Keywords

1 Introduction

Aspergilli are a unique group of saprophytic filamentous fungi that dwell in wide spectrum of habitat such as soil, decaying organic matter, and moist indoor environments. Enormous nutritional flexibility underlies their successful lifestyle. The genus Aspergillus is gifted with rich ability to secrete degradative enzymes, organic acids and secondary metabolites. These fungi are capable of growth in a wide range of pH (between pH 2.0 and pH 11.0), temperature (10–50 °C) and high osmolarity (Kis-Papo et al. 2003). These endowments have fascinated both academic and industrial researchers for over a century. While A. nidulans is an extensively used genetic model organism and an academic favorite, many other Aspergilli are industrial workhorses (Meyer 2008). For instance, A. niger is an avid citric acid producer (Karaffa et al. 2001); A. oryzae is used to brew sake and make soy sauce (Barbesgaard et al. 1992); A. terreus is employed for the production of itaconic acid (Okabe et al. 2009) and lovastatin (Bizukojc and Ledakowicz 2009). Some of them like A. niger and A. oryzae are choice platforms for heterologous protein production due to their exceptional secretion capacity (Su et al. 2012). In contrast, the genus also includes pathogens like A. fumigatus, A. parasiticus, and A. flavus.

Many whole genome sequences now available for Aspergilli are indicative of their acknowledged value. Genome sequences of several industrially and medically important Aspergillus species are freely accessible (Arnaud et al. 2012; Gibbons and Rokas 2013). This has enabled a better appreciation of gene expression, metabolism, and its regulation. Understanding the organization, regulation, and manipulation of fungal genes requires the development of various genetic tools. Due to the industrial importance of these fungi many of such tools may have remained trade secrets. Low transformation efficiency, limited choice of selection markers, and poor frequency of targeted gene insertions are largely responsible for the delay in development of genetic engineering toolkit for these organisms.

A. nidulans was the first Aspergillus to be successfully transformed and this involved the use of protoplasts (Tilburn et al. 1983). Subsequently other DNA delivery methods were described along with modifications to suit individual fungal strains (for details see other chapters of this book). Each of these methods has its own advantages and disadvantages (Prabha and Punekar 2004). Regardless of these variations, the transforming DNA either integrates into the host genome or replicates autonomously. With rare exceptions, integrative transformation is predominant in filamentous fungi including Aspergilli. Selectable markers are necessary and are used to score transformants (and DNA integration events) in a background of the recipient host. The repertoire of convenient selection markers available for Aspergillus transformation is limited. While some markers work well across Aspergilli others are species specific. Several earlier reviews have stressed on various transformation techniques and development of strains with different genetic backgrounds, to study filamentous fungal biology (Fleissner and Dersch 2010; Jiang et al. 2013; Kuck and Hoff 2010; Lubertozzi and Keasling 2009; Meyer 2008; Meyer et al. 2011; Ruiz-Diez 2002; Su et al. 2012; Prabha and Punekar 2004; Weld et al. 2006). However, the description of selectable markers for Aspergillus transformation has attracted limited attention. A comprehensive and up-to-date account on various selection markers and different strategies exploiting these markers in manipulating Aspergilli is given here.

2 Selectable Markers for Aspergillus Transformation

Selectable markers are one of the important molecular tools that distinguish transformed cells from the untransformed host. They generally fall into three categories: resistance markers, nutritional markers, and bidirectional markers. Besides, a few reporter genes also serve as convenient visual markers to select transformants. Selectable markers functionally tested in Aspergilli are described below. Three specific markers, namely bar, agaA, and sC as examples relating to A. niger transformation, are particularly emphasized.

