An efficient marker recycling system for sequential gene deletion in a deep sea-derived fungus Acremonium sp. HDN16-126.

Acremonium species are prolific producers of therapeutic molecules which include the widely used beta-lactam antibiotic, cephalosporin. In light of their significant medical value, an efficient gene disruption method is required for the physiological and biochemical studies on this genus of fungi. However, the number of selection markers that can be used for gene targeting is limited, which constrain the genetic analysis of multiple functional genes. In this study, we established a uridine auxotrophy based marker recycling system which achieves scarless gene deletion, and allows the use of the same selection marker in successive transformations in a deep sea-derived fungus Acremonium sp. HDN16-126. We identified one homologue of Acremonium chrysogenum pyrG (also as a homologous gene of the yeast URA3) from HDN16-126, designated as pyrG-A1, which can be used as a selection marker on uridine free medium. We then removed pyrG-A1 from HDN16-126 genome via homologous recombination (HR) on MM medium with 5-fluoroortic acid (5-FOA), a chemical that can be converted into a toxin of 5-flurouracil by pyrG-A1 activity, thus generating the HDN16-126-△pyrG mutant strain which showed auxotrophy for uridine but insensitivity to 5-FOA and enabled the use of exogenous pyrG gene as both positive and negative selection marker to achieve the scarless deletion of target DNA fragments. We further applied this marker recycling system to successfully disrupt two target genes pepL (encodes a putative 2OG-Fe (II) dioxygenase) and pepM (encodes a putative aldolase) identified from HDN16-126 genome, which are proposed to be functional genes related to 2-aminoisobutyric acid metabolism in fungi. This work is the first application of uridine auxotrophy based scarless gene deletion method in Acremonium species and shows promising potential in assisting sequential genetic analysis of filamentous fungi.

Introduction

Acremonium sp. is a group of filamentous fungi widely distributed in soil, sea water and on decaying plant material in nature [1], [34]. As the marked [beta]-lactam antibiotic of cephalosporin C was first discovered from Acremonium species, this genus has been regarded as one of the most pharmaceutically important filamentous fungi following Penicillium species [[2], [3], [4]].

Chemical and pharmaceutical studies on Acremonium spp. have proven that they are also prolific producers of other secondary metabolites including steroids, terpenoids, metroterprnoids, polyketides, alkaloids and peptides [[5], [6], [7]]. However, the vast majority of the biosynthetic machinery of those metabolites as well as physiological and biochemical effectors of this genus are still obscure. This is partially due to the lack of efficient genetic manipulation system and the limited availability of selection markers for transformation.

As so far, marker recycling system and scarless deletion methods are widely adopted to circumvent the limitation of selection markers [9,10], among which, the pyrG-based marker recycling system is often used in eukaryotic microorganism [11]. PyrG/URA3 orthologues are genes that encode orotidine-5′-monophosphate decarboxylases (OMPdecase) or orotidylate decarboxylase that catalyze the decarboxylation of orotidine 5'-monophosphate to form uridine monophosphate [12,13]. This family of enzymes also could convert 5-fluoroorotic acid (5-FOA), an analogue of the uracil precursor, to 5-fluorouracil, a toxic compound that inhibits DNA and RNA synthesis [12], [13]. Therefore, a pyrG or URA3 gene could be used as a feasible counter-selection marker when it is positioned between homologous sequences, excision of pyrG/URA3 cassette by homologous recombination can be selected with 5-FOA added in medium. Indeed, the pyrG/URA3-based counter-selection system has been successfully applied in Saccharomyces cerevisiae [15], Aspergillus spp.[16], [35], Candida albicans [18], Neurospora crassa [19], Talaromyces versatilis [20], Mucor circinelloides [21] and Colletotrichum orbiculare [22]. However, this system has never been applied to the genus of Acremonium species.

During our continued work in searching pharmaceutical fungal strains from Deep-Sea environments, one fungal strain Acremonium sp. HDN16-126 was obtained from surface sediment sample collected from Mariana Trench. Chemical studies on this strain afforded several antibacterial structures including prenylphenols, macrolides, diketopiperazines and peptaibols (data not published), which indicated talented metabolic capacity of HDN16-126. To facilitate further biochemistry studies on the strain HDN16-126, we established a uridine auxotrophy based scarless gene deletion system. In this method, the native pyrG (pyrG-A1) in HDN16-126 was first deleted from chromosome using a hyg cassette to generate the △pyrG strain, which enables the employment of pyrG-A1 gene as a counter-selection marker. This △pyrG strain was then subjected to take up a gene knock out cassette carrying pyrG-A1 flanked by three homologous arm fragments. After two rounds of screening on uridine free medium (pyrG-A1 worked as complementary for the uridine auxotrophy) and MM medium supplemented with 5-FOA (pyrG-A1 worked as counter-selection marker), the cassette was first knocked-into and then knocked-out from the chromosome, resulting in the recycling of the pyrG-A1 marker and scarless deletion of target genes. Here we will describe the details of the marker recycling system together with its successful application in disrupting two target genes pepL and pepM, which are proposed to be related to 2-aminoisobutyric acid metabolism in fungi.

