Phleomycin Increases Transformation Efficiency and Promotes Single Integrations in Schizophyllum commune

Abstract

Phleomycin is mutagenic by introducing double-strand breaks in DNA. The ble gene of Streptoalloteychus hindustanus, which confers resistance to this substance, is widely used as a selection marker for transformation. Schizophyllum commune grows on 25 μg of phleomycin ml⁻¹ after introduction of a resistance cassette based on the ble gene. However, we here report that growth of resistant colonies on this concentration of phleomycin resulted in aberrant colony morphologies. Apparently, phleomycin was mutagenic despite acquired resistance. Therefore, a new selection system was developed based on resistance to the antibiotic nourseothricin. However, the transformation efficiency was tenfold lower than that obtained with phleomycin as a selection agent. This low transformation efficiency could be rescued by addition of a nonselective concentration of phleomycin during protoplast regeneration. This was accompanied by a higher incidence of single-copy integrations and with an increase of expression of key genes involved in double-strand break repair. Taken together, we conclude that the effect of a nonselective concentration of phleomycin strongly resembles the effect of restriction enzyme-mediated integration (REMI) but, unlike REMI, it does not depend on the presence of a target restriction site.

Phleomycin and other bleomycins are widely used as selection agents for the transformation of algae (6, 9), protista (36), animals (4, 24), and fungi (2, 3, 15, 17, 35). They introduce double-strand breaks in the DNA when activated by metal ions (mainly iron) and oxygen (34). In addition, bleomycins damage RNA and attack cell walls (5). Resistance to phleomycin is conferred by the ble gene of Streptoalloteychus hindustanus. This gene encodes a 14-kDa protein that is capable of sequestering bleomycin-like molecules in a reversible way (12). The basidiomycete Schizophyllum commune can be efficiently transformed by using a phleomycin resistance cassette, in which the ble gene of S. hindustanus is placed under the control of the regulatory sequences of the S. commune glyceraldehyde-3-phosphate dehydrogenase gene (GPD) (30). However, we here show that phleomycin-resistant strains of S. commune are mutated upon exposure to phleomycin. Therefore, a cassette was constructed that confers resistance to a new selection marker, nourseothricin. Addition of a nonselective concentration of phleomycin during protoplast regeneration promoted single-copy integration of the construct and resulted in an increased transformation frequency independent of the selection marker used.

MATERIALS AND METHODS

Strains and growth conditions.

The co-isogenic S. commune strains 4-39 (CBS 341.81) and 4-40 (CBS 340.81) were used, as well as the uracil auxotroph 12-42 and the 4-39 derivative 4-39P. The latter strain has been transformed to phleomycin resistance using plasmid pHYM1.2. This plasmid contains the S. hindustanus ble gene under the control of the upstream and downstream regulatory sequences of the S. commune GPD gene (27). Strains were grown in minimal medium (MM) (7) at 25 or 30°C.

Plasmids used in the present study.

pGEMPhleo and pGEMNour are derivatives of pGEM-T Easy (Promega, Madison, WI). They contain EcoRI fragments encompassing a phleomycin (30) and a nourseothricin resistance cassette, respectively. The latter cassette consists of the nat1 gene of Streptomyces noursei (accession no. S60706) (19) placed under the control of a 230-bp S. commune GPD promoter and a 450-bp S. commune SC3 terminator. Plasmid pNOURAD consists of a pUC20 backbone containing both the phleomycin and the nourseotricin resistance cassettes. These cassettes can be excised from the plasmid using EcoRI. Plasmid pUBB20 consists of a pUC20 backbone with a 2.8-kb BamHI fragment containing the URA1 gene of S. commune (11).

Transformation of S. commune.