2.1 Resistance Markers

Resistance markers allow the growth of an organism in the presence of inhibitors (antibiotic or antimetabolite). These are routinely used to transform wild type/natural isolates of fungal strains; they circumvent the need to create an auxotrophic host background. The only requirement is that the recipient organism is sensitive to the selection pressure applied. These markers are particularly useful for organisms with little genetic information available. Resistance markers by and large are dominant and provide a tool for positive selection—the gene imparting a survival phenotype to the recipient on the respective selection medium. Table 11.1 provides a list of such selectable markers reported till date for Aspergilli; of these, hph, bar, and benA are more commonly used. Interestingly, hsv-1 tk (coding for herpes simplex virus type 1 thymidine kinase) provides for a negative selection—its expression is lethal to the host organism when grown in the presence of ganciclovir. Lethality is due to the conversion of ganciclovir into a cytotoxic nucleoside analogue. In combination with ble, hsv-1 tk was used to provide both positive and negative selection in A. fumigatus (Krappmann et al. 2005).

Table 11.1 Resistance (dominant) markers in Aspergilli
Full size table

Resistance to l-phosphinothricin (PPT) (through the bar/pat gene) was first used as selection marker in Neurospora crassa (Avalos et al. 1989). Inhibition of glutamine synthetase and subsequent accumulation of ammonia are thought to be the major mode of PPT action. Other cellular targets of PPT comprise membrane depolarization (Ullrich et al. 1990) and membrane transport process (Trogisch et al. 1989). The bar gene imparts resistance by acetylating PPT; it serves as a dominant selection marker as many fungi are sensitive to PPT (Ahuja and Punekar 2008). Selection of A. niger transformants using bar gene constructs was achieved with both a homologous promoter (A. niger PcitA, Dave and Punekar 2011) and a heterologous promoter (A. nidulans PtrpC, Ahuja and Punekar 2008). Besides bar transformants, spontaneously resistant A. niger colonies were obtained. Analysis of this spontaneous resistance, caused by mutations in the A. niger glutamate uptake system, was helpful in fine-tuning bar selection conditions. While host sensitivity to PPT may be an issue, the virtues of this marker are many. PPT is an active ingredient of several commercial herbicides (such as Basta, Bayer Crop-Science, India) and is readily accessible. Due to lower toxicity to humans, affordable cost and PPT availability, bar selection is attractive. The compact size of bar (CDS around 0.5 kb) allows construction of expression vectors with larger inserts for genome engineering. The bar marker was successfully employed in A. niger: to select for EGFP expression (Dave and Punekar 2011), for deletion analysis of A. niger citA promoter (Dave and Punekar 2011), and to disrupt agaA from A. niger (see next section for details). It is also applied in introducing ldh (lactate dehydrogenase), gbuA (4-guanidinobutyrase), and adc (arginine decarboxylase) expression constructs in this fungus (unpublished work).

Several other resistance markers (not listed in Table 11.1) are available for filamentous fungi, but so far they have not been tested in Aspergilli. These include bsd (blasticidin S resistance in Rhizopus niveus; Yanai et al. 1991), sur (sulfonylurea resistance in Penicillium chrysogenum and Magnaporthe grisea, Sweigard et al. 1997) markers and more recently nat (nourseothricin resistance in N. crassa, Cryphonectria parasitica and P. chrysogenum; Smith and Smith 2007; Hoff et al. 2010; de Boer et al. 2013) and ergA (terbinafine resistance in P. chrysogenum; Sigl et al. 2010).

Some of the disadvantages of resistant markers which may limit their utility include poor selection with few markers due to background growth, random integration of the marker gene into the host genome, some of the markers like oliC are species specific, some agents are mutagenic to the fungus (i.e., phleomycin), dominant selection can be too harsh requiring a nonselective phase before cultivating under selective pressure, and restricted availability and toxicity of the antibiotic/antimetabolite required for selection. Most importantly, dominant marker genes are susceptible to horizontal transfer in the environment. It is therefore highly desirable to eliminate the resistance gene from the genetically modified fungus before its large scale use.

