpressed additional band is seen at ca. 95 kDa which is not observed with the non-induced control clones or the induced control clones which do not contain plasmid. The majority of the overexpressed β-chain appears in the insoluble protein fraction, however, a slight overexpression is also seen in the soluble protein fraction.
2.5 Expression of Both Chains on Separate Plasmids in a Cell
In order to express both chains in one cell, the lacIq expression cassette and the dnaY expression cassette must firstly be recloned from the helper plasmid pUBS520 onto the expression plasmids. The lacIq expression cassette was cloned onto pAMV-α and the dnaY expression cassette was cloned onto the expression plasmid pAMV-β. In order to ensure a stable multiplication of the expression plasmids, the ampicillin resistance gene from pAMV-α was replaced by the kanamycin resistance gene from pUBS520. The resulting expression plasmids pAMV-αlacIq and pAMV-βdnaY were subsequently cotransformed in various E. coli expression strains.
In order to express the genes which code for the α-chain and the β-chain of the AMV-RT, 3 ml LBampkan medium was inoculated with plasmid-containing clones and incubated at 30° C. in a shaker. At an OD550 nm of 0.5 the cells were induced with 0.5 mM IPTG and incubated for 4 h at 30° C. in a shaker. Subsequently the optical density of the individual expression clones was determined, an aliquot which corresponded to an OD550 nm of 5.0/ml was removed and the cells were centrifuged (10 min, 6000 rpm, 4° C.). The cell pellet was resuspended in 400 μl TE buffer (50 mM TRIS/50 mM EDTA, pH 8.0), the cells were disrupted by ultrasound and the soluble protein fraction was separated from the insoluble protein fraction by centrifugation (10 min, 14000 rpm, 4° C.). Application buffer containing SDS and β-mercaptoethanol was added to all fractions and the proteins were denatured by boiling (5 min 100° C.). Afterwards 10 μl of each was analysed by means of an analytical SDS gel (8%) (Laemmli U. K., 1970, Nature 227: 555-557).
Analysis of the SDS gel surprisingly shows a clear overexpression of the α- and β-chain. Strongly overexpressed additional bands are seen at ca. 63 kDa and ca. 95 kDa which are not observed with the non-induced control clones or the induced control clones which do not contain plasmid. The distribution of the bands in the soluble and insoluble fraction is like that of the experiments in which both chains were expressed separately. The expression output of both chains is overall somewhat less than for separate expression.
3.1 Production of Various Fusion Proteins
In order to efficiently purify the recombinant AMV-RT heterodimers, suitable peptide sequences, so-called tags were fused to the 5′ end of both chains. Tags enable affinity chromatographies to be carried out. A series of two affinity chromatographies which are each specific for one of the two tags additionally allows the isolation of pure heterodimers (Wende W. et al., 1996, Biol. Chem. 377, 625-632). Appropriate primer designs were used to attach eight arginine residues to the α-chain and six histidine residues to the β-chain by means of PCR reactions. The sequences of the sense primers are shown in SEQ ID NO:8 (5′ primer for the α-chain) and SEQ ID NO:9 (5′ primer for the β-chain). The oligonucleotides of SEQ ID NO:2 (β-chain) and SEQ ID NO:3 (α-chain) which had already been used for gene isolation were used as antisense primers.
The PCR mixtures were applied to a 1% agarose gel, the PCR fragments of 1739 bp for the α-chain and 2597 bp for the β-chain were isolated from the agarose gel (QIAEX II, Gel Extraction Kit, Qiagen, Germany), cleaved with the restriction endonucleases EcoRI and PstI and cloned into a vector fragment of the preferred expression plasmid that had also been linearized with EcoRI and PstI and isolated. For this 1 μl (20 ng) vector fragment and 3 μl (100 ng) PCR fragment, 1 μl 10×ligase buffer (Maniatis et al., 1989 Molecular Cloning: A Laboratory Manual, second Edition, Cold Spring Harbor Laboratory Press NY (USA), vol. 27), 1 μl T4 DNA ligase, 4 μl sterile H2Obidistilled were pipetted, carefully mixed and incubated overnight at 16° C. The cloned genes were subsequently examined by means of restriction analysis and sequencing. The resulting Expression Plasmids were named pAMV-αlacIq-Arg and pAMV-βdnaY-His.
