A mus-51 RIP allele for transformation of Neurospora crassa

This report describes the construction and characterization of mus-51, an allele for high-efficiency targeted integration of transgenes into the genome of the model eukaryote Neurospora crassa. Two of the mus-51 strains investigated in this work (RZS27.10 and RZS27.18) can be obtained from the Fungal Genetics Stock Center. The two deposited strains are, to our knowledge, genetically identical and neither one is preferred over the other for use in Neurospora research.


Introduction
Non-homologous end joining, a DNA repair pathway that joins damaged DNA ends without regard for homology, was a hindrance to targeted transgene integration in N. crassa until Ninomiya et al. (2004) discovered that using an NHEJ mutant as a transformation host greatly increases the efficiency of this process.For example, one can achieve targeted transgene integration levels of nearly 100% by including either mus-51 Δ ::hph or mus-51 Δ ::bar in the transformation host's genetic background (Ninomiya et al., 2004;Colot et al., 2006).However, these alleles prevent one from using both hph and bar as selectable markers in consecutive transformations of the same host.Construction of a marker-free mus-51 null allele could eliminate this deficiency.Below, we describe the construction and characterization of a mus-51 null allele called mus-51 RIP70 .Additionally, we describe a method to distinguish between mus-51 + and mus-51 RIP70 genotypes by restriction endonucleasemediated digestion of PCR products.

Strains and growth conditions
Vogel's minimal medium (VMM) (Vogel, 1956) with or without histidine (0.5 g/L) was used for vegetative cultures.Synthetic Crossing Medium (SCM) (Westergaard and Mitchell, 1947) with or without histidine (0.5 g/L) was used for crosses.BDS medium (1% sorbose, 0.05% glucose, and 0.05% fructose) (Brockman and de Serres, 1963) with or without methyl methanesulfonate (MMS, 0.22 µl/ml) was used in a preliminary screen for mus-51 RIP alleles.BDS medium with hygromycin (200 μg/ml) and/or cyclosporin A (5 μg/ml) was used to screen for hygromycin and cyclosporin A resistance.Top and Bottom Agar were used in transformation experiments as previously described (Harvey et al., 2014).The key strains used in this study are listed in Table 1.

Plasmid construction
Plasmid pTH1256.1 was constructed by cloning the hph selectable marker from pCB1004 (Carroll et al., 1994) into the ApaI site of pBM61 (Margolin et al., 1997).Oligonucleotides P518 and P519 (Table 2) were then used to PCR-amplify a 2170 base pair (bp) fragment of mus-51 + .This PCR product was cloned into the NotI site of plasmid pTH1256.1 to create plasmid pSS2.12.Plasmid pSS2.12 thus contains the sequences necessary to insert the 2170 bp mus-51 + fragment next to the his-3 locus while converting his-3 to his-3 + .
Table 2. Oligonucleotides used in this study.

Isolation of mus-51 RIP70
HSS1.21.4 was crossed with FGSC 9716 by simultaneous inoculation of each strain to opposite sides of a 100 mm petri dish containing SCM plus histidine.The petri dish was incubated on a laboratory bench top for five weeks.Ascospores were collected, soaked in sterile water at 4 °C for over 24 hours, heat-shocked at 60 °C for 30 minutes, and plated onto VMM.Histidine-prototrophs were selected and screened for sensitivity to methyl methanesulfonate (MMS).The mus-51 coding region of an MMS-sensitive progeny named RSB1.70was analyzed by Sanger sequencing, found to be mutated, and named mus-51 RIP70 .Next, protoperithecia of F2-26 were fertilized with conidia from RSB1.70.This cross produced progeny RZS27.4,RZS27.6,RZS27.10, and RZS27.18.The mus-51 locus in each of these four strains was PCR-amplified with oligonucleotides P777 and P778 and the PCR products were sequenced by Sanger sequencing with oligonucleotides P699, P974, P975, P1001, P1050, and P1051.Sequences were analyzed with Bioedit 7.2.5 (Hall, 1999).The full sequence of mus-51 RIP70 can be obtained from GenBank under accession number KU860571.

