Patent Application
20090170206
The present invention relates to an improved method for efficient and targeted integration of nucleic acids into chromosomes of cells.
Different cell types are used for different industrial purposes. For example mammalian cell lines are used for antibody production; fungal cells are preferred organisms for production of polypeptides and secondary metabolites; bacterial cells are preferred for small metabolite and antibiotic production; plant cells are preferred for taste and flavor compounds. Recombinant techniques are widely employed for optimization of the productivity of cells and/or processes. This can involve a multitude of options, including, but not limited to over expression of a gene of interest, deletion or inactivation of competing pathways, changing compartmentalization of enzymes, increasing protein or metabolite secretion, increasing organelle content and the like (see for example Khetan and Hu (1999) In: Manual of Industrial Microbiology Biotechnology, Eds. Demain and Davies, pg. 717-724). To be successful with these methods it is crucial that the recombinant construct is stably maintained in the production host. This can be either as part of an episomal vector or via integration in the genome. The latter situation is the preferred solution as this is the most stable situation. Even more preferred is the integration at the predetermined, correct genomic locus. Since in several species, especially most eukaryotic organisms, integration of DNA into the genome occurs with high frequency at random, the construction of industrial production cells by recombinant DNA technology often leads to the unwanted integration of the polynucleotide resulting in genome modifications at random. Moreover, this often results in multiple integrations and thus instable situations. This uncontrolled “at random multiple integration” of a polynucleotide is a potentially dangerous process, which can lead to unwanted modification of the genome of the host, resulting in decreased productivity.
It is therefore highly desirable to be able to construct an industrial production cell line by correct genome targeting of the polynucleotide of interest with high efficiency. Furthermore, now that the sequences of complete genomes of an increasing amount of organisms are becoming available, the opportunity to construct genome-wide over expression and deletion libraries is opened. An important requirement for the efficient construction of such libraries is that the organism in question can be efficiently transformed, that the polynucleotide of interest is correctly targeted with a high frequency and that the required homology needed to direct targeted integration of a nucleic acid into the genome is relatively short.
There are several methods described to decrease the frequency of this unwanted, at random integration of polynucleotides in cells.
Eukaryotic cells have at least two separate pathways (one via homologous and one via non-homologous recombination) through which nucleic acids (in particular of course DNA) can be integrated into the host genome. The yeast Saccharomyces cerevisiae is an organism with a preference for homologous recombination (HR). The ratio of homologous to non-homologous recombination (HR/NHR) of this organism may vary from about 0.9 to 1. Contrary to Saccharomyces cerevisiae, most higher eukaryotic cells (including fungal, plant and mammalian) cells have a preference for non-homologous recombination (NHR). Among these, the HR/NHR ratio ranges between 0.0001 and 0.5. In such organisms, the targeted integration frequency is rather low. Also, the length of homologous regions flanking a polynucleotide sequence to be integrated into the genome of such organisms has to be relatively long, for example at least 2,000 base pairs for disrupting a single gene. The necessity of such flanking regions represents a heavy burden when cloning the DNA construct comprising said polynucleotide and when transforming the organism with it. Moreover, neighboring genes which lie within those flanking regions can easily be disturbed during the recombination processes following transformation, thereby causing unwanted and unexpected side-effects.
Recently, several publications describe the inhibition of the very efficient Non-Homologous End-Joining (NHEJ) pathway, the pathway responsible for random integration of polynucleotides in cells, as a method for improving the HR/NHR ratio (see for example Ninomiya et al., 2004, Proc. Natl. Acad. Sci. USA 101:12248-12253; Krappmann et al., 2006, Eukaryot. Cell. 5:212-215). It is potentially a very powerful method, resulting in very significant improvements (even up to 60-fold) of gene targeting efficiency.
However, there are still some drawbacks to this method. Firstly, it does not work for all species. For example, mammalian cells deficient in ku70, one of the components of the NHEJ pathway, have been isolated (Pierce et al., 2001, Genes and Development 15: 3237-3242). These mutants have a six-fold higher homology-directed repair frequency, but no increase in the efficiency of homology-directed targeted integration. Secondly, although it has a positive effect on the NHR/HR ratio in several fungal species (see for example Ninomiya et al., 2004; Krappmann et al., 2006), in most cases it is limited to 60-90% correct gene targeting. This is an acceptable improvement for working with one or several genes, but not for a High Throughput genome wide analysis and/or modification of gene function. In the individual cases where 100% correct transformants were obtained this involved long flanking regions, which also is not amenable for a High Throughput genome wide analysis and/or modification of gene function. Thirdly, to obtain such strains with improved HR/NHR ratios, one has to modify the recombination machinery of the host cell and this can lead to unwanted side effects (see for example Celli et al., Nat Cell Biol (2006), 8: 885-890).