3.2 Transformation of the expression plasmids pAMV-αlacIq-Arg and pAMV-βdnaY-His in Various E. coli Expression Strains
Competent cells of various E. coil strains were prepared according to the method of Hanahan (J. Mol. Biol. 1983, vol. 166 pp. 557) (see example 2.2).
3.3 Expression of Both Chains with Fused Tags on Separate Plasmids in a Cell
In order to express both chains with tags in a cell, various E. coli expression strains were cotransformed with the expression plasmids pAMV-αlacIq-Arg and pAMV-βdnaY-His.
In order to express the genes which code for the α-chain with an Arg-tag and the β-chain with an His-tag of the AMV-RT, 3 ml LBampkan medium was inoculated with plasmid-containing clones and incubated at 30° C. in a shaker. At an OD550 of 0.5 the cells were induced with 0.5 mM IPTG and incubated for 4 h at 30° C. in a shaker. Subsequently the optical density of the individual expression clones was determined, an aliquot which corresponded to an OD550 of 5/ml was removed and the cells were centrifuged (10 min, 6000 rpm, 4° C.). The cell pellet was resuspended in 400 μl TE buffer (50 mM TRIS/50 mM EDTA, pH 8.0), the cells were disrupted by ultrasound and the soluble protein fraction was separated from the insoluble protein fraction by centrifugation (10 min, 14000 rpm, 4° C.). Application buffer containing SDS and β-mercaptoethanol was added to all fractions and the proteins were denatured by boiling (5 min 100° C.). Afterwards 10 μl of each was analysed by means of an 8% analytical SDS gel (Laemmli U. K., 1970, Nature 227: 555-557).
Analysis of the SDS gel surprisingly shows a clear overexpression of the α- and β-chain. Strongly overexpressed additional bands are seen at ca. 63 kDa and ca. 95 kDa which are not observed with the non-induced control clones or the induced control clones which do not contain plasmid. The distribution of the bands in the soluble and insoluble fraction is like that of the experiments in which both chains were expressed separately without tags in one cell.
3.4 Expression of Both Chains with Fused Tags on a Plasmid
If the genes for the α- and β-chain of the AMV-RT are distributed on two plasmids, differences in the stability of these plasmids could lead to the production of different amounts of the respective chains and thus to a lower yield of αβ-chain heterodimer. Hence with the exception of the gene for β lactamase, the entire genetic information of the two plasmids pAMV- αlacIq-Arg and pAMV-βdnaY-His was combined on a single plasmid pAMV-αβ-1. This plasmid was constructed by inserting the SspI-AflIII fragment of pAMV-βdnaY-His containing the sequence for the T5 promoter, the gene coding for the β-chain with an N-terminal His tag, the sequence for the rrnB terminator and the dnaY gene, into the SalI cleavage site of pAMV-αlacIq-Arg which contains the sequence for the T5 promoter, the gene coding for the α-chain with an N-terminal Arg-tag, the sequence for the rrnB terminator, the kanamycin resistance gene and the lacIq gene. For this purpose 1 μg each of the expression plasmids pAMV-αlacIq-Arg and pAMV-βdnaY-His were cleaved with the restriction endonucleases described above according to the manufacturer's instructions, the restriction mixtures were separated in a 1% agarose gel and the 4124 bp SspI-AflIII fragment of pAMV-βdnaY-His and the 6024 bp fragment of pAMV-αlacIq-Arg were isolated from the agarose gel (QIAEX II, Gel Extraction Kit, Qiagen/Germany). The non-compatible ends were prepared with Klenow polymerase (Roche Diagnostics GmbH) according to the manufacturer's instructions and the two fragments were ligated together as described above. The resulting new expression plasmid pAMVαβ-1 was examined by means of restriction analysis.