Materials and methods

Strains and culture conditions

The fungal strain HDN16-126 was isolated from Mariana Trench sediment sample (−8000 m) and identified as Acremonium sp. based on the internal transcribed spacer (ITS) region of rDNA sequence (GenBank no. MK543173). The strain was cultured on PDA solid medium (BD/Difco), the spores were collected and stored at −80 °C in 20% glycerol.

Saccharomyces cerevisiae BJ5464 [23] was used for yeast homologous recombinant, Escherichia coli DH10B was used for standard cloning (purcased from Solarbio, Beijing) (Table 1).

Table 1

Strains and plasmids used in this study.

Strainsor plasmids	Characteristics	Sources
Strains
Acremonium sp. HDN16-126	Isolated from sediment sample collected from Mariana Trench	Preserved in our laboratory
Saccharomyces cerevisiae BJ5464	Recombination host of the knock out fragments	Preserved in our laboratory
Escherichiacoli DH10B	Cloning host	Solarbio life sciences
Plasmids
pYEU	Recombination vector of the knock out fragments	Preserved in our laboratory
pHyg	Hygromycin resistance gene carrier	Preserved in our laboratory
pYEU-pyrG	pYEU carring pyrG-up, pyrG-dn, hyg, pyrG gene fragments	This study
pYEU-pepL	pYEU carring pepL-up, pepL-dn, pyrG, pepL gene fragments	This study
pYEU-pepM	pYEU carring pepM-up, pepM-dn, pyrG, pepM gene fragments	This study

Bioinformatic analysis of target genes

The genome of Acremonium sp. HDN16-126 was sequenced at Beijing Genomics Institute (BGI) Co. Ltd (Qingdao, China). OMPdecase (GenBank: P14017.1) from the strain Acremonium chrysogenum [24] was selected as a query to identify pyrG gene from the genome of Acremonium sp. HDN16-126 using NCBI local blastp tool (https://www.ncbi.nlm.nih.gov/books/NBK52637/). Similarly, two target genes pepL and pepM which are postulated to be 2-aminoisobutyric acid metabolism relating genes were identified through local blastp tool by using the amino acid sequence of TqaL and TqaM as queries from Penicillium aethiopicum [25].

The sequences relationship of those genes with their orthologues were further predicted and analyzed by ClustalW alignments [26] and cytoscape software [27,28].

Hygromycin and 5-FOA sensitivity assay and uridine complementation assay

Agar dilution method was adopted to identify the lethal doses and proper working concentrations of hygromycin and 5-FOA. Spores of HDN16-126 were inoculated on PDA medium supplemented with hygromycin at gradient concentrations (0, 25, 50, 100, 200, 300 μg/mL) and on MM solid medium (1% glucose, 5% nitrate salts, 0.1% trace elements) supplemented with 5-FOA at gradient concentrations of (0, 0.5, 1.0, 1.5, 2.0 and 2.5 mg/mL), respectively. To confirm the proper dose of uridine for the auxotrophic mutants required to restore their growth, the △pyrG mutant type (MT) strains were inoculated on MM medium with 5-FOA and different concentrations of uridine (0, 1, 2.5, 5, 10 mM) in the second-round selection process. Each culture inoculated with equivalent spores and incubated at 28 °C for 4 days, all experiments were conducted in triplicate and cultures with the same volume of solvent worked as controls.

