Introduction

Agrobacterium tumefaciens, a plant pathogen, is widely used for the transformation of plants. Agr. tumefaciens has the unique ability to transfer DNA (so-called transferred DNA or T-DNA), which is located between two inverted repeats, known as the left border (LB) and right border (RB), to its host. The transferred DNA is transported into the host as a single-stranded DNA molecule. Once inside the host, the DNA is targeted to the nucleus where it randomly integrates into the host genome (for reviews, see Gelvin 2000; Zupan and Zambryski 1995). Agr. tumefaciens is able to transfer its DNA not only to plants, but also to other organisms, including yeast, filamentous fungi and human cells (Bundock et al. 1995; de Groot et al. 1998; Kunik et al. 2001). The number of fungi that has been shown to be transformed by Agr. tumefaciens is rapidly increasing and Agrobacterium-mediated transformation (AMT) has been shown to be a good alternative to protoplast-based transformation methods (Amey et al. 2002; de Groot et al. 1998; Fitzgerald et al. 2003; Meyer et al. 2003). To date, only a limited number of selection markers have been used, including the auxotrophic pyrG marker and dominant selection markers based on resistance to the antibiotics hygromycin, geneticin, or phleomycin (de Groot et al. 1998; Gouka et al. 1999; Pardo et al. 2002; Vijn and Govers 2003). Analysis of the T-DNA integration patterns of transformants obtained from various fungi has revealed that the T-DNA integrates at random and predominantly as a single copy (Abuodeh et al. 2000; Bundock et al. 2002; Covert et al. 2001; Degefu and Hanif 2003; Leclerque et al. 2003; Vijn and Govers 2003), indicating that fungal AMT can be used for insertional mutagenesis and for the generation of DNA-insertion mutant banks.

To optimize AMT for various fungi, several variables [e.g. the influence of co-cultivation period and temperature, the Agrobacterium/spore ratio, the addition or omission of the vir gene inducer acetosyringone (AS) in the Agrobacterium pre-culture] have been found to influence transformation frequency (Abuodeh et al. 2000; Combier et al. 2003; Leclerque et al. 2003; Malonek and Meinhardt 2001; Michielse et al. 2004b; Mullins et al. 2001; Rho et al. 2001). Furthermore, it has been shown that these environmental conditions also have an influence on T-DNA copy number. For example, the addition of AS to the Agrobacterium pre-culture can result in a decrease (or an increase) in single-copy T-DNA integration (Combier et al. 2003; Mullins et al. 2001; Rho et al. 2001). Prolongation of the co-cultivation period was shown to lead to multiple T-DNA integrations in Magnaporthe grisea (Rho et al. 2001). However, the length of the co-cultivation period had no influence on T-DNA copy number in Fusarium oxysporum (Mullins et al. 2001). Applying AMT for the generation of mutant banks, single-copy T-DNA integration is preferred as it enables linkage of an observed phenotype with a single alteration in the genome. For heterologous and homologous protein production in filamentous fungi, it has been shown that the introduction of multiple copies of the gene of interest results in an increase in protein production (Verdoes et al. 1995). Therefore, it would be beneficial to be able to generate and identify multicopy transformants obtained with AMT. Gouka et al. (1999) demonstrated the use of AMT in Aspergillus awamori for the generation of transformants carrying multiple copies of the gene of interest, using a binary vector carrying these multiple copies on its T-DNA.

A selection marker which is frequently used in protoplast-based transformation methods to generate transformants carrying multiple copies of the introduced DNA is the Asp. nidulans amdS selection marker (Hanegraaf et al. 1991; Kelly and Hynes 1985; Penttila et al. 1987; Rodriguez and Yoder 1987). In this study, we demonstrate the use of the Asp. nidulans amdS selection marker and AMT in Asp. awamori to generate and identify multicopy T-DNA transformants.

