Isolation of rfk-2 UV , a mutation that blocks spore killing by Neurospora Spore killer-3.

Neurospora Spore killer-3 ( Sk-3 ) is a selfish genetic element that kills spores to achieve gene drive. Here, we describe the isolation and mapping of rfk-2 UV , a mutation that disrupts spore killing. The rfk-2 UV mutation is located 15.6 cM from mus-52 on Chromosome III. The significance of this discovery with respect to Sk-3 evolution is discussed. Copyright:

(A) Spore killing and gene drive are present in SkS (ISU-3037) × Sk-3 (ISU-3291) crosses. Left: Asci possess a spore killing phenotype. Asci develop asynchronously in N. crassa . Asci with dark pigmented ascospores are more mature than those with light pigmented ascospores. Right: Offspring were randomly collected and examined for Sk-3 gene drive with a PCR-based gene drive assay. All 34 offspring possess the Sk-3 genotype, demonstrating that Sk-3 gene drive occurs in this cross. Lanes: K, Sk-3 control, ISU-3291; S, SkS control, FGSC 10340; two lanes contain a DNA ladder, the remaining 34 lanes correspond to 34 offspring. (B) Spore killing and gene drive are absent from SkS (ISU-3036) × Sk-3 rfk-2 (ISU-4684) crosses. Left: Spore killing is absent from asci. Right: Of the 34 offspring examined, there are 18 with an Sk-3 genotype and 16 with an SkS genotype, demonstrating that rfk-2 disrupts gene drive (χ 2 = 0.12, p-value = 0.73). (C) The Sk-3 interval is on Chromosome III. The relative positions of the centromere, hph , hph , and mus-52 Δ are shown in the diagram. (D) Left: A three-point cross was performed with strains Sk-3 rfk-2 (ISU-4685/6) and Sk-3 mus-52 Δ (ISU-4689). Right: A total of 213 offspring were collected and genotyped for rfk-2 hph and mus-52 Δ alleles. Recombination analysis indicates that rfk-2 is located centromere-proximal of hph and mus-52 Δ. “Y” means that the indicated allele is present in a genotype. Abbreviations: SCO1, genotypes result from a crossover between rfk-2 and hph ; SCO2, genotypes result from a crossover between hph and mus-52 Δ; DCO, genotypes result from a double crossover. (E) Left: A three-point cross was performed with strains Sk-3 rfk-2 (ISU-4687/8) and Sk-3 mus-52 Δ (ISU-4689). Right: A total of 186 offspring were collected and genotyped for rfk-2 hph and mus-52 Δ alleles. Recombination analysis indicates that rfk-2 is located centromere-proximal of hph and mus-52 Δ.

Description

Spore killer-3 ( Sk-3 ) is a selfish genetic element that was discovered over four decades ago in the filamentous fungus Neurospora intermedia (Turner and Perkins 1979). Sk-3 is interesting because it is transmitted to nearly all offspring of an Sk-3 × SkS cross, where Sk-3 refers to a strain carrying the selfish genetic element, and SkS refers to a strain that is sensitive to Sk-3 -based spore killing. The biased transmission of Sk-3 is an example of gene drive that occurs through spore killing (Zanders and Johannesson 2021). Specifically, during an Sk‑3 × SkS cross, spore killing eliminates ascospores (offspring) with an SkS genotype while sparing those with an Sk‑3 genotype. As a result, Sk-3 × SkS crosses produce asci with four viable and four inviable ascospores, rather than the eight viable ascospores typical of Neurospora crosses.

Sk-3 has been mapped to a 30 cM interval of Chromosome III (Turner and Perkins 1979). This interval contains hundreds of genes and it is transmitted to offspring as a single unit (Campbell and Turner 1987). At least two genes within the interval are thought to mediate gene drive. One gene is rsk, which is required for resistance to spore killing but not for spore killing itself (Hammond et al. 2012). The second gene has yet to be identified, but it is believed to encode Sk‑3 ’s killer (Hammond et al. 2012).