Preparation the protoplast of HDN16-126

HDN16-126 was cultured in PDA solid medium at 28 °C for 5 days, and the spores were collected and stored at −20 °C in 20% glycerol (4 × 105/mL). 1 mL spores were then inoculated into 40 mL PDB + YE liquid medium and incubated at 28 °C, 180 rpm for about 10 h. Mycelia was harvested by centrifugation at 4000 rpm for 15 min at 4 °C. The supernatant was removed and the mycelia was washed with 15 mL Osmotic buffer (1.2 M MgSO4·7H2O, 10 mM sodium phosphate, pH 5.8) and centrifuge at 4000 rpm for 15 min at 4 °C. Then the sediment was re-suspended with 10 mL Osmotic buffer containing 0.04 g Yatalase and 0.06 g Lysing Enzymes from Trichoderma (Takara Bio). The resulting mixture was transferred in a 50 mL sterilized conical flask. The flask was incubated at 28 °C, 80 rpm overnight, then the culture liquid was transferred into 50 mL sterile centrifuge tube and added isovolumetric trapping buffer (0.6 M sorbitol, 0.1 M Tris-HCl, pH 7.0) softly, the mixture was centrifuged at 4000 rpm at 4 °C for 15 min. Then the protoplast layer was obtained and re-suspended in STC buffer (1.2 M sorbitol, 10 mM CaCl2, 10 mM Tris-HCI, pH 7.5) with 2x volume, followed by centrifuge at 4000 rpm at 4 °C for 15 min. The supernatant was discarded and the protoplast was harvested and stored in 1 mL STC buffer for transformation.

Generation of pyrG-A1 deletion mutant

Yeast homologous recombination [29] was used for the construction of pyrG knock out cassette (Fig. S2). The cassette was designed to carry four fragments: upstream (left flank sequence of pyrG-A1 gene's promoter), downstream (right flank sequence of pyrG-A1 gene's terminator), hyg (hygromycin resistance gene) and pyrG-A1 (partial sequence of pyrG-A1 expression cassette). For the construction, hyg gene fragments with its promoter and terminator were PCR amplified from the pHyg plasmid using primer pair of hyg-F/hyg-R and pyrG-A1 fragments together with its upstream and downstream flanking sequences amplified from genomic DNA of HDN16-126 using primer pairs of pyrg-F/pyrg-R, pyrg-up-F/pyrg-up-R, pyrg-dn-F/pyrg-dn-R (Table S1), respectively. To facilitate yeast homologous recombination and assemble these DNA fragments into one construct, 30-bp homology regions were added to primers thus generating short overlap sequences between these fragments and the cloning vector pYEU (Fig. S1). The cloning vector pYEU was linearized by SpeI/PmlI and co-transformed together with above four fragments into S. cerevisiae BJ5464 by using Frozen-EZ Yeast Transformation II Kit (Solarbio, Beijing). Transformants were selected on semisynthetic medium (dropout uridine) and the constructed plasmid pYEU-pyrG cassette was rescued from yeast cells by Zymoprep™ II-Yeast Plasmid Miniprep and re-transformed into E.coli DH10B by CaCl2-transformation method. The propagated plasmids were finally extracted from E.coli and verified by PCR and sequencing.

The pyrG-A1 knock out cassette was PCR amplified from plasmid pYEU-pyrG using primers of kz-pyrg-F/kz-pyrg-R and transformed into the protoplast by the polyethylene glycol (PEG)-CaCl2 method [30]. 20 μL PCR product of pyrG-A1 knock out cassette was added into 100 μL protoplast, mixed gently and placed on ice for 1 h, then 600 μL PEG was added and the mixture was incubated at room temperature for 30 min to accomplish the first round of homologous recombination. The reaction product was cultivated on PSA plates (PDA plates supplemented with 218.6 mg/mL sorbital) supplemented with hygromycin at 28 °C for 3–4 days, and recombinants were picked up and verified by PCR using primer yz-hygMT-F/yz-hygMT-R. The correct recombinants were further screened on MM medium plus uridine and 5-FOA, where the uridine auxotrophy mutants (HDN16-126-△pyrG) were picked up, followed by PCR verification using the primer pair of yz-pyrgup-F1/yz-pyrgdn-R1 and yz-pyrgup-F2/yz-pyrgdn-R2 (Fig. 2C).

Fig. 2

Construction of markerless △pyrG mutant. (A) Morphology of the WT spores with spores in dormancy period, spores in germination and protoplasts of the WT; (B) Flow chart of construction procedure of △pyrG mutant; (C) PCR verification of the MT, the length of PCR product from MT was about 0.7 kb, and the length of PCR product from WT was about 1.8 kb; (D) Growth of △pyrG on MM medium, MM plus uridine (10 mM) medium and MM plus 5-FOA and uridine medium, the WT on MM medium was used as a control.