The HR/NHR ratio can also be improved by over expressing components of the HR pathway. An example of this method is given by Shaked et al. (2005, Proc Natl. Acad. Sci. USA. 102:12265-12269). They show that by over expression of yeast RAD54 the HR frequency can be improved a 100-fold. Still, this results only 1-10% correct transformants, which makes this method not amenable for a High Throughput genome wide analysis and/or modification of gene function.
Another method is the so-called bipartite gene-targeting method (Nielsen et al., 2006, 43: 54-64). This method is using two overlapping non-functional parts of a selection marker. Upon correct homologous recombination the selection marker becomes functional. They tested the method in the fungal species with the most efficient homologous recombination system, Aspergillus nidulans, with 24% correct gene targeting in WT cells. The method results in a 2.5-fold improvement over the standard method, but even in Aspergillus nidulans only 62% of the transformants obtained is correct. Also, rather long flanking regions are used to obtain correct targeting. This is an acceptable improvement for working with one or several genes, but not for a High Throughput genome wide analysis and/or modification of gene function.
Liu et al. (2001, J. Bacteriol., 183: 1765-1772) describe another method, which uses a second selection marker to enrich for transformants with targeted gene disruption in Acremonium chrysogenum. The method results in a 10-fold improvement over the standard method, but still only 8% of the transformants obtained is correct.
Still another method is described by Kang and Khang (US 2005/0181509). This is a variation on the method of Liu et al. (2001). Here they apply a negative selection marker, i.e. the herpes simplex virus thymidine kinase (HSVtk) gene, as the second selection marker. If the selection procedure would work correctly, polynucleotides that integrate at random in the genome would kill the cells as the HSVtk gene would convert the 5-fluoro-2′-deoxyurine in the agar plates to a toxic compound. Again, this method increases the frequencies of correct targeting in cells, but it is limited to 50% of the cells. More importantly there is a very high percentage of false positives obtained (9-100%), which makes this method unsuitable for a High Throughput genome wide analysis and/or modification of gene function. Kang and Khang (US 2005/0181509) also describe the testing of the diphtheria toxin A (dtA) gene. This gene has been applied in plants and mammalian cells as a second marker to increase the frequency of correct gene targeting to 1-2% (see for examples Terada et al., 2004, Plant Cell Rep. 22:653-659; Yagi et al., 1993, Anal. Biochem. 214:77-86). However, they failed to get this marker functional in fungal species.
Surprisingly, we found that by combining a method that increases the HR/NHR ratio with the use of a second selectable marker, we can get correct gene targeting at 100% in cells which normally have a preference for NHR. Moreover, we surprisingly found that the diphtheria toxin A (dtA) gene does work in filamentous fungal cells and can be used efficiently in fungal species as a second (and lethal) marker to enrich for cells wherein a correct gene targeting event has occurred.
The present invention discloses a method to construct eukaryotic cells having a target sequence in a chromosomal DNA sequence replaced by a desired replacement sequence comprising:
modifying a parent eukaryotic cell with a preference for NHR to provide a eukaryotic cell having an increased HR/NHR ratio as compared to the parent cell,
providing a DNA molecule comprising a first DNA fragment comprising a desired replacement sequence flanked at its 5′ and 3′ sides by DNA sequences substantially homologous to sequences of the chromosomal DNA flanking the target sequence and a second DNA fragment comprising an expression cassette comprising a gene encoding a selection marker and regulatory sequences functional in the eukaryotic cell operably linked thereto;
transforming the modified eukaryotic cells with the DNA molecule;
growing the cells to obtain transformed progeny cells having the DNA molecule inserted into the chromosome,
deselecting the progeny cells in which the DNA molecule is inserted in the chromosome via a non-homologous recombination event by expression of the selection marker; and
obtaining cells wherein the target sequence in the chromosomal DNA sequence is replaced by the desired replacement sequence.
The parent eukaryotic cell with a preference for NHR may be any eukaryotic cell having a HR/NHR ratio ≦0.5, preferably ≦0.2, more preferably <0.1, most preferably ≦0.05. The exact ratio may vary between different loci in one species. Suitable loci for determining the ratio's above are the niaD locus (and its homologues in all species) and the KU70 locus (and its homologues in all species).
The HR/NHR ratio is determined by analysis of a set of transformants to determine which part of the transformants has correct gene targeting (HR) and which part of the transformants has incorrect recombination (NHR). Several methods are available to obtain these data:
The preferred method for quickly scanning many transformants is PCR.