Sk‑3 is one of two complex selfish genetic elements known to exist in Neurospora fungi. The second is Sk‑2 (Turner and Perkins 1979). Sk‑2 shares many similarities with Sk‑3 . Sk-2 is transmitted to offspring in a biased manner, resides on a similar interval of Chromosome III, and uses rsk for resistance to spore killing but not spore killing itself. Despite these similarities, Sk‑2 and Sk-3 are distinct elements. For example, Sk-2 ’s rsk allele ( rsk ) provides resistance to spore killing by Sk-2 but not Sk-3 , and Sk-3 ’s rsk allele ( rsk ) provides resistance to spore killing by Sk‑3 but not Sk-2 (Hammond et al. 2012).

A recent finding suggests that some of the similarities between Sk-2 and Sk-3 , such as their complex genomic rearrangements, may have evolved by convergent evolution (Svedberg et al. 2018). Other similarities, such as the role of rsk in the drive mechanisms of both Sk-2 and Sk-3 , appear to be the result of descent from a common ancestral selfish genetic element. However, a complete understanding of the evolutionary relationship between Sk-2 and Sk-3 will likely require additional knowledge, such as the identity of Sk-3 ’s killer. The Sk-2 killer is encoded by rfk-1 and spore killing is absent in Sk‑2 rfk-1 Δ × Sk crosses (Rhoades et al. 2019). In contrast, deletion of the most likely rfk-1 ortholog from an Sk-3 strain had no effect on spore killing, leaving the identity of Sk-3’s killer unknown (Svedberg et al. 2018).

Here, to help identify Sk-3 ’s killer, we performed a genetic screen for required for killing ( rfk ) mutations (see methods). The genetic screen uses Sk‑3 rsk Δ × SkS crosses, which abort development before the production of viable ascospores (Hammond et al. 2012; Harvey et al. 2014). We isolated a few candidate rfk mutations with our genetic screen and chose the most promising candidate, rfk-2 , for additional analysis. As demonstrated in Figure 1 (A and B), rfk-2 disrupts spore killing and gene drive.

To determine the approximate genomic location of rfk-2 , we performed two sets of three-point crosses (Figure 1C). Recombination analysis of 213 offspring from the first set of crosses ( rfk-2 hph × mus-52 Δ) indicates that rfk-2 is located 2.8 cM from hph and 15.0 cM from mus-52 Δ (Figure 1D). For the second set of crosses ( rfk-2 hph × mus-52 Δ), recombination analysis of 186 offspring indicates that rfk-2 is located 7.5 cM from hph and 16.1 cM from mus-52 Δ (Figure 1E).

In addition to providing genetic distances from physical positions on Sk-3 Chromosome III, our recombination data indicate that rfk-2 is located centromere-proximal of hph , hph , and mus‑52 Δ (Figure 1, D and E). This is somewhat surprising given that Sk-2 rfk-1 is located at the junction of Sk-2 and SkS sequences on the right arm of Chromosome III (Rhoades et al. 2019). rfk-1’s location within Sk-2 allows it to escape inactivation by a genome defense process called meiotic silencing by unpaired DNA (MSUD) (Aramayo and Selker 2013; Hammond 2017; Rhoades et al. 2019), and thus, given the importance of rfk-1 's location, we initially predicted that Sk-3 's killer gene would be found centromere-distal of hph , hph , and mus‑52 Δ. Our finding that rfk-2 is centromere-proximal to all three of these genetic markers indicates that Sk-2 and Sk-3 may have evolved different relative positions for their killer genes. In summary, the future cloning and characterization of rfk-2 should help clarify the organizational patterns of critical gene drive genes within Sk-2 and Sk-3 , as well as the evolutionary relationship between these two complex selfish genetic elements.