Polymerase chain reaction (PCR)
Genomic DNA was isolated from lyophilized mycelia with the IBI Scientific Plant Genomic DNA Mini Kit.PCR was performed with New England Biolabs Phusion High-Fidelity DNA polymerase or MidSci Bullseye Taq DNA polymerase.When restriction endonuclease-mediated digestion of PCR products was needed to differentiate between two products of similar size, 2.5 μl of a completed PCR reaction was digested with a restriction endonuclease in a 25 μl reaction under standard conditions.

The csr-1 + gene deletion assay
A 3718 bp csr-1 + deletion vector was obtained by PCR-amplifying the csr-1 Δ ::hph locus from strain P8-65 with oligonucleotides P583 and P584.The PCR product was purified with an IBI scientific Gel/PCR DNA Fragment Extraction Kit and 500 ng were electroporated into conidia as previously described (Margolin et al., 1997) with selection for hygromycin resistance.Hygromycin-resistant transformants were then screened for resistance to cyclosporin A.

Results and Discussion
The mus-51 RIP70 allele is 84% identical to wild type Repeat-induced point mutation (RIP) introduces transition mutations into repeated DNA sequences within the nuclei of sexual cells just prior to meiosis (Cambareri et al., 1989;Selker, 1990).Therefore, we used RIP in an attempt to generate a mus-51 null allele at its native location on chromosome IV by first placing a 2170 bp fragment of mus-51 + next to the his-3 + locus on chromosome I.We then placed the resulting his-3 + ::mus-51 2170 -carrying transformant (HSS1.21.4) through a sexual cross with strain FGSC 9716.Strains mutated in mus-51 are sensitive to the DNA damaging agent methyl methanesulfonate (MMS) (Ninomiya et al., 2004).We thus selected RSB1.70, an MMS sensitive-progeny of HSS1.21.4 × FGSC 9716 (data not shown), for further analysis.Sanger sequencing confirmed RSB1.70 to carry a mutated mus-51 allele, which we named mus-51 RIP70 .
The mus-51 RIP70 allele contains 341 transition mutations spread over a 2134 bp region of chromosome IV (Figure 1).The mutations begin 211 bp before the start of the mus-51 + coding sequence and they end 255 bp before the end of the coding sequence (Figure 1).If we assume that all RIP mutations result from C to T transition events, 228 mutations must have originated on the coding strand and 113 mutations must have originated on the template strand (Figure 1, red lines).The high number of mutations suggests that mus-51 RIP70 is a null allele.Moreover, the mus-51 RIP70 allele encodes 30 early stop codons and over 100 amino acid substitutions relative to mus-51 + .