Eukaryotic host cells having an increased HR/NHR ratio as compared to a parent cell may be obtainable by modifying the parent eukaryotic cell by increasing the efficiency of the HR pathway and/or by decreasing the efficiency of the NHR pathway. The HR/NHR ratio thereby is increased at least 2 times, preferably at least 4 times, more preferably at least 10 times.
In the context of the invention, the HR pathway is defined as those genes and elements being involved in the control of the targeted integration of polynucleotides into the genome of a host, said polynucleotides having a certain homology with a certain predetermined site of the genome of a host to where the integration is targeted. The NHR pathway is defined as all genes and elements being involved in the control of the integration of polynucleotides into the genome of a host irrespective of the degree of homology of the polynucleotides with a genomic sequence of the host.
Increasing the efficiency of the HR pathway and/or decreasing the efficiency of the NHR pathway may be effectuated by up and/or down regulating the expression of genes involved in HR and/or NHR. The expression level of a DNA sequence, e.g. a gene, is up or down regulated, respectively, when the expression level of this DNA sequence in the obtained eukaryotic cell is higher or lower, respectively, than the expression level of the same DNA sequence in the parent eukaryotic cell, preferably at least 2 times higher or lower, more preferably at least 4 times higher or lower, even more preferably at least 10 times higher or lower. When down regulated, the expression of the DNA sequence most preferably is not detectable.
The up and/or down regulation of the expression level of a DNA sequence may be monitored by quantifying the amount of corresponding mRNA present in a cell by Northern blotting (see for instance Molecular Cloning: A Laboratory Manual, Sambrook et al., New York: Cold Spring Harbour Press, 1989) and/or by quantifying the amount of corresponding protein present in a cell by Western blotting. mRNA and protein levels may also be measured by ‘omics’ techniques like transcriptomics and/or proteomics. The difference in mRNA amount may also be quantified by DNA array analysis (Eisen, M. B. and Brown, P. O. DNA arrays for analysis of gene expression. Methods Enzymol. 1999: 303: 179-205). The up and/or down regulation of the expression level of a DNA sequence may be obtained by subjecting the parent eukaryotic cell to recombinant genetic manipulation techniques and/or classical mutagenesis techniques. Up regulation of the expression of a DNA sequence using recombinant genetic manipulation techniques preferably comprises overexpression of the DNA sequence. Down regulation of the expression of a DNA sequence using recombinant genetic manipulation techniques preferably comprises inactivation of the DNA sequence. Inactivation may be done by replacing the DNA sequence by a non-functional variant thereof or by deleting part or all of the DNA sequence from the genome.
In one embodiment, the up and/or down regulation of the expression level of a DNA sequence may be inducible. This can be achieved by replacing the endogenous regulatory regions of the DNA sequence, i.e. a gene encoding the component involved in HR and/or NHR, by new regulatory regions, preferably comprising a repressible or inducible promoter, more preferably comprising a promoter that can be switched on/off, for instance by glucose repression, ammonia repression, and/or pH repression. Examples of fungal glucose-repressed promoters are the Penicillium chrysogenum pcbAB promoter (Martin et al., 1999, Antonie Van Leeuwenhoek. 75:21-31) or the Aspergillus niger glucoamylase promoter. Examples of on/off switchable promoters are described in the following publications: An activator/repressor dual system allows tight tetracycline-regulated gene expression in budding yeast (Belli et al., 1998, Nucl. Acid Res. 26: 942-947); a light-switch able gene promoter system (Shimizu et al., 2002, Nat. Biotech. 20:1041-1044).
A suitable eukaryotic cell may subsequently be selected by monitoring the expression level of the relevant DNA sequence. Optionally, the eukaryotic cell may be selected by measuring the efficiency of the NHR and/or of the HR pathways and/or the HR/NHR ratio. In the context of the invention, the efficiency of the HR pathway of a filamentous fungus may be measured by the efficiency of the targeted integration of a given polynucleotide sequence into a predetermined site in the genome of the filamentous fungus using given homology region(s). In the context of the invention, the efficiency of the NHR pathway of a filamentous fungus may be measured by the efficiency of the non-targeted integration of a given polynucleotide sequence in the genome of the filamentous fungus irrespective of any homology region(s).
In a preferred embodiment, eukaryotic cells having an increased HR/NHR ratio as compared to a parent cell are obtainable by decreasing the efficiency of the NHR pathway. This can be achieved by down regulating the expression of a gene involved in NHR in the eukaryotic cell as comparison to the expression of said gene in the parent eukaryotic cell measured under identical conditions.
According to a preferred embodiment, the eukaryotic cell having an increased HR/NHR ratio as compared to a parent cell is deficient in at least one of its endogenous genes that are equivalents of the following yeast genes involved in the NHR pathway: KU70, KU80, RAD50, MRE11, XRS2, LIG4, LIF1, NEJ1 and SIR4 (van den Bosch et al., 2002, Biol. Chem. 383: 873-892 and Allen et al., 2003, Mol. Cancer. Res. 1:913-920).