2.2 Nutritional Markers

Nutritional markers are usually developed from metabolic pathways. The selection involves the complementation of an auxotrophic strain with the corresponding wild type allele. Development of auxotrophic recipient Aspergillus strains is often tricky due to the nutritional versatility of these organisms, poor understanding of their metabolism, and restricted range of available genetic tools. The nutritional markers described so far in Aspergilli are listed in Table 11.2. Markers like argB and riboB are commonly deployed in Aspergillus transformation. Disruption of argB (coding for ornithine carbamyltransferase of arginine biosynthesis pathway) leads to arginine auxotrophy and this is exploited as a transformation marker by complementing a matching argB − strain (Table 11.2). Either an argB mutant or a deliberately argB disrupted recipient is used (Lenouvel et al. 2002). Insertion of multiple copies of the bacterial aspA gene relieved the nutritional deficiency in NAD-glutamate dehydrogenase negative A. nidulans (Hunter et al. 1992). Thus, the aspA marker is unique in providing an efficient means of selecting multi-copy transformants.

Table 11.2 Nutritional markers in Aspergilli
Full size table

Arginase (encoded by agaA) catabolizes l-arginine to l-ornithine and urea; this defines the only route for arginine utilization in most fungi. The A. niger agaA gene when disrupted using a PtrpC-bar cassette (by homologous recombination) results in an arginase negative phenotype. The D-42 strain (agaA::bar) is therefore unable to grow on arginine as sole nitrogen source (Dave et al. 2012). While agaA mutants are reported from N. crassa (Morgan 1970) and A. nidulans (Bartnik et al. 1977), the A. niger D-42 strain is the first example of targeted agaA disruption in filamentous fungi. Combination of D-42 strain and arginase expression construct (PcitA-agaA) defines agaA as a novel nutritional marker for A. niger transformation. Details of applying this selection marker may be found in an accompanying protocol (see Chap. 39), while the strategy to identify agaA transformants is shown in Fig. 11.1. Depending on the integration event, the transformants obtained on arginine plates may either be agaA + bar + (ectopic) or agaA + bar − (homologous). Such an agaA/bar combination describes a two-way selection with a potential for marker reuse.

Fig. 11.1
figure 1

Strategy for agaA based selection of transformants in A. niger. The PcitA-agaA gene construct complements and confers growth phenotype on arginine to D-42 strain (agaA::bar). Two types of transformants are expected depending upon the nature of integration event

Full size image

The inability of A. niger D-42 strain (agaA:: bar) to grow on arginine is consistent with a single pathway for arginine utilization. Both biochemical and bioinformatics approaches support the absence of a functional arginine decarboxylase in A. niger. But a pathway to catabolize agmatine was shown to exist in this fungus (Kumar 2013). Introducing a functional arginine decarboxylase is expected to confer the ability to utilize l-arginine to the D-42 mutant. Therefore, arginine decarboxylase could become another novel marker based on fungal arginine catabolism. Initial attempts to express bacterial arginine decarboxylase in A. niger D-42 strain for this purpose, have been unsuccessful (unpublished). However, as in S. cerevisiae, an alternative route for polyamine biosynthesis was reconstituted by introducing E. coli arginine decarboxylase (Klein et al. 1999); we still hope to develop this gene as a successful marker for A. niger.

2.3 Bidirectional Markers

At times it is advantageous if the same marker can be used for both gain and loss of function. All such two-way selection markers reported to date happen to be nutritional markers (Table 11.3). Such markers are called bidirectional markers as they provide a two-way selection. pyrG and amdS are frequently used bidirectional markers. As is typical for bidirectional markers, pyrG allows both positive and negative selection—pyrG mutants are auxotrophic for uridine/uracil but are resistant to 5-fluoroorotic acid (5-FOA). The cloned amdS gene was first used as homologous transformation marker in A. nidulans (Tilburn et al. 1983). The selection is based on utilization of acetamide by amdS expressing transformants. For many Aspergilli that cannot utilize acetamide, amdS based selection works directly in the wild type background. In this sense, amdS represents a nutritional marker which is dominant. Since amdS transformants become sensitive to fluoroacetamide, amdS was also exploited as a bidirectional marker (Michielse et al. 2005). Bidirectional markers are convenient tools as they can be repeatedly exploited in the same host. This feature is particularly useful when the range of available markers is restricted. The list of bidirectional markers available for fungi, and for Aspergilli in particular, however is limited (Table 11.3).