Materials and methods

Strains, plasmids and growth conditions

Asp. awamori CBS115.52 (CBS, The Netherlands) was used as a recipient strain for transformation. Agr. tumefaciens LBA1100 (pAL1100ΔT-DNA, Δtra, Δocc; Beijersbergen et al. 1992), carrying pUR5750AmdS (this study, Fig. 2a) or pSDMAmdSΔBB (Michielse, Arentshorst, Ram and van den Hondel, submitted for publication; Fig. 2b) was used for AMT and was grown in LB medium (Sambrook et al. 1989) containing spectinomycin (250 μg/ml) and kanamycin (100 μg/ml) at 28°C. Introduction of plasmids into LBA1100 was performed as described by Mattanovich et al. (1989). Escherichia coli XL1-Blue (Stratagene) was used for construction of pUR5750AmdS and was grown in LB medium containing kanamycin (25 μg/ml) at 37°C. The binary vector pUR5750AmdS, containing both the hygromycin and amdS selection markers, was constructed by inserting an XbaI fragment of p3SR2 (Corrick et al. 1987), corresponding to the amdS gene and its promoter and terminator sequences, into pUR5750 (de Groot et al. 1998) previously digested with XbaI.

AMT in Asp. awamori

AMT was performed as described by de Groot et al. (1998), with minor adjustments (Michielse et al. 2004b). Transformants were either selected on agar plates containing MM (Punt and van den Hondel 1992) supplemented with 100 μg/ml hygromycin or on agar plates containing acetamide as a sole nitrogen source (Kelly and Hynes 1985) with or without 100 μg/ml hygromycin, as indicated. All plates were supplemented with 200 μM cefotaxim to inhibit the growth of Agr. tumefaciens.

DNA isolation and Southern analysis

Fungal chromosomal DNA isolation and Southern analysis were performed as described by Kolar et al. (1988) and Michielse et al. (2004a), respectively. Chromosomal DNA was digested with BglII and probed with a 3.1-kb XhoI–HindIII fragment of pAN7.1 (Punt et al. 1987), corresponding to the hygromycin cassette, or with a 2.6-kb XbaI fragment of p3SR2 (Corrick et al. 1987), corresponding to the amdS expression cassette.

Results and discussion

The use of the dominant Asp. nidulans amdS selection marker (Hynes and Pateman 1970) to identify multicopy T-DNA transformants of Asp. awamori obtained with AMT was assessed. Asp. awamori was transformed with Agrobacterium strain LBA1100 carrying the binary vector pUR5750AmdS, which contains both the hygromycin and amdS expression cassettes on its T-DNA. Asp. awamori was also transformed with Agrobacterium strain LBA1100 carrying the binary vector pSDMAmdSΔBB, which contains only the amdS expression cassette on its T-DNA. Depending upon the binary vector used, transformants were selected on agar plates containing medium supplemented with hygromycin, hygromycin and acetamide, or only acetamide.

Selection of transformants obtained with Agr. tumefaciens carrying pUR5750AmdS on medium containing only hygromycin yielded the highest transformation efficiency (Table 1). Double selection of the transformants obtained with the same binary vector on acetamide medium containing hygromycin resulted in a six-fold decrease in the transformation frequency (Table 1). The lower transformation efficiency is most likely due to more stringent selection conditions to obtain transformants. It is known that the use of the amdS marker requires high expression levels of the amdS gene to allow growth on medium containing acetamide. High expression levels can be the result of either multicopy integration or integration at a certain locus, which allows efficient transcription (Hanegraaf et al. 1991; Kelly and Hynes 1985; Penttila et al. 1987; Rodriguez and Yoder 1987). The selection of transformants obtained with the binary vector which carried only the amdS expression cassette between the T-DNA borders (pSDMAmdSΔBB), using acetamide-containing agar plates, resulted in a low transformation frequency (Table 1). About 20 co-cultivations were performed, yielding only three transformants. Some background growth of Asp. awamori under these selection conditions was observed. It is likely that Asp. awamori contains a low level of endogenous acetamidase activity, which caused this background growth and complicated the identification of transformants. These putative transformants grew very slowly and each remained a fluffy colony of white mycelium with very few spores. It is possible that the identification of putative amdS positive transformants may be facilitated by performing the co-cultivation on black filter paper rather than on white nitrocellulose membranes, as both are equally efficient (Covert et al. 2001). Alternatively, the background growth might be reduced if fungal and bacterial cells are recovered from the filters after co-cultivation and subsequently spread onto selection media. This method has been shown to reduce the background growth of Beauveria bassiana in AMT (Leclerque et al. 2003). Since variation in co-cultivation conditions has been shown to influence transformation frequency and T-DNA copy number (Abuodeh et al. 2000; Combier et al. 2003; Leclerque et al. 2003; Malonek and Meinhardt 2001; Michielse et al. 2004b; Mullins et al. 2001; Rho et al. 2001), it might be possible to increase the transformation frequency obtained by AMT in Asp. awamori based on the amdS selection marker by using different co-cultivation conditions.