An example of equivalents of the yeast KU70 and KU80 genes, respectively, are the genes hdfA or mus51 and hdfB or mus52, respectively, from filamentous fungi (see WO05095624 and Ninomiya et al., 2004). Mammalian equivalents of the yeast genes are also known as Ku70 or Ku80 genes (see for example Pierce et al., 2001).
The eukaryotic cell having an increased HR/NHR ratio as compared to a parent cell as specified hereinabove is preferably used as a host cell for transformation using the DNA molecule as specified herein below.
The first DNA fragment of the DNA molecule comprises a desired replacement sequence flanked at its 5′ and 3′ sides by DNA sequences substantially homologous to sequences in the chromosomal DNA flanking the target sequence.
With the term “substantially homologous” as used in this invention is meant that a DNA sequence flanking the replacement sequence has a degree of identity to a chromosomal DNA sequence flanking the target sequence of at least 80%, preferably at least 90%, over a region of not more than 3 kb, preferably not more than 2 kb, more preferably not more than 1 kb, even more preferably not more than 0.5 kb, even more preferably not more than 0.2 kb. even more preferably not more than 0.1 kb, even more preferably not more than 0.05 kb, most preferably not more than 0.03 kb. The degree of required identity may thereby depend on the length of the substantially homologous sequence. The shorter the homologous sequence, the higher the percentage homology may be.
It will be obvious to the skilled person that, in order to achieve homologous recombination via a double cross-over event, these flanking sequences need to be present at both sides of the replacement sequence and need to be substantially homologous to sequences at both sides of the target sequence in the chromosome.
The nature of the replacement sequence may vary depending on the intended use. The replacement sequence may for instance confer a selectable phenotype to the eukaryotic cell. In that case, the replacement sequence comprises a selection marker. Preferably, the selection marker is a positive selection marker. A preferred positive selection marker is the amdS gene. A selection marker as replacement sequence preferably is used when the target sequence needs to be inactivated.
The replacement sequence may also be a modified version of the target sequence, for instance to provide for altered regulation of a sequence of interest or expression of a modified gene product with altered properties as compared to the original gene product.
The replacement sequence may also constitute additional copies of a sequence of interest being present in the genome of a eukaryotic cell, to obtain amplification of that sequence of interest.
The replacement sequence may be a sequence homologous or heterologous to the eukaryotic cell of interest. It may be obtainable from any suitable source or may be prepared by custom synthesis.
The target sequence may be any sequence of interest. For instance, the target sequence may be a sequence of which the function is to be investigated by inactivating or modifying the sequence. The target sequence may also be a sequence of which inactivation, modification or overexpression is desirable to confer a eukaryotic cell with a desired phenotype.
The second DNA fragment comprises an expression cassette providing for expression of a selection marker. However, only a non-homologous recombination event will lead to actual integration of the selection marker cassette. This implicates that expression of this selection marker product will occur upon integration of the DNA molecule comprising the first and second DNA fragment into the chromosome of the eukaryotic cell via non-homologous recombination. Integration of the expression cassette thus will occur when the DNA molecule in integrated at a site that is not homologous to the target sequence.
Cells that express the selection marker are deselected, i.e. are removed from the population of transformed progeny cells.
In the context of the invention, the second DNA fragment encoding a selection marker confers a detectable phenotype to the recipient eukaryotic cell. Examples of detectable phenotypic selection markers are Green Fluorescent Protein (eGFP), Yellow Fluorescent Protein (eYFP), Cyan Fluorescent Protein (eCFP), Blue Fluorescent Protein (eBFP), Red Fluorescent Protein (RFP), β-galactosidase (β-gal), glucoronidase (GUS). Preferably, such detectable phenotypic selection markers are selectable markers. Examples of selectable markers are, but not limited to, tn5-ble (phleomycinR), hyg (hygromycinR), kan (kanamycinR), gen (G418R), amdS (acetamide utilisation).
In a preferred embodiment, the second DNA fragment encoding a detectable phenotypic selection marker is a lethal selection marker, which confers a lethal phenotype to the eukaryotic cell. Such a lethal phenotype may be achieved in various ways. A lethal phenotype for instance may directly be achieved when the lethal selection marker encodes a compound that is toxic to the cell, e.g. diphtheria toxin A or suppressor tRNA's. A lethal phenotype may also conditionally be achieved when the lethal selection marker encodes an enzyme that converts a particular substrate into a toxic compound. Examples of such conditional lethal markers, in combination with a substrate that can be converted into a toxic compound, are: nitrate reductase (niaD gene) in combination with chlorate, orotidine-5′-monophosphate decarboxylase (pyrG gene) in combination with fluoro-orotic acid, acetamidase (amdS gene) in combination with fluoro-acetamide, or thymidine kinase (tk gene) in combination with 5-fluoro-2′-deoxyuridine.