The correct expression plasmid according to restriction analysis was transformed in the E. coli K-12 strain LE392 as described above and subjected to an expression control. The protein content of the cells after 4 hours growth under induced conditions was subsequently examined by means of SDS-PAGE. According to SDS-PAGE analysis the level of the expression output and the relative proportion of soluble and insoluble fractions are comparable to the expression of the genes for the α- and β-chain on separate plasmids, but the amount of expressed α- and β-chain appears to be more homogeneous.
Furthermore the Arg-tag of the α-chain was replaced by a His-tag like that of the β-chain for the purification procedure. For this purpose an intermediate construct pAMV-αlacIq-His was prepared in which the EcoRI-NheI fragment from pAMV-αlacIq-Arg was replaced by the EcoRI-NheI fragment from pAMV-βdnaY-His. Subsequently, like the construction of pAMVαβ-1, the entire genetic information of the two plasmids pAMV-αlacIq-His and pAMV-βdnaY-His with the exception of the gene for β lactamase was combined on a single plasmid pAMVαβ-2. The new expression vector was named pAMVαβ-2. Cells were transformed as described above with pAMVαβ-2and subjected to an expression control under standard conditions. The expression output was not increased under these conditions.
4.1 Increase of the Expression of Active AMV-RT by Changing the Expression Conditions
Particular growth and induction conditions have positive effects on the expression of active AMV-RT. Afterwards the growth temperature was lowered from 30° C. to 15° C. during the induction phase, the IPTG concentration was reduced from 0.5 mM to 0.15 mM to induce expression and the induction time was increased from 4 h to 26 h. The protein content of the cells after the induction phase was examined as described above by SDS polyacrylamide gel electrophoresis.
Afterwards the total expression yield of α- and β-chain was, as expected, substantially reduced in the SDS-PAGE analysis, but the content of soluble expressed α- and β-chain was considerably increased in comparison to the expression experiments under standard growth and induction conditions. This increase in the expression of active AMV-RT was also confirmed in the subsequent purification and activity determination.
4.2 Increasing the Expression of Active AMV-RT by Coexpression of Helper Genes
4.2.1. Coexpression of the Gene for the Tryptophan-tRNA (tRNAtrp)
One property of the AMV-RT is to use an endogenous cell tRNA for tryptophan (tRNAtrp) as a primer for the polymerase reaction after infection of a eukaryotic host cell (Leis et al., 1993, in: Reverse Transcriptase, Cold Spring Harbor Monograph Series, eds.: Skala, A. M. and Goff, S. P., Cold Spring Harbor, N.Y. (USA), 33-48). However, whether the endogenous E. coli tRNAtrp can be used by the AMV-RT as a primer has not been proven. In E. coli the tRNAtrp is only coded by a single gene trpT, the expression of which is adapted to the normal requirements of the cell. In order to exclude a potential deficiency of tRNAtrp in the cell, the trpT gene according to SEQ ID NO:10 was isolated by means of PCR from E. coli LE392 (the primers used for this are shown in SEQ ID NO:11 and 12), recleaved with PstI for insertion into pAMV-αlacIq-His and ligated into the vector fragment of pAMV-αlacIq-His that was also linearized with PstI as described above. Clones which have integrated the trpT gene at the PstI restriction endonuclease cleavage site were checked by means of restriction analysis and sequencing. In this intermediate construct pAMV-αlacIq-His-trpT the gene for the α-chain and the gene for the E. coli tRNAtrp form one transcription unit, the expression of which is regulated by the IPTG-inducible T5 promoter. Subsequently, similarly to the construction of pAMVαβ-1 or pAMVαβ-2, the entire genetic information of the two plasmids pAMV-αlacIq-His-trpT and pAMV-βdnaY-His with the exception of the gene for β-lactamase was combined on a single plasmid pAMVαβ-3. Cells were transformed as described above with pAMVαβ-3 and subjected to an expression control using the modified expression conditions. Afterwards the yield of active AMV-RT is significantly increased.