Generation of pepL and pepM deletion mutant

To tested whether uridine auxotrophy could be applied to scarless gene deletion, pepL and pepM were selected as target genes for disruption. The pepL disruption cassette was designed to contain four fragments: upstream and downstream of pepL amplified by the primer L-UP-F/L-UP-R and L-DN-F/L-DN-R; the pepL gene fragment amplified by the primer pair of L-F/L-R; and the pyrG-A1 together with promoter (~1.5 kb) and terminator (~800 bp) amplified by primer pair of PYRG-F/PYRG-R (Table S1). Four fragments were integrated into the plasmid pYEU using yeast homologous recombination method as described in this paper with the order as upstream-downstream-pyrG-pepL, thus generating pYEU-pepL (Fig. S1). The pepL knock out cassette was PCR amplified from plasmid pYEU-pepL using primers kz-L-F/kz-L-R and transformed into the protoplast of HDN16-126-△pyrG by the polyethylene glycol (PEG)-CaCl2 method as described before for homologous recombination. After one round of screening on uridine free medium and one round of screening on medium with 5-FOA, the mutant strain with scarless deletion of pepL (△pepL) was obtained. The △pepM mutant was generated by the same method as above, both mutants were verified by PCR primers (yz-L-F/yz-L-R for △pepL and yz-M-F/yz-M-R for △pepM).

Results

Bioinformatic analysis of pyrG, pepL and pepM

PyrG, pepL and pepM homologous sequences were identified by NCBI local blastp program. The 1104 bp open reading frame encoding a protein of 368 amino acids was identified from HDN16-126 and designated as pyrG-A1 based on the high similarity with OMPdecase from Acremonium chrysogenum (81.71% identity). Protein interaction analysis of OMPdecase was conducted using A. chrysogenum, Trichoderma sp., Fusarium sp. and other fungal species as references. As shown in Fig. 1A, OMPdecase-HDN16-126 clustered with homologues from A. chrysogenum and several other species, which showed a high possibility of the similarity in functions among those genes (E value: 50).

Fig. 1

Protein interaction network of pyrG-A1 (A), pepL (B) and pepM (C).

The gene of pepL encoding a putative 2OG-Fe (II) dioxygenase and pepM encoding a putative aldolase were identified based on their similarities to Penicillium aethiopicum derived TqaL (57.95%) and TqaM (67.32%), respectively. The protein interaction network as shown in Fig. 1B and C denoted their functional similarities with the homologous genes from reference filamentous strains (E value: 50).

Antibiotic sensitivity of Acremonium sp.

Wild type Acremonium sp. (WT) was subjected to agar dilution test with hygromycin (0–300 μg/mL) and 5-FOA (0–2.5 mg/mL). Growth of WT could be totally inhibited on PDA medium with 50 μg/mL of hygromycin and MM medium with 0.5 mg/mL of 5-FOA, which determined the working concentration of 50 μg/mL for hygromycin and 0.5 mg/mL for 5-FOA in transformants and mutants selection (Fig. S6A).

Generation of pyrG-A1 mutant

A 6732 bp pyrG-A1 knock out cassette was amplified from plasmid pYEU-pyrG (Fig. S4) and dissolved in 20 μL STC buffer with final concentration of 400 ng/μL. Protoplast harvested from 5-day culture of HDN16-126 was transformed with above pyrG-A1 knock out cassette solution. After the first round of recombination (Fig. 2B), 12 mutants were picked up from the PSA plus 50 μg/mL hygromycin selective medium and all the mutants were verified by PCR to be the correct clones with the hyg gene cassette integrated into the correct locus by homologous recombination (Fig. S7). Strikingly, the homologous recombination rate in this step was much higher than literature reports for other filamentous fungi where a big portion of wrong insertions were generated by non-homologous recombination [31]. Then the 12 mutants were further selected on MM plates plus uridine and 5-FOA (0.5 mg/mL), where the hyg gene and pyrG-A1 gene would be removed from the genome with 5-FOA working as a counter selection marker. After this second round of selection, one pyrG-A1 deficient strain HDN16-126-△pyrG was obtained and verified by PCR test. As shown in Fig. 2C, the length of the PCR product for the mutant strain was about 0.7 kb and for the wild type strain was about 1.8 kb, which confirmed the successful removal of pyrG-A1 from HDN16-126 chromosome.

The HDN16-126-△pyrG strain was unable to grow on MM, but grow on MM supplemented with 10 mM uridine (Fig. 2D). This strain also showed low sensitivity to 5-FOA treatment, which is in consistent with the absence of the pyrG-A1 cassette. In addition, the morphology of mycelium showed no significant difference compared with that of the WT except for a slightly darker pigmentation on the mycelium (Fig. 2D). Together, we concluded that by two rounds of selection with hygromycin and 5-FOA treatments, the pyrG-A1 gene could be scarlessly knocked out from HDN16-126 genome.