The expression cassette providing for expression of the selection marker comprises regulatory sequences operably linked to the selection marker-encoding gene The term “operably linked” refers to a juxtaposition wherein the components described are in a relationship permitting them to function in their intended manner. A regulatory sequence such as a promoter, an enhancer or another expression regulatory signal “operably linked” to a coding sequence is positioned in such a way that expression of a polypeptide from its coding sequence is achieved under conditions compatible with the regulatory sequences.
The regulatory sequences of the selection marker expression cassette preferably are heterologous to the chromosome of the eukaryotic cell of interest, i.e. the regulatory sequences are from a different eukaryotic species than the eukaryotic cell of interest to be transformed. The use of a homologous regulatory sequence in this context may result in a targeted integration event at a chromosomal site corresponding to the homologous regulatory sequence. Such an integration event is undesirable because it decreases the percentage correct targeting to the site comprising the targeting sequence.
Regulatory sequences used may drive constitutive expression; this will enable to use the negative selection directly after transfection. Alternatively, regulatory sequences may be used that drive regulatable or inducible expression of the selection marker gene; this allows for a two step procedure. Firstly, the transfection and subsequent isolation of transformants is performed, under conditions that the transcription of the selection marker gene does not occur. Secondly, the isolated transformants are transferred to conditions which induce the transcription of the negative selection marker gene, thereby selectively killing all the isolates that underwent random integration events.
The DNA molecule may comprise the first and second DNA fragment in any order and preferably is linear. If the replacement sequence does not comprise a selection marker, such a marker may be provided on a separate DNA molecule.
A eukaryotic cell of interest is transformed with the DNA molecule comprising the first and second DNA fragment, and, optionally, a DNA molecule comprising a selection marker, using techniques commonly known in the art. Briefly, eukaryotic cells are transformed by contacting the eukaryotic cells with a suitable amount of the DNA molecule(s), preferably in linear form, and selecting colonies of transformed cells by culturing the cells on a selective medium enabling growth of transformed cells only.
Upon transformation, the DNA molecule comprising the first and second DNA fragment integrates in the chromosome of the eukaryotic host cell by a homologous or a non-homologous integration event. A homologous integration event occurs at the target sequence in the host chromosome by a double cross-over event at the homologous sequences flanking the replacement and targeting sequence. Such an event ensures that the second DNA fragment comprising the selection marker expression cassette is not integrated into the chromosome. A non-homologous integration event results in integration of the complete DNA molecule comprising first and second DNA fragments. Cells wherein a non-homologous integration event has occurred are deselected when the selection marker gene is expressed upon integration. This expression of the selection marker gene may occur simultaneously with selection of the transformants or may occur in a later stage after transformants have been selected. In the latter case, expression of the selection marker gene may be induced by a suitable inducer.
The eukaryotic cell is a eukaryotic cell with a preference for NHR. Preferably, the eukaryotic cell is a fungal, plant or animal cell. More preferably, the fungus is of the genus Aspergillus, Penicillium, Acremonium, Trichoderma, Chrysosporium, Mortierella, Kluyveromyces or Pichia; most preferably of the species Aspergillus niger, Aspergillus nidulans, Aspergillus oryzae, Aspergillus terreus, Penicillium chrysogenum, Penicillium citrinum, Acremonium chrysogenum, Trichoderma reesei, Mortierella alpina, Chrysosporium lucknowense, Kluyveromyces lactis, Pichia pastoris or Pichia ciferrii. More preferably, the plant is of the genus Arabidopsis, Nicotiana, Solanum, Lactuca, Brassica, Oryza, Asparagus, Pisum, Medicago, Zea, Hordeum, Secale, Triticum, Capsicum, Cucumis, Cucurbita, Citrullis, Citrus, Sorghum; most preferably of the species Arabidopsis thaliana, Nicotiana tabaccum, Solanum lycopersicum, Solanum tuberosum, Solanum melongena, Solanum esculentum, Lactuca sativa, Brassica napus, Brassica oleracea, Brassica rapa, Oryza glaberrima, Oryza sativa, Asparagus officinalis, Pisum sativum, Medicago sativa, Zea mays, Hordeum vulgare, Secale cereale, Triticum aestivum, Triticum durum, Capsicum sativus, Cucurbita pepo, Citrullus lanatus, Cucumis melo, Citrus aurantifolia, Citrus maxima, Citrus medica, Citrus reticulata. More preferably, the animal cell is of the genus Homo, Rattus, Mus, Sus, Bos, Danio, Canis, Felis, Equus, Salmo, Oncorhynchus, Gallus, Meleagris, Drosophila, Caenorhabditis; most preferably of the species Homo sapies, Rattus norvegicus, Mus musculus, Sus scrofa, Bos taurus, Danio rerio, Canis lupus, Felis catus, Equus caballus, Salmo salar, Oncorhynchus mykiss, Gallus gallus, Meleagris gallopavo, Drosophila melanogaster, Caenorhabditis elegans.