Table 11.3 Bidirectional markers from Aspergilli
Full size table

As a bidirectional marker pyrG is convenient and is frequently used. The availability and cost of 5-FOA is sometimes a concern. sC is superior as a marker in this regard. ATP sulfurylase (encoded by sC) activates inorganic sulfate to form adenosine 5-phosphosulfate (APS) (Fig. 11.2a). The sC gene sequence is highly conserved in Aspergilli (Varadarajalu and Punekar 2005) making it easy to use across species. Since ATP sulfurylase is the first committed step in sulfate assimilation, sC mutants are incapable of using sulfate as source of sulfur. At the same time these mutants display selenate (a toxic analogue of sulfate) resistance. Both sC − (selenate resistant) and sC + (sulfate utilization) phenotypes can thus be selected for. Homologous transformation with the sC marker was first reported in A. nidulans (Buxton et al. 1989) and later in A. fumigatus (De Lucas et al. 2001). A heterologous sC gene was used to transform A. niger (Buxton et al. 1989) and A. oryzae (Yamada et al. 1997). A mutant (sC −) background is a prerequisite to use sC marker. Spontaneous sC mutants are easily isolated on selenate media as they are resistant to selenate. Mutations in the sB gene (encoding sulfate permease) could also result in selenate resistance. However, only sC mutants are chromate sensitive and can be distinguished from sB mutants (Fig. 11.2b, c). Although sC has a long CDS, the high degree of sequence conservation (across Aspergilli) at this genomic locus (Varadarajalu and Punekar 2005) could in principle be used for marker assisted homologous integration events.

Fig. 11.2
figure 2

Selection of sC mutants. (a) Sulfate assimilation pathway in Aspergillus, (b) strategy to distinguish sC − and sB − mutants, and (c) spontaneous A. niger sC − mutant (one out of 106 spores spread per plate)

Full size image

2.4 Visual Marker Systems

Some reporter genes may also serve as selection markers since they allow the transformants to be visually distinguished. Examples of well-known reporters include lacZ (β-D-galactosidase acting on X-Gal), gusA (β-glucuronidase acting on X-Gluc), and laccase genes. They were used as reporters in A. nidulans (Kanemori et al. 1999) and A. parasiticus (Miller et al. 2005). Heterologous laccases were used as reporters in A. nidulans and A. niger (Mander et al. 2006). Since then various laccase genes from A. niger genome have been cloned and functionally annotated (Ramos et al. 2011). Laccases oxidize artificial substrates like ABTS (2,2-azino-di-(3-ethylbenzthiazoline sulfonate)), ADBP/DMA (4-amino-2,6-dibromophenol/3,5-dimethylaniline), and DMPPDA (N,N-dimethyl-p -phenylenediamine sulfate) thereby help locate the transformants on agar plates containing these substrates. The green fluorescence protein (GFP) and its variants have been extensively used as reporters in Aspergilli (Jiang et al. 2013; Nitsche et al. 2013). However, utility of GFP as a visual marker is constrained because scoring transformants requires their exposure to UV light. In a rare example, the bar-egfp protein fusion gene was created to combine selectable and visible markers for Beauveria bassiana (Jin et al. 2008).

3 Selection Markers in Genome Manipulation Strategies

Transformation with the help of different selectable markers enables genetic manipulation of fungi to study various cellular metabolic processes. Gene knock-in (gene insertion, replacement, or over-expression) and knock-out (gene deletion) approaches have led to the elucidation of many metabolic/gene functions. Industrial strain development may require introduction of one or more steps (genes) of a metabolic pathway. Multiple selectable markers or multiple use of the same marker often becomes a necessity. For instance, a quadruple auxotrophic host (niaD − sC − ΔargBadeA −/adeB −) was developed to construct more auxotrophic strains of A. oryzae (as also other deuteromycetes wherein sexual crossing is impossible) (Jin et al. 2004). Transformation in Aspergillus is almost always an integrative event. Therefore one is quickly restricted by the availability of fresh markers. This limitation may be overcome by devising suitable strategies and judicious use of available markers. Aspects of selectable marker exploitation are highlighted below and include locus specific (homologous) DNA integration events for gene disruption, deletion, or insertion as well as marker rescue.