4.2.2. Coexpression of Chaperone Genes
In E. coli there are two main chaperone systems comprising the GroESL machinery and a 4 component system consisting of DnaK, DnaJ, GrpE and ClpB (Kedzierska, 1999). Both systems play an important role in the correct folding of newly formed proteins as well as in the renaturing of proteins that have aggregated as a result of stress (Hartl F. U., 1996, Nature 381, 571-580; Bukau H. and Horwich A. L., 1998, Cell 92, 351-366; MogK A. et al., EMBO J. 18, 6934-6949; Zolkiewski M., 1999, J. Biol. Chem. 274, 28083-28086; Goloubinoff P. et al., 1999, Proc. Natl. Acad. Sci. USA 96, 13732-13737).
In a first step the groESL operon from E. coli should be overexpressed in the AMV-RT production strains. For this the EcoRI-HindIII fragment from pOF39 (Fayet O., Louarn J.-M., Georgopoulos C., 1986, Mol. Gen. Genet. vol. 202, pp. 335-345 was integrated in the SspI cleavage site of the plasmid pAMV-βdnaY-His. Non-compatible ends were prepared with Klenow polymerase (Roche Diagnostics) according to the manufacturer's instructions before ligation. The sequence of groESL is shown in SEQ ID NO:13. In this new construct pAMV-βdnaY-His-groESL, the groESL operon forms an artificial transcription unit containing the 3′-situated gene for β lactamase. The expression is then either under the control of the endogenous bla constitutive promoter that is now on the 5′ side of the groESL operon and/or under the control of the σ32-dependent promoter of the groESL operon. Subsequently the entire genetic information of the two expression plasmids pAMV-αlacIq-His-trpT and pAMV-βdnaY-His-groESL with the exception of the gene for β-lactamase was again combined as described above on a single plasmid pAMVαβ-4.
Cells were transformed with pAMVαβ-4 as described above and subjected to an expression control under the modified expression conditions. The co-overproduction of GroESL results in an increase of the biomass and of the amount of active AMV-RT. Three to four-fold higher values were obtained compared to the previously best production strains after purification and activity testing.
After the co-overproduction of GroESL in the AMV-RT production strains had proven to be a positive measure, the other main chaperone system of E. coli was additionally co-overproduced in a second step. In addition to the supposed general advantages of this co-overproduction this could compensate for a disadvantage of the GroESL machinery i.e. its exclusion volume of circa 65 kDa (Deuerling E. et al., 1999, Nature 400, 693-696). This should be particularly important for the correct folding of the β-chain of the AMV-RT (93 kDa) provided it cannot be divided into single domains that are formed independently of one another. The genes DnaK, DnaJ and GrpE were combined in an artificial operon corresponding to the physiological combination (Diamant S. and Goloubinoff P., 1998, Biochemistry 37, 9688-9694; Pierpaoli E. V. et al., 1998, J. Biol. Chem. 273, 6643-6649), whereas the gene for ClpB forms its own transcription unit. Both transcription units were placed under the control of IPTG-inducible T5 promoters in order to coordinate the expression with the genes for the subunits of the AMV-RT.
For technical reasons the cloning process required a number of intermediate steps on the path to the final construct pCHAP-5. Thus the pKKT5 derivatives pCHAP-1 and pCHAP-2 were firstly constructed. pCHAP-1 contains the genetic information for the dnaKJ operon from E. coli starting with the start codon for dnaK up to the stop codon for dnaJ; pCHAP-2 carries the artificial transcription unit from the coding regions of the genes for GrpE and ClpB as an insert; the corresponding DNA fragments were amplified by PCR from the genomic DNA of E. coli K12KE392. The sequence of the dnaKJ operon is shown in SEQ ID NO:14, the corresponding primers used to isolate the dnaKJ operon are shown in SEQ ID NO:15 and 16. The sequence of the grpE gene is shown in SEQ ID NO:17, the corresponding primers for the isolation of the grpE gene are shown in SEQ ID NO:18 and 19. The sequence of the clpB gene is shown in SEQ ID NO:20, the corresponding primers for the isolation of the clpB gene are shown in SEQ ID NO:21 and 22. In order to construct pCHAP-1 the PCR fragment containing the dnaKJ operon was recleaved with SmaI and BamHI and, as described above, ligated