The mus-51 RIP70 allele increases the efficiency of targeted transgene integration
The efficiency of targeted transgene integration when using mus-51 RIP70 in the genetic background of a transformation host was measured with a csr-1 + -gene deletion assay.Deletion of csr-1 + enhances resistance to cyclosporin A (Bardiya and Shiu, 2007), allowing one to identify csr-1 Δ genotypes by screening on medium containing the compound.We transformed four strains, RZS27.4 (mus-51 + ), RZS27.6 (mus-51 + ), RZS27.10 (mus-51 RIP70 ), and RZS27.18(mus-51 RIP70 ), with a csr-1 Δ ::hph deletion vector.While nearly all mus-51 RIP70 transformants (99.5%) were resistant to cyclosporin A, only 37.5% of the mus-51 + transformants were resistant (Figure 2).To confirm that cyclosporin A resistance resulted from replacement of csr-1 + allele by the csr-1 Δ ::hph deletion vector, PCR was used to examine the csr-1 locus in a cyclosporin A-resistant and a cyclosporin Asusceptible transformant from each transformation host.While all of the cyclosporin A-resistant isolates carried the csr-1 Δ ::hph allele at the csr-1 locus, none of the susceptible isolates did (Figure 3).These results confirm that the mus-51 RIP70 allele can be used for high-efficiency targeted integration of transgenes in N. crassa.The mus-51 RIP70 allele improves the efficiency of transgene integration by homologous recombination.A csr-1 + deletion assay was performed with two mus-51 + strains (RZS27.4 and RZS27.6) and two mus-51 RIP70 strains (RZS27.10 and RZS27.18).Each strain was transformed with a csr-1 + deletion vector.The csr-1 Δ ::hph deletion vector replaces the entire coding region, as well as upstream regulatory sequences, with an hph selectable marker (Bardiya and Shiu, 2007).Only transformants resulting from replacement of csr-1 + with csr-1 Δ ::hph should be resistant to cyclosporin A. A total of 96 transformants were isolated, 24 for each of the transformation hosts.Conidial suspensions were prepared from each transformant and tested for growth on minimal BDS medium with or without hygromycin and cyclosporin (please note that hygromycin was an unnecessary addition because all transformants were previously shown to be resistant to hygromycin).Cyclosporin A resistance is indicated by the formation of large dense colonies with rough borders, while susceptibility is indicated by small thin colonies with smooth borders.A) A total of 6 out of 24 RZS27.4transformants were resistant.B) A total of 12 out of 24 RZS27.6 transformants were resistant.C) A total of 21 out of 23 RZS27.10 transformants were resistant.Strain 13 did not grow on minimal medium so it was not included.D) A total of 24 out of 24 RZS27.18transformants were resistant.Key: i) P6-07, ii) P8-65, iii) P6-07 and P8-65 [i.e., a suspension of conidia from both strains].Numbers 1-24 refer to the 24 transformants from each transformation host.Red font is used to indicate resistance to cyclosporin A.

The restriction endonuclease RsaI can distinguish between mus-51 RIP70 and mus-51 + alleles
One disadvantage of the mus-51 RIP70 allele is the inability to identify the allele by screening for growth on a common antibiotic.To address this issue, we devised a simple PCR-assay to distinguish between mus-51 RIP70 and mus-51 + alleles.In this assay, the mus-51 locus is PCR-amplified with oligonucleotides P974 and P975 and the resulting 478 bp PCR product is digested with the restriction endonuclease RsaI.The digested product is then analyzed by standard agarose gel electrophoresis.RsaI will only digest the mus-51 + PCR product (Figure 4).This procedure can also be combined with the conidial-PCR method of Henderson et al., (2005) for increased efficiency.

Figure 1 :
Figure 1: The mus-51 RIP70 allele contains 341 transition mutations.Positions 2,520,251 through 2,522,899 on chromosome IV of the N. crassa reference genome (version 12) are depicted in the diagram.The 2170 bp fragment, which was inserted next to the his-3 + locus on chromosome I, is delineated by black vertical lines labeled "transgene start >" and "> transgene end".The coding sequence of mus-51 + is marked with the black vertical lines labeled "ATG>" and ">TGA".The short vertical red lines denote the locations of C to T transitions on the coding strand (shown above chromosome) or the template strand (shown below chromosome).

Figure 4 :
Figure 4: The mus-51 RIP70 allele can be distinguished from mus-51 + by RsaI-based digestion of PCR products.A) Partial DNA sequences for mus-51 + and mus-51 RIP70 are shown.Note that oligonucleotides P974 and P975 anneal perfectly to both alleles (red arrows).Although these primers amplify a PCR product of identical length from mus-51 + and mus-51 RIP70 , an RsaI recognition site (5' GTAC 3'), which exists in mus-51 + but not mus-51 RIP70 (red rectangle), can be used to distinguish the PCR products.B) Oligonucleotides P974 and P975 were used to PCR-amplify the mus-51 region from the 11 strains described in Figure3B.The PCR products were digested with RsaI and analyzed by agarose gel electrophoresis.The mus-51 + allele produces two fragments with sizes of 369 and 109 bp, but the 109 bp fragment is often too faint to detect without additional gel staining.The mus-51 RIP70 allele produces a single fragment of 478 bp.Lanes are labeled as in Figure3B.

Table 1 .
Strains used in this study.