Monokaryotic mycelium was grown for 3 days in 100 ml of MM in a 250-ml Erlenmeyer flask at 250 rpm at 25°C. The culture was diluted twofold in MM, homogenized for 15 s in a blender (Waring Products, Inc., Waring Laboratory, Torrington, CT), and grown for another 24 h in 100 ml of MM in a 250-ml Erlenmeyer flask. Mycelium was pelleted at 4,500 × g for 5 min and washed with 1 M MgSO₄. Mycelium (3 g [wet weight]) was added to 10 ml of filter-sterilized protoplasting mixture (1 M MgSO₄, 10 mM malate buffer [pH 5.8], and 1.5 mg of lysing enzymes [Plant Research International, Wageningen, The Netherlands] ml⁻¹). The suspension was gently shaken (10 rpm) for 2.5 h at 30°C, diluted twofold with sterile water, and incubated for an additional hour. Cell debris was removed by centrifugation for 2 min at 300 × g. The supernatant was incubated for 15 min at room temperature after the addition of 1 volume 1 M sorbitol and sieved through glass wool to remove remnants of cell debris. The protoplasts were pelleted by centrifugation for 15 min at 2,200 × g and carefully resuspended in 25 ml of 1 M sorbitol. After centrifugation for 10 min at 2,200 × g, protoplasts were divided into aliquots in 100-μl portions in 1 M sorbitol with 50 mM CaCl₂ to a final concentration of 10⁸ protoplasts ml⁻¹. Portions were frozen at a cooling rate of 1°C min⁻¹ using the freezing container Mr. Frosty (Nalgene, Lima, OH). For transformation, portions were thawed on ice, and 5 μg of circular plasmid DNA was added that was dissolved in 20 μl of TE (10 mM Tris [pH 8], 3 mM EDTA). After 15 min of incubation on ice, 1 volume of polyethylene glycol 4000, buffered with 10 mM Tris (pH 7), was slowly mixed with the protoplasts at room temperature and then incubated for 5 min. Regeneration medium (MM + 0.5 M MgSO₄) was added to a total volume of 3 ml. After overnight regeneration at 25°C, 7 ml of MM containing 1% low-melting-point agarose (at 37°C) was added. After mixing, regenerated protoplasts were spread on square plates containing 40 ml of solidified MM. Regeneration medium contained no antibiotic, 25 μg of phleomycin ml⁻¹, or 10 μg of nourseothricin ml⁻¹. Selection plates were incubated at 30°C for 5 days.

Isolation of genomic DNA and RNA.

Strains were grown on a porous polycarbonate membrane (diameter, 76 mm; pore size, 0.1 μm; Osmonics; GE Water Technologies, Trevose, PA) on MM with 1.5% agar for 4 days at 25°C. Colonies were harvested and ground in liquid nitrogen. Homogenized mycelium was lyophilized for RNA extraction. Then, 2 mg of mycelial powder was extracted with TRIzol reagent (Invitrogen/Life Technologies, Carlsbad, CA) according to the instructions of the manufacturer. RNA was purified by using the RNeasy kit (Qiagen, Valencia, CA), and RNA concentrations were estimated by using a nanodrop spectrophotometer. For DNA isolation, 200 mg of homogenized mycelium was mixed with 0.9 ml of extraction buffer (2% sodium dodecyl sulfate, 24 mg of 4-aminosalycilic acid ml⁻¹, and 20% 5× RNB [121.1 g of Tris-HCl, 73.04 g of NaCl, and 95.1 g of EGTA liter⁻¹; pH 8.5]). After incubation at 65°C for 15 min, the mixture was extracted with 0.4 ml of phenol-chloroform (1:1) and 0.5 ml of chloroform. The phases were separated by centrifugation for 10 min at 14,000 × g. The DNA was precipitated with 0.8 volume of isopropanol and pelleted at 14,000 × g for 10 min. After an overnight wash with 70% ethanol, the DNA was dissolved in 300 μl of TE containing 10 μg of RNase ml⁻¹.

Southern blotting.

DNA hybridizations were performed as described previously (31).

cDNA synthesis and real-time PCR analysis.

Total RNA was treated with RNase-free DNase to degrade contaminating genomic DNA. cDNA was synthesized by using random hexamer primers and Moloney murine leukemia virus reverse transcriptase according to the manufacturer's instructions (Fermentas, St. Leon-Rot, Germany). Real-time PCR was performed using an ABI Prism 7900HT SDS and Sybr green chemistry (Applied Biosystems, Foster City, CA). Expression levels were related to that of the actin gene ACT1 (accession no. AF156157) by using the Q-gene method that takes the efficiency of primer pairs into account (22). Primers were designed according the recommendations of the PCR master mix manufacturer (Applied Biosystems). ACT1 levels were determined with the primer sequences 5′-TGGTATCCTCACGTTGAAGTA-3′ and 5′-GTGTGGTGCCAGATCTT-3′, with an amplification efficiency of 1.93. The KU70 and RAD52 primer pairs consisted of 5′-CAGACGTGCCACTTGTT-3′/5′-CCCACAATGATGATCTTCTTCTTCT-3′ and 5′-GAAGAGTGGGCGGTTTA-3′/5′-CCTGCCCGTACCCAATA-3′, respectively. In both cases the amplification efficiency was 1.65.