Methods

Strains and alleles used in this study

Sk-3 was introgressed into N. crassa for genetic analysis shortly after its discovery in N. intermedia (Turner and Perkins 1979). Only N. crassa strains were used in the present study. The rid genotype suppresses a genome defense process called RIP, which mutates duplicated sequences during sexual reproduction (Freitag et al. 2002; Aramayo and Selker 2013). mus-51 Δ, and mus-52 Δ alleles suppress NHEJ, thereby increasing the efficiency of genetic transformation (Ninomiya et al. 2004). The sad-2 Δ allele inhibits MSUD, which suppresses the expression of unpaired genes during meiosis (Aramayo and Selker 2013; Hammond 2017). The his-3 and leu-1 genes are required for histidine and leucine biosynthesis, respectively, and fl controls macroconidiation (Perkins et al. 2000).

Culture conditions and ascus analysis

Vegetative propagation was performed on VMM/VMA and crosses were performed on SCA as previously described (Harvey et al. 2014; Rhoades et al . 2020). For imaging asci, syringe needles were used to dissect asci from perithecia into 50% glycerol at 18 days post fertilization. Asci were imaged by standard light microscopy.

Screen for

To isolate Sk‑3 rfk‑2 , we made one change to a previously developed screen for Sk-2 rfk mutations (Harvey et al. 2014). Specifically, we irradiated Sk‑3 rsk Δ conidia (from strain ISU-4677) instead of Sk-2 rsk Δ conidia. We then followed the protocol as previously described by fertilizing Sk protoperithecia with the UV irradiated conidia, incubating the mating cultures for four weeks, collecting shot ascospores from the lids of crossing plates, geminating ascospores on VMA, and transferring individual germlings (offspring) to culture tubes containing VMA for vegetative propagation. Each offspring was genotyped for Sk-3 and examined for an ability to kill spores in crosses with an Sk mating partner (strains F2-23, F2-26, ISU-3036, and/or ISU-3037). Offspring with an Sk-3 genotype that displayed defects in spore killing were considered rfk mutant candidates. The rfk-2 mutation was first identified in strain MAV214. The following series of crosses was used to move rfk-2 from MAV214 into strain ISU-4684: Cross 1) MAV214 × ISU-3036 = ISU-4678; Cross 2) ISU-4678 × ISU-4679 = ISU-4681; Cross 3) ISU-4681 × F2-23 = ISU-4682; Cross 4) ISU-4682 × ISU-3291 = ISU-4683; and, Cross 5) ISU-4683 × ISU-3291 = ISU-4684.

Genetic modifications

A standard electroporation-based transformation procedure was used to make genetic modifications to N. crassa (Margolin et al. 1997; Rhoades et al. 2020). All transformation vectors were constructed by double-joint (DJ)-PCR (Yu et al. 2004; Hammond et al. 2011), using oligonucleotide PCR primers. Sk-3 genomic DNA was used for amplification of left and right DJ-PCR fragments. The genome sequence of Sk-3 strain FGSC 3194 (Svedberg et al. 2018) was used for primer design and Chromosome III position information. Plasmid pTH1256.1 was used for amplification of hph center fragments for DJ-PCR (GenBank: MH550659.1). Plasmid pNR28.12 was used for amplification of nat center fragments for DJ-PCR (GenBank: MH553564.1). Both plasmids can be obtained from the Fungal Genetics Stock Center (McCluskey et al. 2010). Strain ISU-3291 and transformation vectors v14b, v260, and v337 were used to create rsk Δ ::hph , leu-1 Δ ::nat , and mus-52 Δ ::nat alleles, respectively. Strain ISU-4684 and transformation vectors v322 and v324 were used to create hph and hph alleles, respectively. Primers for v14b construction: 1001b/1002b (center), 1003b/1004b (left), 1005b/1006b (right), and 1007b/1008b (nested). Primers for v260 construction: 297/298 (center), 1907/1908 (left), 1909/1910 (right), and 1911/1912 (nested). Primers for v322 construction: 585/586 (center), 2204/2158 (left), 2159/2160 (right), and 2161/2162 (nested). Primers for v324 construction: 585/586 (center), 2169/2170 (left), 2171/2172 (right), and 2173/2174 (nested). Primers for v337 construction: 297/298 (center), 2205/2214 (left), 2215/2208 (right), and 2216/2217 (nested).