Homologous recombination is necessary for targeted integration of the DNA along with the selectable marker. Homologous recombination occurs by a single crossover event (type I integration) resulting in the insertion of foreign DNA (and the marker) into the target locus or by a double crossover event (type III integration) leading to replacement of target gene by selection marker. Both these strategies are extensively employed with minor modifications. However, homologous recombination events are rare while ectopic integrations (illegitimate recombination, type II integration) are more common in Aspergilli. Frequency of homologous recombination increases when longer homologous flanking sequences, corresponding to the target locus, are used. While 30–50 bp flanking sequences serve well in S. cerevisiae (Hua et al. 1997), much longer flanking sequences (0.5–2.0 kb) are required in Aspergilli (Meyer 2008). The frequency of homologous integration was significantly improved when the components of the nonhomologous end-joining (NHEJ) pathway (of DNA recombination) were deleted in N. crassa (Ninomiya et al. 2004). This was extended to Aspergilli by marker-assisted knock out in A. nidulans (nkuA::argB; Nayak et al. 2006), A. niger (kusA::amdS and kueA::pyrG; Honda et al. 2011; Meyer et al. 2007), A. oryzae (ku70/80::ptrA; Takahashi et al. 2006), A. fumigatus (ku80::pyrG; Krappmann et al. 2006), and A. sojae (ku70/80::ptrA;Takahashi et al. 2006). This approach has also enabled the genome-wide deletion project in A. nidulans (Meyer et al. 2011).

3.1 Split Marker

The split marker technique, initially developed in S. cerevisiae (Fairhead et al. 1996), has found application in many filamentous fungi (Kuck and Hoff 2010). In this strategy, instead of a complete marker sequence, two partially overlapping fragments of the marker are used. The fragments are so designed that the functional marker is generated only after a recombination event (Fig. 11.3a). This technique requires two PCR products (generated either by two step PCR or by ligation fusion PCR), each comprising a fusion of flanking sequence of the target gene and an appropriate partial–inactive fragment of the marker DNA. The two PCR products have internal sequence overlaps such that a complete marker CDS is generated after recombination. On transformation, three crossover events are required for homologous integration of the split marker replacing the target gene (Kuck and Hoff 2010). The probability of this event increases when the fragments are in close proximity and when they are at the desired locus. Ectopic integration of either of the fragments of such a split marker will be nonfunctional and hence those transformants are not selected in the screen. The split marker approach significantly improves the chances of homologous over ectopic integrations; but the transformation frequency decreases notably as three recombination events are required. An hph split marker was used to disrupt pyrG and acuB genes of A. niger (Nielsen et al. 2007); interestingly, pyrG was later rescued (also see Sect. 3.3 below) by eliminating the hph marker. Disruption of A. fumigatus genes involved in trehalose biosynthetic pathway (tpsA and tpsB, Al-Bader et al. 2010) and also of photoreceptors (lreA and fphA, Fuller et al. 2013) was achieved through the split marker strategy.

Fig. 11.3
figure 3

Use of selection markers in fungal genome manipulation. Strategies for split marker (a), dual selection (b), and marker rescue (c). [T flanking homologous regions of the target gene, M1 positive selectable marker, M2 negative selectable marker, black bars direct DNA repeats]

Full size image

3.2 Dual Selection

Dual selection strategy makes use of two markers—one for positive and another for negative selection. This allows one to distinguish between homologous (positive selection) and ectopic (negative selection) integrations of a gene-cassette. The positive selection marker (often resistance or prototrophy) is designed to be locus specific while the other marker (often lethality or auxotrophic) provides for negative selection. The ectopic transformants express both the markers whereas a targeted gene replacement results in the loss of negative marker and is amenable to positive selection (Fig. 11.3b). Although tested in many fungi (such as Gardiner and Howlett 2004; Khang et al. 2005), dual selection is reported in only two species of Aspergilli. The niaD, areA, and the tannase genes of A. sojae were individually disrupted by dual selection using pyrG and oliC markers (Takahashi et al. 2004). Homologous integration at pyrG locus of A. awamori was achieved through the hph and amdS twin marker system (Michielse et al. 2005).