into a vector fragment of pKKT5 which had also been linearized with SmaI and BamHI. pCHAP-2 was constructed by means of a three-fold ligation with the EcoRI-PstI fragment of the grpE gene, the PstI-HindIII fragment of the clpB-gene and a vector fragment of pKKT5 linearized with EcoRI and HindIII. p-CHAP-3 in which the clpB gene is present alone as a transcription unit, is derived from pCHAP-2 by ligating the PstI-HindIII fragment from pCHAP-2 into the vector fragment of pKKT5 linearized with EcoRI and HindIII as described above. Before the ligation reaction the non-compatible ends of the two fragments were prepared with Klenow polymerase (Roche Diagnostics) according to the manufacturer's instructions. pCHAP-4 is a pCHAP-1 derivative whose insert was extended by the grpE gene from pCHAP-2 and thus the artificial transcription unit comprises the genes for DnaK, DnaJ and GrpE. As a result of the Shine Dalgarno sequence which is suboptimal in this case, the expression of grpE should be reduced compared to pCHAP-2 and thus be better adapted to the expression of dnaKJ (Diamant & Goloubinoff, 1998; Pierpaoli et al., 1998). In order to construct pCHAP-4 the EcoRI-AvaI fragment from pCHAP-2 was inserted into the BamHI cleavage site of pCHAP-1 after the non-compatible ends of the two fragments had been prepared with Klenow polymerase (Roche Diagnostics) according to the manufacturer's instructions. The final construct p-CHAP-5 is a pCHAP-4 derivative which contains the insert of pCHAP-3 as additional genetic information. For this the BspLU11I-NdeI fragment in pCHAP-4 was replaced by the BspLU11I-SspI fragment from pCHAP-3 by restriction and ligation as already described several times. In order to ensure the compatibility of the ends, the overhanging ends generated by NdeI were previously filled in with Klenow polymerase (Roche Diagnostics) according to the manufacturer's instructions.
The effect of combining the expression plasmid pAMVαβ-4 with the various helper plasmids pCHAP-1 to 5 on the overproduction of active AMV-RT was examined. At least under the modified standard expression conditions all helper plasmids considerably increased the previous yields of active AMV-RT and as expected the helper plasmid pCHAP-5 gave the best result. This was confirmed by SDS-PAGE as well as by subsequent purification and activity determination.
5.1. Test for Reverse Transcriptase Activity (Test A)
During the purification, the reverse transcriptase activity in the fractions was detected by means of a non-radioactive test system. The “reverse transcriptase assay non-radioactive” (Roche Molecular Biochemicals, cat. No. 1468120) was used for this. The incubation period was shortened to 30 minutes.
5.2. Test for Reverse Transcriptase Activity (Test B)
The specific reverse transcriptase activity of the pools was determined by a radioactive test system. Reverse transcriptase activity was determined in a test volume of 100 μl (50 mM Tris/HCl, pH 8.3 (37° C.), 40 mM KCl, 6 mM MgCl2, 0.5 mM dTTP, 0.04 OD260 nm poly (A)×dT15, 0.1 μM [3H]-dTTP). AMV-RT (5 μl) was added in suitable dilutions. After incubating for 10 min at 37° C., the reaction was stopped with 10% TCA solution (500 μl). The radioactively-labelled product that formed was washed on a nitrocellulose filter after precipitation. The incorporation rate of radioactivity was measured in a scintillation counter and the RT activity of the sample was calculated. One enzyme unit was defined as the amount of AMV-RT which incorporated 1.0 nMol TMP into acid insoluble product in 10 min at 37° C.
5.3. Test for DNA Polymerase
The activity of DNA polymerase from E. coli was determined by measuring the nick translation. The DNA polymerase was detected by means of a non-radioactive nick translation test. The nick translation was carried out in a test volume of 50 μl (50 mM Tris/HCl, pH 7.5, 10 mM MgCl2, 0.1 mM DTE, 28.875 μM DIG-dUTP, 1.444 μM bio-16-dUTP, 95.865 μM dTTP, 20 μM dATP, 20 μM dCTP, 20 μM dGTP, 1 μg pBR322, 1 pg DNaseI). After adding the samples (1 μl) the reaction mixture was incubated for 30 min at 37° C. Afterwards the reaction mixture was transferred to streptavidin-coated microtitre plates. Subsequent treatment and evaluation of the test was carried out analogously to the “reverse transcriptase assay, non-radioactive” (Roche Molecular Biochemicals, Cat. No. 1468120).