Table 1 Transformation frequency and T-DNA copy number obtained after different selection conditions. amdS Aspergillus nidulans acetamidase gene, hph hygromycin resistance gene
Full size table

It should be noted that the addition of AS to induce expression of the virulence genes of Agr. tumefaciens during co-cultivation is an absolute requirement to obtain transformants. Furthermore, all transformants obtained were mitotically stable (data not shown).

All transformants obtained were tested for their ability to grow on agar plates containing acetamide or hygromycin. A total of 25 [pUR5750AmdS] transformants, of which ten were obtained after selection on minimal medium with hygromycin (designated pUR5750AmdS-H) and 15 obtained after selection on acetamide medium supplemented with hygromycin (designated pUR5750AmdS-HA), were replica-plated onto agar plates containing either minimal medium supplemented with hygromycin or acetamide medium. All 25 transformants were able to grow on minimal medium containing hygromycin and no growth differences between the transformants were observed under these conditions (Fig. 1a,c). However, growth differences between the transformants were observed on acetamide medium (Fig. 1b,d). The ten pUR5750AmdS-H transformants remained small on acetamide medium and, even after prolonged incubation, these colonies did not grow further and did not sporulate. Seven of the 15 pUR5750AmdS-HA transformants (Fig. 1d, numbers 1, 2, 3, 6, 7, 9, 15; identified in caption) grew much faster on acetamide medium compared with the pUR5750AmdS-H transformants, suggesting that these transformants may contain more than one T-DNA copy. Also, the three [pSDMAmdSΔBB] transformants were able to grow on agar plates containing acetamide and no major growth differences between the three transformants were observed (data not shown).

Fig. 1a–d
figure 1

Growth phenotype of transformants grown on hygromycin and acetamide selection plates. a, b Ten pUR5750AmdS transformants were obtained after selection on hygromycin-containing agar plates (top row, from left to right: numbers 1–5; bottom row: numbers 6–10) and replica-plated onto hygromycin-containing agar plates (a) or acetamide-containing agar plates (b). c, d pUR5750AmdS transformants obtained after selection on acetamide- and hygromycin-containing agar plates (top row, from left to right: numbers 1–5; middle row: numbers 6–10; bottom row: numbers 11–16) and replica-plated onto hygromycin-containing agar plates (c) or acetamide-containing agar plates (d)

Full size image

Southern analysis was performed to determine the T-DNA copy number. Chromosomal DNA was digested with BglII and probed with hygromycin (hph) and amdS genes to determine single and multicopy T-DNA integration events. A single T-DNA integration event was expected to result in the detection of one fragment when probed with the hph gene (≥4,674 bp) and in two fragments (≥4,563 bp, ≥820 bp) when probed with the amdS gene (Fig. 2a). Based on previous studies, we know that 0–30% of the transformants contain more than one T-DNA copy when selected on medium containing hygromycin (Michielse et al. 2004a; Michielse, unpublished data). In this study, we found that all ten pUR5750AmdS-H transformants contained a single T-DNA integration. The Southern analysis result of a representative pUR5750AmdS-H transformant is shown in Fig. 2c,d (lane 1).