3.3 Marker Rescue

The list of markers at our disposal is not extensive and therefore recycling/reusing a marker is of essence. The issue is particularly acute when the same recipient has to be transformed repeatedly. The problem was cleverly addressed by the advent of a “blaster cassette” tool for sequential deletion of multiple genes in yeast. This method allows the rescue and reuse of a single selectable marker, mostly a bidirectional marker. The yeast URA-blaster cassettes are tripartite sequences consisting of URA3 gene flanked on both sides by direct DNA repeats (Alani et al. 1987; Fonzi and Irwin 1993). Transformants become prototrophic for uracil upon URA3 integration. Forced excision of URA3, through recombination between the two direct repeats, is achieved in the presence of 5-FOA. The desired locus continues to remain disrupted as one copy of the direct repeat is left behind (Fig. 11.3c). This rescues the URA marker and the organism is rendered auxotrophic for uracil. The URA-blaster cassette is again available for the next round of gene integration/disruption.

A URA3 homolog was used to develop the corresponding blaster cassette (pyrG-blaster) for Aspergilli. The pyrG-blaster was successfully used to disrupt rodA gene of A. fumigatus (d’Enfert 1996) and lacA and glaA genes of A. niger (Storms et al. 2005). Deletion of A. nidulans aroC was achieved with a pyrG-blaster module that combines ET cloning (RecE and RecT mediated recombinogenic engineering approach) to rescue the marker (Krappmann and Braus 2003). Marker rescue based on the cre/loxP recombination system is also available for Aspergilli. Here, a loxP direct DNA repeat is placed on both sides of pyrG marker. The Cre recombinase recognizes and catalyzes reciprocal recombination between the pair of lox repeats, thereby rescuing the marker. Marker rescue using cre/loxP system was used to disrupt pabaA and veA loci of A. fumigatus (Krappmann et al. 2005) and ligD gene of A. oryzae (Mizutani et al. 2012). Successive disruption of yA and wA genes was achieved in A. nidulans using cre/loxP blaster cassettes (Forment et al. 2006). As an improvement, a self-excising marker cassette that employs the prokaryotic β-rec/six site-specific recombination system was adopted to recover the ptrA marker after disrupting abr2 and pksP genes of A. fumigatus (Hartmann et al. 2010).

Marker rescue strategy may also be used to transiently disrupt a gene. As noted above, NHEJ disruption favors homologous recombination. However, such NHEJ disrupted strains display increased sensitivity towards DNA damaging conditions such as γ irradiation (Meyer et al. 2007). Restoring the NHEJ function after the required genetic manipulation is therefore desirable. Transient disruption and subsequent recovery of NHEJ function was demonstrated in A. nidulans. The pyrG-blaster module served to disrupt nkuA; required genetic manipulation was done and finally nkuA function was regained by recombining the flanking direct repeats of nkuA itself (Nielsen et al. 2008).

4 Conclusions

The importance of Aspergilli as an industrially and medically important group of fungi is well established. A growing interest in both basic and applied research on Aspergilli acknowledges this fact. Many whole genome sequences are now available and await exploitation. Both genome annotation and strain development require genetic manipulations. Selectable markers are central to these objectives. A range of selection markers are now available to engineer Aspergillus genomes. More are being developed based on the knowledge of the fungal genomes. Selectable markers find direct applications in strain improvement programs for industrially important Aspergilli. Strategies for safe and judicious use of available markers through locus specific integration, marker rescue, and self-excision subsequent to site-specific recombination continue to emerge.