5.4 Test for Contaminating Activities
The test for the presence of contaminating foreign activities was carried out in a solution composed of 10 mM Tris/HCl, pH 7.5, 10 mM MgCl2, 1 mM DTE.
Suitable samples of the individual enzyme fractions were incubated with the corresponding nucleic acids. So-called nicking activity was detected by incubation with the plasmid pBR322 (1 μg) for 2-16 hours at 37° C. Unspecific nucleases were detected by incubation with lambda-DNA/EcoRI, HindIII (1 μg) for 2-16 hours at 37° C. Unspecific RNases were detected by incubation for 2-4 hours at 37° C. with MSII-RNA (5 μg).
For the test for contamination by exonucleases, the samples were incubated for 4 hours at 37° C. with 4 μg [3H]-labelled DNA and afterwards the released [3H]-labelled nucleotides were determined.
6.1. AMV-RT from E. coli LE392 pAMV-αlacIq-Arg/pAMV-βdnaY-His Construct
6.1.1. Purification
E. coli cells which overexpress both chains of the AMV-RT (see above) were used as the starting material to purify the recombinant AMV-RT.
The AMV-RT was purified at 4° C. The purification was carried out by chromatographic methods after cell lysis and separation of the nucleic acids. The purification process yields a recombinant AMV-RT which is free of contaminating enzyme activities and in RT-PCR has the same functionality as an AMV-RT purified from native material.
Buffer
200 ml buffer A was added to ca. 50 g E. coli LE392 cells (pAMV-αlacIq-Arg/pAMV-βdnaY-His) which were thawed and suspended. Two tablets of Complete (Roche Molecular Biochemicals, cat. No. 1697498) were added to the suspension. Subsequently the cells were lysed by means of ultrasound (Branson sonicator) while cooling (temperature: <10° C.). The degree of lysis of the cell suspension that was achieved was typically 40-50%.
Precipitation of Nucleic Acids
Afterwards the nucleic acids were removed by means of polymin precipitation. 5 ml of a 10% polymin P solution was added dropwise. If the precipitation was incomplete, further dropwise addition was carried out. After incubation for 30 min at 4° C., centrifugation was carried out (30 min, 13000 rpm, 4° C.).
Chromatographic Purifications
Affinity chromatography on a Ni-chelate column: The clear centrifugation supernatant was diluted with buffer B (1+1) and absorbed to a nickel-loaded chelating sepharose ff column (2.6 cm×10 cm, Pharmacia) which had been equilibrated with buffer B, it was then washed with ca. 500 ml buffer B, afterwards with 200 ml buffer B+10 mM imidazole and 200 ml buffer B+20 mM imidazole. The enzyme was eluted with a linear gradient of buffer B+20 mM imidazole and buffer C in a total volume of 500 ml. The flow rate was 5 ml per minute, the fraction size was 20 ml per fraction. The enzyme eluted between 50 mM and 200 mM imidazole. The pool of active fractions was dialysed against buffer D.
Chromatography on Heparin-sepharose
The dialysed pool was subsequently absorbed to a heparin-sepharose ff column equilibrated with buffer D (1.6 cm×10 cm, Pharmacia) and washed with ca. 200 ml buffer D, then with ca. 200 ml buffer D+300 mM KCl. The enzyme was eluted with a linear gradient of buffer D+300 mM KCl and buffer D+1 M KCl in a total volume of 200 ml. The flow rate was 2.5 ml per min, the fraction size was 10 ml. The AMV-RT eluted at a KCl concentration of 500 mM to 700 mM.
Chromatography on S-sepharose ff
The RT-active fractions were pooled and dialysed against buffer E. The dialysate was loaded onto a S-sepharose ff column equilibrated with buffer E (1.6 cm×10 cm, Pharmacia). After washing with ca. 200 ml buffer E, the enzyme was eluted with a linear gradient of buffer E and buffer E+1 M KCl in a total volume of 400 ml. The flow rate was 2.5 ml per minute, the fraction size was 10 ml.
