Genetic dissection of Nucleoporin 160 ( Nup 160 ) , a gene involved in multiple phenotypes of reproductive isolation in Drosophila

Kazunori Maehara, Takayuki Murata, Naoki Aoyama, Kenji Matsuno and Kyoichi Sawamura* Graduate School of Life and Environmental Sciences, University of Tsukuba, 1-1-1 Tennodai, Tsukuba, Ibaraki 305-8572, Japan Department of Biological Science and Technology, Tokyo University of Science, 2641 Yamazaki, Noda, Chiba 278-8510, Japan Research Institute for Science and Technology, Tokyo University of Science, 2641 Yamazaki, Noda, Chiba 278-8510, Japan Faculty of Life and Environmental Sciences, University of Tsukuba, 1-1-1 Tennodai, Tsukuba, Ibaraki 305-8572, Japan


INTRODUCTION
A century ago, Quackenbush (1910) claimed to have observed unisexual broods in Drosophila melanogaster.It turned out later that his D. melanogaster flies actually included two species, D. melanogaster and D. simulans (Sturtevant, 1919).The beauty of the latter new species was that it could be crossed with D. melanogaster, the most-studied and best-understood species of that genus (Provine, 1991).In crosses between D. melanogaster females and D. simulans males, only sterile female hybrids are obtained, as male hybrids die during larval development.In the reciprocal cross, sterile male hybrids appear, as most female hybrids die during embry-onic development.Sturtevant (1920) intercrossed the species using chromosome anomalies and was able to deduce the genetic causes of hybrid inviability, but "the complete sterility of surviving F 1 hybrids frustrated Sturtevant and his vision of comprehensively exploring the genetics of interspecific differences" (Barbash, 2010).Genetic tricks and the serendipitous discovery of rescue mutations were needed before further studies could shed light on his questions (Provine, 1991;Sawamura, 2000;Barbash, 2010).
Thanks to recent advances in molecular biology techniques and genomic sequencing (Adams et al., 2000;Drosophila 12 Genomes Consortium, 2007), detailed study of speciation has become feasible.As a result, several genes for hybrid inviability and sterility have recently been isolated in this pair of species and characterized at the molecular level (reviewed in Sawamura, 2012).The D. melanogaster gene Hybrid male rescue (Hmr) encodes a DNA-binding protein that is involved in hybrid inviability and female sterility (Hutter and Ashburner, 1987;Barbash and Ashburner, 2003;Barbash et al., 2003).D. simulans Lethal hybrid rescue (Lhr) encodes a heterochromatin protein that causes hybrid inviability (Watanabe, 1979;Brideau et al., 2006;Prigent et al., 2009).D. melanogaster zygotic hybrid rescue (zhr) consists of heterochromatic 359-bp repetitive sequences and causes hybrid inviability in crosses involving D. simulans females (Sawamura et al., 1993;Ferree and Barbash, 2009).JYalpha, a gene located on different chromosomes in D. melanogaster and D. simulans, causes male sterility of introgression homozygotes (Muller and Pontecorvo, 1940;Masly et al., 2006).D. simulans Nucleoporin 96 (Nup96) causes inviability when it is hemizygous in the hybrid; the hybrid males cannot be rescued by the Lhr mutation (Presgraves et al., 2003).D. simulans Nucleoporin 160 (Nup160) causes inviability and female sterility when introgressed into D. melanogaster; the hybrid males with the introgression (or deficiency) cannot be rescued by Lhr and introgression homozygous (or hemizygous) females are sterile (Tang and Presgraves, 2009;Sawamura et al., 2010).
In the present study, we excised the P transposable element from the P{EP}Nup160 EP372 insertion and examined whether the new mutations cause hybrid inviability and female sterility.Because Nup160 has been implicated as the cause of morphological anomalies by deficiency mapping (Sawamura et al., 2010), we also examined the abdomen, wings, and bristles of flies heterozygous for Nup160 sim and excision derivatives.Finally, we determined the lethal phase of hybrid males and measured the total duration of development of homozygous carriers.

Generation of excision mutations
The P{EP} element, which includes the mini-white gene (w + ), was excised from Nup160 EP372 in the germline of males in the w background by conventional methods using the defective P element Δ2-3 as the transposase source (Robertson et al., 1988).
A total of 219 white-eyed males were screened.The original Nup160 EP372 and 24 homozygous lethal derivatives were maintained using the CyO balancer chromosome.Homozygous viable derivatives seemed to be revertants and were discarded except for 10 lines kept as controls.The genetic symbol w is omitted hereafter unless such indication is necessary, because all experiments were conducted in the w background.