PCR-based assay for gene drive

Genomic DNA was isolated from offspring and control strains with IBI Scientific’s mini genomic DNA kit for plants and fungi. PCR primer set 49/50 amplifies a 596 bp product from Sk-3 genotypes and an 896 bp product from SkS genotypes. PCR products were examined by standard agarose-gel electrophoresis with ethidium bromide staining.

Three-point crosses

The position of rfk-2 was mapped relative to three markers: hph , hph , and mus-52 Δ. hph was created by inserting the hygromycin resistance cassette ( hph ) between genes ncu05694 and ncu05695 at approximately 1.0 Mb on Chromosome III of ISU-4684. hph was created by inserting hph between genes ncu07875 and ncu07876 at approximately 1.6 Mb on Chromosome III of ISU-4684. mus-52 Δ was created by replacing mus-52 in strain ISU-3291 with nat , a nourseothricin resistance cassette. Offspring were genotyped for hph or hph with hygromycin resistance assays, for mus-52 Δ with nourseothricin resistance assays, and for rfk-2 with spore killing assays.

rid; mus-51

rid; Sk-3 rsk Δ ::hph rfk-2 D ::hph a

rid; Sk-3 rfk-2 D ::hph a

rid; Sk-3 rfk-2

rid; Sk-3 rfk-2

rid; Sk-3 rfk-2 Δ ::bar a

rid; Sk-3 rfk-2 Δ:: hph; mus-51 Δ ::bar a

rid; Sk-3 rfk-2 Δ:: hph; mus-51 Δ ::bar a

rid; Sk-3 rfk-2 Δ:: hph; mus-51 Δ ::bar a

rid; Sk-3 rfk-2 Δ:: hph; mus-51 Δ ::bar a

rid; Sk-3 rsk Δ ::hph rfk-2

Strain	Genotype	Source
F2-23	rid; fl A	Hammond et al. 2012
F2-26	rid; fl a	Hammond et al. 2012
F3-14	rid; fl; Sk-3 A	Hammond et al. 2012
FGSC 10340	rid; mus-51 RIP70 a	Smith et al. 2016
ISU-3036	rid; fl; sad-2 Δ ::hph A	Samarajeewa et al. 2014
ISU-3037	rid; fl; sad-2 Δ ::hph a	Samarajeewa et al. 2014
ISU-3291	rid; Sk-3; mus-51 Δ ::bar A	P8-42 × F3-14
ISU-4677	rid; Sk-3 rsk Δ ::hph; mus-51 Δ ::bar A	Trans. ISU-3291 with v14bc
ISU-4678	rid; Sk-3 rsk Δ ::hph rfk-2 UV ; mus-51Δ::bar; sad-2 D ::hph a	ISU-4773 × ISU-3036
ISU-4679	rid; Sk-3 leu-1 Δ ::nat-1; mus-51 Δ ::bar A	Trans. ISU-3291 with v260
ISU-4681	rid; Sk-3 rfk-2 UV ; mus-51Δ::bar; sad-2 D ::hph a	ISU-4678 × ISU-4679
ISU-4682	rid; Sk-3 rfk-2 UV a	ISU-4681 × F2-23