Generation of pepL and pepM mutant

With the uridine auxotroph mutant HDN16-126-△pyrG in hand, we continued to test whether the marker recycling system could be used for sequential disruption of multiple genes. Two genes of pepL and pepM which are proposed to be 2-aminoisobutyric acid metabolism relating genes were selected as targets for disruption. The pepL and pepM disruption cassettes were designed as shown in Fig. 3A. After first round of selection on MM-sorbital dropout uridine medium, the cassettes with pyrG-A1 gene flanked by three homologous fragments were integrated into the upstream next to pepL and pepM respectively. Each 12 single clones were picked up from pepL and pepM disruption mutants, within which the expression of pyrG-A1 from the cassette could complement the uridine auxotrophy and restore the cell growth. Then all the clones were transferred to MM plus uridine and 5-FOA medium for dropping out of pyrG-A1. 10 out of 12 clones for △pepL and 8 out of 12 clones for △pepM were grown. All the mutants were verified to be correct by PCR using primer yz-L-F/yz-L-R or yz-M-F/yz-M-R (Fig. 3B), where the length of PCR products from △pepL and △pepM mutants were 1105 bp and 1078 bp respectively, while as control, the PCR products from WT were 2460 bp and 2399 bp in correlation with pepL and pepM. The △pepL, △pepM mutants were then tested for their growth on MM with or without uridine (10 mM) and results showed they could only grow on uridine supplied MM medium and compared with WT strain, both △pepL and △pepM mutants showed less spore production and much darker pigmentation on mycelium (pink for mutants and pale for WT, Fig. 3C). The disruption efficiency for gene pepL and pepM were 83.3% and 66.7%, respectively, and these data demonstrate that the marker recycling system can be applied to genetic manipulation of Acremonium species for generating scarless gene deletion and replacement mutations.

Fig. 3

Construction of pepL and pepM scarless deletion mutants. (A) Two rounds of homologous recombination and selection on MM-sorbital dropout uridine medium and MM with 5-FOA medium; (B) PCR verification of △pepL and △pepM mutant; (C) Growth of △pepL, △pepM on MM plus uridine (10 mM) medium compared with WT strain.

Discussion

In general, traditional homologous recombination techniques are commonly used to generate gene disruption and replacement [32] in fungi. During this process, resistant marker will be retained on the chromosome after one round of recombination, making it impossible to use the same selection marker during the next cycle of gene manipulation, which means increasingly more selection markers are required for completing multiple gene disruptions. Consequently, the physiological and biochemical studies of many fungal strains are limited by the availability of suitable selection markers.

In this study, we report the establishment of a uridine auxotrophy based scarless gene deletion method in Acremonium sp. HDN16-126, a fungus collected from the Mariana Trench (−8000 m). Importantly, the pyrG-A1 marker was successfully recycled from the chromosome, which allows the use of the same selection marker in successive transformations. The pyrG-A1-based marker recycling system showed several advantages over other gene deletion systems, such as sit-directed recombination methods including Flp-FRT and Cre-loxP recombinations, which have been widely used in prokaryotes and eukaryotes [22]. Those tools could easily conduct the excision of target DNA between two repeated specific sites, while for a legacy, one of the repeated sites is left on the chromosome, which makes it an unstable factor as the site specific sequences accumulated in the genome. Compared to above methods, the pyrG-A1-based marker recycling system could create much more stable multiple knock-out mutants.

Using this pyrG-A1 based marker recycling system, sequential disruption of two target genes pepL and pepM were achieved in HDN16-126-△pyrG strain. To our best knowledge, this is the first report of the application of uridine auxotrophy based marker recycling system in deep sea-derived Acremonium species. In addition, the reintroduction of pyrG-A1 could restore the growth of HDN16-126-△pyrG mutant to wild-type levels, which indicated the potential of this system to access disruption and function analysis of any gene of interest in filamentous fungi. With the help of this efficient gene manipulation system, more in depth physiological and biochemical investigations including the pepL and pepM functions in amino acids metabolism are at the right time in progress.

CRediT authorship contribution statement

Ruonan Sun: Experiment operation, Writing – original draft. Hengyi Xu: Experiment operation. Yanyan Feng: and. Xuewen Hou: Formal analysis, and chemical study of strains. Tianjiao Zhu: and. Qian Che: and. Blaine Pfeifer: Checking and confirming all procedures. Guojian Zhang: and. Dehai Li: Designed the study, supervised the laboratory work, and, contributed to the critical reading of the manuscript.

Declaration of competing interest

The authors declare that they have no conflicts of interest.