RESULTS

Phleomycin is mutagenic to resistant S. commune strains.

The occurrence of morphologically aberrant colonies during growth on phleomycin-containing medium indicated that this compound was mutagenic to S. commune despite acquired resistance. To study this effect, a resistant strain (4-39P) was grown for 7 days at 30°C on MM plates with or without 25 μg of phleomycin ml⁻¹. Mycelial plugs were taken from four different spots from the resulting colonies and were cultured on MM plates. Colonies derived from mycelium that had been grown in the absence of phleomycin were morphologically indistinguishable from the wild type (Fig. 1, C1 to C4). In contrast, inocula taken from 4-39P colonies that had been grown in the presence of phleomycin developed morphologically aberrant colonies in three independent experiments. In one case three of four colonies showed a characteristic thin phenotype, and one showed the typical streak phenotype (Fig. 1, C5 to C8) (23). These phenotypes were stable and segregated 1:1 in a cross with a compatible wild-type strain. Morphological aberrant colonies were not observed when these experiments were repeated with the wild-type strain 4-39 and the phleomycin-resistant strain 4-39P using nonselective concentrations of phleomycin (0 to 5 μg ml⁻¹). These experiments show that phleomycin is mutagenic to resistant S. commune strains at selective concentrations of the antibiotic.

FIG. 1.

Colonies of S. commune resulting from inocula taken from mycelium of the phleomycin-resistant strain 4-39P that had been grown in the absence (C1 to C4) or presence (C5 to C8) of phleomycin.

Phleomycin increases transformation efficiency independently of the selection marker used.

Nourseothricin resistance is mediated by the nat1 gene of S. noursei. It has been used as a selection marker in several fungi, including the yeast Saccharomyces cerevisiae, the filamentous ascomycete Podospora anserina, and the basidiomycetous yeasts Cryptococcus neoformans and Ustilago maydis (8, 13, 14, 21). The coding sequence of nat1 was cloned between the S. commune GPD promoter and the SC3 terminator, resulting in plasmid pGEMNour. Incubating 10⁷ protoplasts of S. commune strain 4-39 with 5 μg of this vector routinely resulted in four transformants. However, transformation of 4-39 with pGEMPhleo, in which the nat1 gene is replaced by the ble gene of S. hindustanus, resulted in 10-fold more transformants. pNOURAD was constructed to determine what caused this difference in transformation efficiency. This vector contains both the nourseothricin and the phleomycin resistance cassettes. Transformation of 4-39 with pNOURAD, using phleomycin during regeneration, yielded similar numbers of transformants as with pGEMPhleo. This was irrespective of the antibiotic that was present in the selection plate (i.e., either phleomycin or nourseothricin). The construct pNOURAD was then introduced in strain 4-39 using 10 μg of DNA and 2 × 10⁷ protoplasts. After the addition of regeneration medium, the mixture was divided into four portions that were regenerated in the presence of phleomycin (25 μg ml⁻¹) or nourseothricin (8 μg ml⁻¹) or in the absence of either antibiotic (two portions). Protoplasts regenerated in the presence of phleomycin or nourseothricin were plated on MM containing the same antibiotic (Table 1). Protoplasts regenerated in the absence of antibiotic were plated on MM containing either phleomycin or nourseothricin. Protoplasts regenerated with phleomycin and nourseothricin yielded 64 and 7 resistant colonies, respectively (Table 1). Regeneration without antibiotics resulted in six transformants on phleomycin and four on nourseothricin. These results indicate that the presence of phleomycin during regeneration affects transformation efficiency. Strain 4-39 was transformed with pGEMNour to test whether this effect depends on the presence of the phleomycin resistance cassette. Regeneration in the absence of phleomycin, followed by selection on nourseothricin (8 μg ml⁻¹), resulted in three transformants. In contrast, 37 transformants were obtained when regeneration was performed in the presence of 25 μg of phleomycin ml⁻¹. Similar results were obtained when nonselective concentrations of 1 to 5 μg of phleomycin ml⁻¹ were used during regeneration. A 10-fold-increased transformation efficiency was also obtained when protoplasts of an uracil auxotrophic strain were transformed with plasmid pUBB20 (containing the URA1 gene with its own regulatory sequences) and regenerated in the presence of 1 to 25 μg of phleomycin ml⁻¹.