ISU-4683	rid; Sk-3 rfk-2 UV a	ISU-4682 × ISU-3291
ISU-4684	rid; Sk-3 rfk-2 UV ; mus-51 Δ ::bar a	ISU-4683 × ISU-3291
ISU-4685	rid; Sk-3 rfk-2 UV v322 Δ:: hph; mus-51 Δ ::bar a	Trans. ISU-4684 with v322
ISU-4686	rid; Sk-3 rfk-2 UV v322 Δ:: hph; mus-51 Δ ::bar a	Trans. ISU-4684 with v322
ISU-4687	rid; Sk-3 rfk-2 UV v324 Δ:: hph; mus-51 Δ ::bar a	Trans. ISU-4684 with v324
ISU-4688	rid; Sk-3 rfk-2 UV v324 Δ:: hph; mus-51 Δ ::bar a	Trans. ISU-4684 with v324
ISU-4689	rid; Sk-3 mus-52 Δ ::nat; mus-51 Δ ::bar A	Trans. ISU-3291 with v337
MAV214	rid; Sk-3 rsk Δ ::hph rfk-2 UV ; mus-51Δ::bar a	ISU-4677-UV × F2-26
P8-42	rid his-3; mus-51 Δ ::bar a	Hammond et al. 2011
PCR Primer Number	Sequence
49	CCGCTGGTTTGTGGTTCTTGATG
50	CAGCCACGGATCGCTTATCGTTT
297	GAGGGAGTGTGGGAAATGGTGTC
298	GTTGGTTAGGTGGGAACGCTTGT
585	CCGTCCACGCCCTTAATACGACT
586	CTTGATTGACAGCGAACGAAACC
1001b	CTCTGCTCTTCTTTCCCCGCTCCAACTGATATTGAAGGAGCAT
1002b	AACCTCGATCTCAAATGAAGCCGCAACTGGTTCCCGGTCGGCAT
1003b	ATAGGGGTGAAAAAGTGGCCTTC
1004b	ATGCTCCTTCAATATCAGTTGGAGCGGGGAAAGAAGAGCAGAG
1005b	ATGCCGACCGGGAACCAGTTGCGGCTTCATTTGAGATCGAGGTT
1006b	CCAGGCACCATCCAAGACAGTT
1007b	CTGGTCGCTTTTTGCTCTGTTTTCC
1008b	GTAATTCCAGGTGCCCAAGCTCA
1907	TGGGTGAATGTCTTGGGAAAGGA
1908	TGAATGCTAAAAGACACCATTTCCCACACTCCCTCGCTTCGAGGAGCTGGAATTATCAAA
1909	GCTGGCTGCAATACAAGCGTTCCCACCTAACCAACGGGCGATGCAAACAATGCTCTTT
1910	CACCTCACATCACACGCTCACCT
1911	GGACCTCGGGCAAGGATTGTAAG
1912	CTTTTCCCAAACTGCTCGCTCCT
2158	AGTCGTATTAAGGGCGTGGACGGCCGACGGATTTAGAACGAGGGC
2159	GGGTTTCGTTCGCTGTCAATCAAGTCCCCGAAGATAATACCCAAAGAGT
2160	AGTTTAGAAAACGGCGGCGGAG
2161	GGCAGAGTGGGTCCTAGCGATA
2162	AGGGTAAAATGTACGGACGAAGCT
2169	GGCGACTGTGGAATGGTAAGCG
2170	AGTCGTATTAAGGGCGTGGACGGCGTAGTGTAGGAAGCTCGGTCA
2171	GGGTTTCGTTCGCTGTCAATCAAGAGAAATGAGGCTGATAGGTAGACGT
2172	TTGACCCGACGTTCAAGATGCA
2173	TGCATTCGACTCACTTGGCATGG
2174	ACATCTTGCTGCTTCATTTCCCCT
2204	TTTCAATTTGGGAGCCGGGACTT
2205	GGGTATGTCAGGGCAAGAACGAC
2208	GCGTAATTGAGAGGCTCCCAACA
2214	TGAATGCTAAAAGACACCATTTCCCACACTCCCTCTCATTCGCGGTGGATTTCTAGGC
2215	GCTGGCTGCAATACAAGCGTTCCCACCTAACCAACTTCAAGAATGTCGAAGGCTGCCA
2216	GAGAATTGCGGGCGGGGAAGGAC
2217	GCCCCACTGTAGAGTTCACAAAGGACG