By using eukaryotic cells with a increased HR/NHR ratio, the method of the invention advantageously ensures an increased targeting of a polynucleotide to a chromosomal target site of interest. The method of the invention further advantageously allows the provision of transformed eukaryotic cells that are enriched in cells wherein the correct targeted integration event has occurred. In particular, at least 50% of the transformed clones has the replacement sequence targeted to the target sequence in the chromosome, as compared to 1-2% in a transformation with a targeting construct without a second fragment comprising a selection marker.
Standard procedures were carried out as described elsewhere (Sambrook et al., 1989, Molecular cloning: a laboratory manual, 2nd Ed., Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.). DNA for plasmid construction was amplified using a proofreading polymerase, Phusion (Finnzymes) following the manufacturer's protocol; while verification of constructed strains and plasmids was achieved by using Taq polymerase. Restriction enzymes were from Invitrogen or New England Biolabs. For routine cloning, Escherichia coli strains Top10 and DH10B (Invitrogen) were employed. The Gateway system of Invitrogen was applied according to the manufacturer's manuals. Verification of the constructed plasmids was carried out by restriction analysis and subsequent sequencing.
To delete the chromosomal copy of the Penicillium chrysogenum hdfA gene (see WO05095624), the fungal equivalent of the yeast KU70 gene involved in the Non-Homologous End-Joining pathway, the integration vector pdel-hdfA was constructed. To this end 2500 bp genomic DNA fragments both directly upstream and directly downstream of the hdfA gene were PCR amplified. The upstream region (alias left flanking; PCR1A in
The deletion fragment was isolated from plasmid pdeI-hdfA as a SfiI fragment and transfected to Penicillium chrysogenum protoplasts. Penicillium chrysogenum protoplasts were produced according to standard protocols (see for examples Cantoral et al., 1987, Biotechnology 5: 494-497; Swinkels et al., 1997, WO97/06261) however Glucanex™ (Sigma) was applied as lysing enzyme. After transfection the protoplasts were plated out on selective regeneration agar plates with acetamide as the sole nitrogen source (for an exact description of the media, see Swinkels et al., 1997). Acetamide utilizing colonies were transferred to fresh acetamide plates (without saccharose to induce sporulation) and 300 of these were subsequently screened for phleomycin sensitivity (on mineral medium agar as described for Penicillium chrysogenum in US 2002/0039758, without phenyl acetic acid but supplemented with 15 g/L agar to solidify and 50 mg/L phleomycin). The thus obtained transformants should be enriched for targeted integration at the hdfA locus as the random transformants would be phleomycin resistant (see
Confirmation of hdfA Deletion
Six of the 9 putative hdfA mutants were screened with colony PCR to determine if the deletion was correct. To this end a piece of the sporulated colony was resuspended in 50 μL of water and boiled for 10 minutes at 98 degrees Celsius. The cell debris was spun down and one μl of the supernatant was used in 3 PCR reactions (see
Obtaining a Marker-Free hdfA Mutant
In order to obtain a clean hdfA mutant the amdS selection marker cassette had to be removed. To facilitate this, a 1000 bp repeat was introduced during cloning (see
Phenotype of the Marker-Free hdfA Mutant
To determine if the hdfA mutants had no obvious phenotype 5 items were checked: penicillin V production, morphology, growth rate, sporulation and phleomycin sensitivity. Penicillin V production was checked in liquid medium as described in US 2002/0039758. Four strains were checked: the WT, SA1 (hdfA mutant containing amdS), AS1 and AS2 (hdfA mutants both marker-free). The productivities were respectively 100%, 114%, 105% and 100%. So, the hdfA deletion causes no apparent change towards penicillin production. Morphology was checked by growing the mutants on plate. No significant difference could be observed versus the WT. So, the hdfA deletion causes no apparent change towards morphology. Growth rate was checked during the Penicillin V productivity tests as described above by measuring the OD600. No significant difference could be observed versus the WT. So, the hdfA deletion causes no apparent change towards growth rate. Sporulation was checked by visually comparing the coloring and structure of the colonies on agar plates. No significant difference could be observed versus the WT. So, the hdfA deletion causes no apparent change towards sporulation. As the hdfA gene product possibly has a role in repair mechanisms the sensitivity versus phleomycin, a known inducer of DNA breaks, was checked. This was done by testing the growth on mineral medium agar plates as described for Penicillium chrysogenum in US 2002/0039758, without phenyl acetic acid but supplemented with 15 g/L agar to solidify and 0-50 mg/L phleomycin. No significant difference could be observed versus the WT. So, the hdfA deletion causes no apparent change towards phleomycin sensitivity. Finally, we concluded that the hdfA mutant has no apparent phenotype and can be used for further studies.