TABLE 1.

Number of transformants after introduction of pNOURAD in S. commune strain 4-39

Portion	Regenerationa	Plating	No. of transformants
1	Phleomycin	Phleomycin	64
2	Nourseothricin	Nourseothricin	7
3	-	Phleomycin	6
4	-	Nourseothricin	4

^aRegeneration was performed in the presence or absence (-) of phleomycin or nourseothricin, as indicated.

Phleomycin was added at different time points during regeneration of pGEMPhleo-transformed protoplasts to assess whether timing of addition of the antibiotic affects transformation efficiency. Addition of phleomycin at the start of regeneration or after 1 or 2 h yielded 42, 48, and 40 transformants, respectively. In contrast, 18, 12, and 6 transformants were obtained when the antibiotic was added 3, 4, and 8 h after the start of the regeneration. Apparently, phleomycin influences transformation efficiency, especially during the first 3 h of regeneration.

Phleomycin reduces the number of integration events.

The copy number of pNOURAD was analyzed in genomic DNA of transformants that had been regenerated in the absence or presence of 25 μg of phleomycin ml⁻¹. In the absence of the antibiotic, only 1 of 11 transformants had a single integration of the construct (Fig. 2A). The incidence of single integrations was increased sixfold when regeneration of protoplasts was performed in the presence of 25 μg of phleomycin ml⁻¹ (Fig. 2B). Reducing the amount of antibiotic to 15 or 3 μg ml⁻¹ gave results similar to those obtained at 25 μg of phleomycin ml⁻¹ (data not shown).

FIG. 2.

Southern analysis of transformants of S. commune regenerated in the absence (A) or presence (B) of 25 μg of phleomycin ml⁻¹. DNA was digested with XhoI that cuts once in the used construct. Blots were hybridized with a fragment of the GPD promoter, which is present both in the genomic DNA and in pNOURAD. The endogenous GPD fragment is represented by the lowest band on the blot. Arrows indicate strains with a single integration of the transforming construct.

Phleomycin upregulates key proteins from two DNA repair pathways.

Expression of S. commune orthologs of S. cerevisiae Rad52 and KU70 was determined to assess whether phleomycin affects expression of genes involved in homologous recombination (HR) and nonhomologous end joining (NHEJ), respectively. These genes were retrieved from the Joint Genome Institute S. commune sequence database. The genomic DNA of S. commune contained a sequence (JGI; augustus-scaffold_9.g449), which was 52% identical to the region spanning amino acid residues 54 to 206 of Rad52 of S. cerevisiae. Similarly, a 900-bp sequence (JGI; FGENESH1_PM.C_SCAFFOLD_9000006) showed 25% identity with KU70. Quantitative PCR showed that the predicted S. commune KU70 gene was upregulated (3.19 ± 0.28)-fold when mycelium had been exposed for 1.5 h to 25 μg of phleomycin ml⁻¹. Similarly, the predicted RAD52 was upregulated (2.11 ± 0.23)-fold.

DISCUSSION

Expression of the ble gene of S. hindustanus under the control of the S. commune GPD promoter and terminator confers resistance to phleomycin (30). Expression could not be detected when genes encoding hygromycin B phosphotransferase (hph) and aminoglycoside phosphotransferase (apt) were expressed flanked by the same promoter and terminator (29). This was also found for gentamicin acetyltransferase (acc1) (our unpublished results). Introduction of introns and the increase of GC content in AT rich regions (20) resulted in detectable mRNA levels of hph in S. commune. Nonetheless, selection on hygromycin remained problematic due to high background of nontransformed colonies (our unpublished results). The phleomycin resistance gene was thus the only efficient dominant selectable marker for S. commune at the time.