Chromatography on Hydroxylapatite
The RT-active fractions were pooled and dialysed against buffer E. The dialysate was loaded onto a HA-ultrogel column equilibrated with buffer E (1.6 cm×10 cm, Biosepra). After washing with ca. 200 ml buffer E, the enzyme was eluted with a linear gradient of buffer E and buffer E+0.5 M potassium phosphate in a total volume of 400 ml. The flow rate was 2.5 ml per minute, the fraction size was 10 ml.
The RT-active fractions were pooled and dialysed against storage buffer. Application buffer containing SDS and β-mercaptoethanol was added to the purified protein and the sample was denatured by boiling (5 min, 100° C.). Subsequently 20 μl aliquots were analysed by an analytical SDS gel (4-20%) (Laemmli U K., 1970, Nature 227: 555-557). The α- and β-subunits of AMV-RT were found in equimolar ratios.
The described method yields a stable AMV-RT with an equimolar distribution of the α- and β-subunits. The enzyme obtained is functional in RT-PCR.
6.1.2. Functional Test in RT-PCR
The recombinant AMV reverse transcriptase that was obtained was examined in a functional test. The functional test consisted of a reverse transcription (RT) coupled with a polymerase chain reaction (PCR). For this 5 units of the recombinant AMV reverse transcriptase was used like the enzyme mixture of the Titan TM One Tube PCR System (cat. No. 1888382, Roche Molecular Biochemicals). A 1.8 kb fragment of the human dystrophin gene was amplified. 10 ng human muscle RNA was used as a template. The primers (400 nM) were the Dys primer 2reV (5′GAG TGA ATA CAG TTT GCC CAT GGA TTG-3) and the Dys primer 8for (5′-AAG AAG TAG AGG ACT GTT ATG AAA GAG AAG-3′). The target was amplified in a RT-PCR using the following program: 50° C. for 30 min, 94° C. for 2 min followed by 10 cycles (94° C. for 10 sec, 58° C. for 30 sec, 68° C. for 1 min 10 sec) and 20 cycles (94° C. for 10 sec, 58° C. for 30 sec, 68° C. for 1 min 10 sec; +10 sec/cycle). Subsequently it was incubated for 7 min at 68° C. The reaction products of the RT-PCR were separated after stopping the reaction on a 1% agarose gel (FIG. 1).
6.2 AMV-RT from E. coli LE392 pAMVαβ-4+pCHAP-5 Construct
6.2.1 Purification
E. coli LE392 pAMVαβ-4+pCHAP-5 cells which overexpress both chains of the AMV-RT (see above) were used as the starting material to purify the recombinant AMV-RT.
The AMV-RT was purified at 4° C. The purification was carried out by chromatographic methods after cell lysis and separation of the nucleic acids. The purification yields a recombinant AMV-RT which is free of contaminating enzyme activities and in RT-PCR has the same functionality as an AMV-RT purified from native material.
Buffer
Ca. 50 g E coli LE392 pAMV-αβ-4+pCHAP-5 cells were mixed with 400 ml buffer A, thawed and suspended. Two tablets of Complete (Roche Molecular Biochemicals, cat. No. 1697498) were added to the suspension. Subsequently the cells were lysed by means of ultrasound (Branson sonicator) while cooling (temperature: <10° C.). The degree of lysis of the cell suspension that was achieved was typically 40-50%.
Precipitation of Nucleic Acids
Afterwards the nucleic acids were removed by means of polymin precipitation. 5 ml of a 10% polymin P solution was added dropwise. If the precipitation was incomplete, further dropwise addition was carried out. After incubation for 30 min at 4° C., centrifugation was carried out (30 min, 13000 rpm, 4° C.).
Chromatographic Purifications
Affinity chromatography on a Ni-chelate column:
The clear centrifugation supernatant was absorbed to a nickel-loaded chelating sepharose ff column (2.6 cm×10 cm, Pharmacia) which had been equilibrated with buffer A, it was then washed with ca. 500 ml buffer A, afterwards with 500 ml buffer B and 500 ml buffer C. The enzyme was eluted with buffer D in a total volume of 500 ml. The flow rate was 5 ml per minute, the fraction size was 20 ml per fraction. The pool of active fractions was dialysed against buffer E.