To delete the chromosomal copy of the Penicillium chrysogenum hdfB gene (see WO05095624), the fungal equivalent of the yeast KU80 gene involved in the Non-Homologous End-Joining pathway, the integration vector pdeI-hdfB was constructed. The construction was generally the same as outlined in
The deletion fragment was isolated from plasmid pdeI-hdfB as a SfiI fragment and transfected to Penicillium chrysogenum protoplasts (produced according to standard protocols, i.e. Cantoral et al., 1987, Biotechnology 5: 494-497; Swinkels et al., 1997, WO97/06261) however Glucanex™ (Sigma) was applied as lysing enzyme. After transfection the protoplasts were plated out on selective regeneration agar plates with acetamide as the sole nitrogen source (media described in Swinkels et al., 1997). Acetamide utilizing colonies were transferred to fresh acetamide plates (without saccharose to induce sporulation) and 375 of these were screened for phleomycin sensitivity (on mineral medium agar as described for Penicillium chrysogenum in US 2002/0039758, without phenyl acetic acid but supplemented with 15 g/L agar to solidify and 50 mg/L phleomycin). The thus obtained transformants should be enriched for targeted integration at the hdfA locus as the random transformants would be phleomycin resistant (outlined for hdfA in
Confirmation of hdfB Deletion
Eight of the 12 putative hdfB mutants were screened with colony PCR to determine if the deletion was correct. To this end, part of the sporulated colony was resuspended in 50 μl of water and boiled for 10 min at 98° C. The cell debris was spun down and one μl of the supernatant was used in 3 PCR reactions (similar set-up as outlined for hdfA in
Obtaining Marker-Free hdfB Mutants
In order to obtain a clean hdfB mutant the amdS selection marker had to be removed. To facilitate this, a 1000 bp repeat was introduced during cloning (see above). This 1000 bp is exactly the same as the first 1000 bp of the right flanking region used in the deletion construct. Thus, if a recombination would take place at this 1000 bp a clean hdfB delection should be obtained in which the flanking regions are retained and the complete hdfB locus will be removed. The amdS selection marker has the advantage that one can select for the presence (i.e. enabling growth on acteamide as the sole nitrogen source) as well on the absence (i.e. fluoroacetamide would generate toxic fluor in the presence of amdS) of the gene. Spores of the 5 hdfB strains were plated on fluoroacetamide containing agar plates (see WO9706261) and growing colonies were isolated and checked with colony PCR. To this end PCR template material was obtained as described above and the PCR was run using the oligonucleotides SEQ ID NO 29 and SEQ ID NO 30. If amdS would be present this should give a fragment of 5.5 kb, while in the recombined situation (i.e. marker-free hdfB deletion) this should give a fragment of 1 kb (as outlined for hdfA in
Phenotype of the Marker-Free hdfB Mutant
As for the hdfA mutant we determined if the hdfB mutants had no obvious phenotypes. The same 5 items were checked: penicillin V production, morphology, growth rate, sporulation and phleomycin sensitivity. Penicillin V production was checked in liquid medium as in US 2002/0039758. Two marker-free derivatives were checked per amdS-positive hdfB strain and compared to the productivity of the WT strain (Table 1).
The productivities all are around 100+/−10% of the WT strains. So, the hdfB deletion causes no apparent change towards penicillin production. Morphology was checked by growing the mutants on plate. No significant difference could be observed versus the WT. So, the hdfB deletion causes no apparent change towards morphology. Growth rate was checked during the PenicillinV productivity tests as described above by measuring the OD600. No significant difference could be observed versus the WT. So, the hdfB deletion causes no apparent change towards growth rate. Sporulation was checked by visually comparing the coloring and structure of the colonies on agar plates. No significant difference could be observed versus the WT. So, the hdfB deletion causes no apparent change towards sporulation. As the hdfB gene product possibly has a role in repair mechanisms the sensitivity versus phleomycin, a known inducer of DNA breaks, was check. This was done by testing the growth on mineral medium agar plates as described for Penicillium chrysogenum in US 2002/0039758, without phenyl acetic acid but supplemented with 15 g/L agar to solidify and 0-50 mg/L phleomycin. No significant difference could be observed versus the WT. So, the hdfB deletion causes no apparent change towards phleomycin sensitivity. Finally, we concluded that, like the hdfA mutant, the hdfB mutant has no apparent phenotype and can be used for further studies.