It is shown here that resistant colonies remain susceptible to mutagenesis by phleomycin. Inocula derived from a resistant strain developed aberrant morphologies when they were grown in the presence of a selective concentration of phleomycin. Mutants with a thin phenotype (25) predominated. They are characterized by a high, radial growth rate and their inability to produce normal aerial hyphae. In addition, fruiting body formation is abolished in dikaryons homozygous for the mutation (28, 32, 38). Transposition of the class II transposon scooter-2 into the thn1 gene, encoding a putative regulator of G protein signaling (RGS), was shown to be responsible for the thin phenotype (10). It can thus be that phleomycin is mutagenic not only because of its ability to make double-strand DNA breaks but also because it mobilizes transposons. The frequent occurrence of colonies with a thin phenotype might be due to a hot spot for integration of scooter-2 in the thn1 gene. In addition, the frequency may be explained by the fact that thin mutants overgrow the wild type (28). However, Southern analysis of phleomycin-treated colonies showed unique hybridization patterns of scooter-2 in three out of four cases. In contrast, the patterns were identical in untreated colonies (our unpublished data). This indicates that treatment with phleomycin causes rearrangements and/or transposition of scooter-2. Transposition as a consequence of DNA injury has been observed with transposons of various species (see the study by Arnault and Dufournel [1]).

Although the stoichiometry between bleomycins and the ble gene-encoded protein is 1:1 to inactivate the antibiotic, a sixfold excess of the resistance protein is needed for the protection of DNA in vitro (12). Taking this into account, 27 mg of the ble-encoded product liter⁻¹ should be produced to fully protect the S. commune DNA during phleomycin selection. This level of resistance protein accumulation is unlikely (our unpublished data). Therefore, it may well be that the genomic DNA of S. commune is not completely protected during selection, allowing mutations to occur.

Different candidates were considered in order to replace the phleomycin selection marker. One candidate, nourseothricin, is an aminoglycoside antibiotic produced by S. noursei that consists of a complex of the streptothricin sulfates C and D. This complex inhibits ribosomal protein synthesis (14). Several genes confer resistance to nourseothricin through acetylation, including the nat1 gene from S. noursei (19). The 573-bp nat1 gene lacks AT-rich regions and contains an overall GC content of 70.7%, which is similar to that of the ble gene of S. hindustanus (69.5%) that was shown to be functional in S. commune. Indeed, nat1 could be used to confer resistance to nourseothricin when placed under the control of the homologous GPD promoter and terminator of S. commune. Since nourseothricin is not mutagenic, we prefer to use this selection marker from now on.

It came as a surprise that the number of nourseothricin-resistant transformants was tenfold lower than that obtained with phleomycin as a selection marker. This was due to the fact that phleomycin increases the number of transformants during regeneration, in particular in the first 3 h of this process. Nonselective concentrations of phleomycin that were shown not to induce morphological mutants also had this effect, and it was independent of the selection marker that was used (uracil prototrophy or nourseothricin resistance). The presence of phleomycin during regeneration also reduced the copy number of the construct that was used for transformation. The effect of phleomycin thus strongly resembles restriction enzyme-mediated integration (REMI; see reference 18). REMI also results in a higher percentage of single-copy integration events and, in some organisms, is accompanied by an increase in the transformation frequency. However, in contrast to REMI, phleomycin does not depend on a restriction site.

Why low concentrations of phleomycin are nonmutagenic but do affect transformation efficiency is not yet understood. It is tempting to speculate that it has an effect on the DNA repair systems. HR and NHEJ are used by the cell to repair double-strand DNA breaks (33). It has been proposed that the proteins Rad52 (HR) and KU70/80 (NHEJ) function as “gatekeepers” for these pathways (16, 37). Expression of mus11, which is the orthologue of Rad52 in Neurospora crassa, can be induced by mutagens (26). In the case of S. commune it was shown that predicted KU70 and RAD52 genes were three- and twofold upregulated upon the addition of phleomycin.

We conclude from these findings that nonselective, nonmutagenic concentrations of phleomycin can be used to increase the number of transformants or to favor ectopic single integrations, as is needed for random tagged mutagenesis. The presented data also suggest that for gene targeting phleomycin should not be added to the regeneration medium.