Chromatography on Heparin-sepharose
The dialysed pool was subsequently absorbed to a heparin-sepharose ff column (1.6 cm×10 cm, Pharmacia) equilibrated with buffer E+250 mM NaCl and washed with ca. 200 ml buffer E+250 mM NaCl. The enzyme was eluted with a linear gradient of buffer E+250 mM NaCl and buffer E+1 M NaCl in a total volume of 200 ml. The flow rate was 2.5 ml per min, the fraction size was 10 ml. The AMV-RT eluted at an NaCl concentration of 500 mM to 700 mM.
The RT-active fractions were pooled and dialysed against storage buffer. Application buffer containing SDS and β-mercaptoethanol was added to the purified protein and the sample was denatured by boiling (5 min, 100° C.). Subsequently 20 μl aliquots were analysed by an analytical SDS gel (4-20%) (Laemmli U K., 1970, Nature 227: 555-557). The α- and β-subunits of AMV-RT were found in equimolar ratios (
The described method yields a stable AMV-RT with an equimolar distribution of the α- and β-subunits. The enzyme obtained is functional in RT-PCR.
6.2.2. Functional Test in RT-PCR
The recombinant AMV reverse transcriptase that was obtained was examined in a functional test. The functional test consists of a reverse transcription (RT), followed by a polymerase chain reaction (PCR). 10 units of the recombinant AMV reverse transcriptase was used for this. A 8 kb, 10 kb, 12 kb and a 13.5 kb fragment of the human dystrophin gene was amplified.
1 μg human muscle RNA was used as a template. The primers (400 nM) were the Dys primer 2 for (5′-CAA TCC ATG GGC AAA CTG TAT TCA CTC-3′) and Dys primer 5 rev (5′-CGT CCC GTA TCA TAA ACA TTC AGC AGC-3′) for 8 kb, Dys primer 8 for (5′-AAG AAG TAG AGG ACT GTT ATG AAA GAG AA-3′) and 5 rev for 10 kb, Dys primer 8 for and Dys primer 9 rev (5′-AGC AGG TAA GCC TGG ATG ACT GAC TAG AAG-3′) for 12 kb and Dys primer 8 for and 10 rev (5′-AAT CAA TCA ACC AAC CGA AAT CTC ACT CTG-3′) for 13.5 kb. The cDNA synthesis was carried out for 60 min at 42° C.
The cDNA synthesis was carried out according to the instructions in the product information for the AMV reverse transcriptase (cat. No. 1495062, Roche Molecular Biochemicals).
The Expand Long Template PCR System (cat. No. 1681834, Roche Molecular Biochemicals) was used for the PCR. The target was amplified using the following PCR program: 94° C. for 2 min, followed by 10 cycles (94° C. for 10 sec, 60° C. for 30 sec, 68° C. for 10 min) and 20 cycles (94° C. for 10 sec, 60° C. for 30 sec, 68° C. for 10 min 10+10 sec/cycle). Subsequently it was incubated for 5 min at 68° C. After stopping the reaction, the reaction products of the RT-PCR were separated on 1% agarose gel (FIG. 3). Lanes 3 and 6 show a DNA molecular weight marker X (cat. no. 1498037, Roche Molecular Biochemicals).
| Number | Date | Country | Kind |
|---|---|---|---|
| 100 46 960 | Sep 2000 | DE | national |
| Number | Date | Country |
|---|---|---|
| 0688870 | Dec 1995 | EP |
| 0743365 | Nov 1996 | EP |
| 0774515 | May 1997 | EP |
| 0885967 | Dec 1998 | EP |
| WO9807869 | Feb 1998 | WO |
| WO 9847912 | Oct 1998 | WO |
| WO0042199 | Jul 2000 | WO |
| Number | Date | Country | |
|---|---|---|---|
| 20020115147 A1 | Aug 2002 | US |