To determine the homologous recombination frequency of hdfA and hdfB deletion strains compared to the WT a deletion cassette for the niaD locus, encoding nitrate reductase, was constructed. To this end the two 1800 bp fragments of the niaD locus were amplified, these were separated by the amdS expression cassette and should enable the exchange of part of the niaD gene for the amdS expression cassette, thereby rendering the transformants chlorate resistant (due to the deletion of niaD) and able to use acetamide as the sole nitrogen source. Three PCR reactions were performed (see Table 2) and cloned into the Gateway pDONR vectors of Invitrogen, enabling later efficient fusion via the Multi-Gateway system (see www.Invitrogen.com).
The three obtained vectors were used to perform a multi-site Gateway reaction into pDESTR4-R3, yielding plasmid pDESTR4R3NamdSN (see
To determine the effect of a second selectable marker the deletion construct pDESTR4R3NphleoNamdS (see
The three obtained vectors were used to perform a multi-site Gateway reaction into pDESTR4-R3, yielding plasmid pDESTR4R3NphleoN. The PgpdA-amdS fragment can be isolated as a KpnI-EcoRI fragment after PCR amplification, using the oligonucleotides of SEQ ID NO's 39 and 40, from the plasmid pHELY::A1 (WO04106347) and cloned into the EcoRI-KpnI sites of pDESTR4R3NphleoN, yielding plasmid pDESTR4R3NphleoNamdS. This plasmid was linearized with NotI and transfected to Penicillium chrysogenum protoplasts (produced according to standard protocols, i.e. Cantoral et al., 1987, Biotechnology 5: 494-497; Swinkels et al., 1997, WO97/06261) however Glucanex™ (Sigma) was applied as lysing enzyme. After transfection the protoplasts were plated out on selective regeneration agar plates with phleomycin (on mineral medium agar as in US 2002/0039758, without phenyl acetic acid but supplemented with 1M Saccharose, 15 g/L agar to solidify and 50 mg/L phleomycin). After sporulation the absence of the amdS gene was checked on acetamide and fluoroacetamide agar plates (WO9706261). A final check was done with PCR, using two specific primer sets which should give only fragments in mutants with a targeted deletion (
A second example of very efficient gene targeting by combining the effect strains with an optimized frequency of homologous recombination and the use of a second selectable marker was obtained by using two constructs named pDONR201amdSni800 and pDONR201amdSni1200 (
The vectors pDONR201amdSni800 and pDONR201amdSni1200 were linearized with either MluI or NruI, and transfected to Penicillium chrysogenum protoplasts (produced according to standard protocols, i.e. Cantoral et al., 1987, Biotechnology 5: 494-497; Swinkels et al., 1997, WO97/06261) however Glucanex™ (Sigma) was applied as lysing enzyme. After transfection the protoplasts were plated out on selective regeneration agar plates with nitrate as the sole nitrogen source (per liter: NaCl, 3.0 g; KH2PO4, 1.0 g; FeSO4.7H20, 10.0 mg; MgSO4.7H2O, 0.5 mg; Glucose, 10.0 g; Saccharose, 342.0 g; Oxoid agar, 15.0 g; NaNO3, 1 g; 1 M phosphate buffer pH6.8, 10 ml). Eleven transformants were obtained (see Table 6) and these were transferred to fresh plates, without saccharose to induce sporulation. After sporulation the absence of the amdS gene was checked on acetamide agar plates (see WO9706261) and none of the transformants was able to grow, showing that no non-homologous recombination occurred and only 100% correct gene targeting was obtained. The results clearly show the beneficial effect of combining strains with improved HR/NHR ratios and the use of a second selectable marker. The gene targeting frequency in WT strains is around 1% (see example 3). Strains with improved HR/NHR ratios, but transformed without a second selectable marker show 47-56% correct gene targeting (see example 3). Strains with normal HR/NHR ratios (i.e. WT strains), but transformed with a second selectable marker show 16-63% correct gene targeting (examples 1 and 2). However, when using strains with improved HR/NHR ratios and a second selectable marker surprisingly a 100% correct gene targeting is obtained (this example).
| Number | Date | Country | Kind |
|---|---|---|---|
| 06075904.0 | Apr 2006 | EP | regional |
| Filing Document | Filing Date | Country | Kind | 371c Date |
|---|---|---|---|---|
| PCT/EP2007/052392 | 3/14/2007 | WO | 00 | 10/8/2008 |