Fig. 2
figure 2

a Schematic representation of pUR5750AmdS T-DNA. LB Left border, RB right border. B BglII restriction site, Pgpd Asp. nidulans glyceraldehyde-3-phosphate dehydrogenase promoter, hph hygromycin resistance gene, TtrpC Asp. nidulans trpC terminator, amdS Asp. nidulans acetamidase enzyme, nptII neomycin/kanamycin resistance gene. b Schematic representation of pSDMAmdSΔBB T-DNA. c, d Southern analysis of [pUR5750AmdS] and [pSDMAmdSΔBB] transformants. Chromosomal DNA was digested with BglII and hybridized with a hygromycin (hph) probe (c) and amdS probe (d). Lane 1 pUR5750AmdS H-1 lane 2 pUR5750AmdS HA-1, lane 3 pUR5750AmdS HA-3, lane 4 pUR5750AmdS HA-2 lane 5 pSDMAmdSΔBB 6

Full size image

Of the 15 pUR5750AmdS-HA transformants, eight single, four double and three triple T-DNA integrations were found. The double T-DNA integration occurred as inverted repeats at either the LB repeat (see Fig. 2c,d; 9,348-bp fragment in c, two times in d, lane 2, fragments of ≥4,563 bp and ≥820 bp) or the RB repeat (see Fig. 2c,d; two times in c, ≥4,674-bp fragments, two times in d, lane 3, 9,126-bp and 820-bp fragments). In three of the 15 transformants, multiple fragments were found for the hph and the amdS probes, indicating a triple T-DNA integration as either tandem or inverted repeats (Fig. 2c,d, lane 4).

The transformants obtained using [pSDMAmdSΔBB] were also analyzed by Southern analysis to determine the T-DNA copy number (Fig. 2d, lane 5). With this vector, a single T-DNA integration would be expected to result in the detection of two fragments (Fig. 2a; ≥1,940 bp, ≥875 bp). This pattern was observed in two of the three transformants obtained (data not shown). Southern analysis of the third transformant revealed the presence of four hybridizing bands. Fragments of 1,750 bp and 2,815 bp were observed (Fig. 2d, lane 5), indicating inverted and tandem integration, respectively. Furthermore, two fragments of 8.5 kb and 8 kb (two times ≥1,940 bp) were observed. This hybridization pattern indicates that this transformant contains three T-DNA copies which are linked.

Based on the Southern analysis, the pUR5750AmdS-HA1, HA2, HA3, HA6, HA7, HA14 and HA15 transformants were identified as multicopy transformants. Six out of these seven transformants grew well on acetamide plates, indicating a good correlation between the Southern data and the growth phenotype. pUR5750AmdS-HA transformant 9 grew and sporulated well on agar plates containing acetamide medium. However, its sporulating phenotype was different from the other transformants (Fig. 1d). Southern analysis revealed that this transformant contained a single T-DNA copy (data not shown). It could be that, in this transformant, the location of the T-DNA integration into the genome resulted in this phenotype. Conversely, pUR5750AmdS-HA transformant 14 was identified as a multicopy transformant, based on Southern analysis (data not shown). However, on agar plates containing acetamide, this transformant did not grow better than the single copy pUR5750AmdS-H transformants (Fig. 1). This could be due to the location of the T-DNA integration into the genome, such that the amdS gene is not expressed highly enough to result in better growth of this transformant than the wild-type Asp. awamori strain on agar plates containing acetamide.

Based on our results, it can be concluded that the amdS marker can be used as a selection marker for AMT in Asp. awamori and that with this marker growth on agar plates containing acetamide correlates with the T-DNA copy number enabling the identification of multicopy T-DNA transformants.