Genetic characterization of mutations
To check whether derivatives have mutations in Nup160 and the neighboring gene RfC38, they were made heterozygous with Nup160 EP372 , Nup160 e00704 , and RfC38 k13807 .At the opposite end of Nup160, the region containing Csl4s has not been examined genetically, because no appropriate mutations or deletions are known.Flies heterozygous for Nup160 sim and each derivative were produced by crossing introgression carrier females (Int(2L)D+S, Nup160 sim /CyO) to male derivative heterozygotes (Fig. 1B), and morphological anomalies (abdomen, wing, and bristle defects) and female fertility were examined as described (Sawamura et al., 2010).To test hybrid inviability, derivative carrier females were crossed to D. simulans Lhr males (Fig. 1C).All experiments were conducted at 25°C.

Molecular characterization of mutations
In the original homozygous lethal Nup160 EP372 chromosome, an 8-kb P{EP} element is inserted in the reverse orientation into the 5'UTR of the Nup160 gene, which is also in the forward orientation adjacent to the 5'UTR of the Csl4 gene, at the site designated 2L: 11,123,814-11,123,822 (Berkeley Drosophila Genome Project coordinates, http:// genome.ucsc.edu/).GCCGGTGCC is the target site duplication of the P element.
DNA was extracted from derivative homozygotes (if viable) or heterozygotes with CyO, and DNA fragments around the Nup160 EP372 insertion site were amplified with the polymerase chain reaction (PCR).PCR primers and conditions are available upon request.When PCR products were separated on an agarose gel, a single band (in homozygotes) or double bands (in heterozygotes, one from the mutation allele and the other from the wild-type Nup160 allele on CyO) were expected.In four homozygotes and ten heterozygotes, the target DNA band from each gel was purified and sequenced.In nine heterozygous derivatives, double bands were not obtained despite the use of several primer pairs, presumably because of a large deletion or a large P element remnant.In these cases, DNA was extracted from derivative carrier flies heterozygous for Int(2L)D+S, Nup160 sim , and the PCR products of regions of interest (outside the large deletion or the large P element remnant) were directly sequenced.Heterozygosity (derived from D. melanogaster and D. simulans alleles) suggests that the derivative retains the corresponding region.If DNA from the adjacent positions to the insertion (~2L: 11,123,793 or 2L: 11,123,885~) was present in the derivative chromosome, a partial P remnant was suspected to remain at either side of the Nup160 EP372 insertion site.
Derivatives having exactly the same sequences were treated as being from the same excision event if they were descendants of a single start vial containing target males that carried both Nup160 EP372 and Δ2-3.If they were descendants of independent start vials, they were treated as independent mutations, because the same excision event may have occurred more than once.

Determination of hybrid lethal phase
We made a y w; Int(2L)D+S/CyO, y + strain by conventional crosses.To determine the lethal phase of Nup160 sim carrier hybrid males, heterozygous females were crossed to D. simulans Lhr males (Fig. 2), and the viability of yellow (y) offspring (i.e., Int(2L)D+S carrier males) was examined at different developmental stages.All other offspring must have the y + phenotype, which is distinguishable from y by mouth hook and denticle bands color during early development.Because sexing larvae by the size of gonadal imaginal discs is difficult in sterile interspecific hybrids (Shen, 1932), larvae were sexed based on Malpighian tubule color, which was white (w) in males.

Measurement of total development time
There is a possibility that homozygous Nup160 sim introgression affects not only female reproduction but also non-reproductive characteristics (e.g., development) in both sexes.To examine if Nup160 sim homozygotes develop normally, eggs were collected at 2-hr intervals from Int(2L)D+S/ CyO females crossed with Int(2L)D+S/CyO males, and emerging flies of introgression homozygotes and heterozygotes were counted every 2 hr.

Phenotypic effects of derivatives
Hybrid viability, female fertility, and several aspects of adult morphology were examined using appropriate genotypes of the Nup160 EP372 derivatives as shown in Tables 2 and 3.The results are summarized in Table 1.The original Nup160 EP372 does not lead to hybrid inviability, female sterility, or morphological anomalies (Tang and Presgraves, 2009;Sawamura et al., 2010).The three null mutations of Nup160 (i.e., Df(2L)Nup160EP372M180, Df(2L)Nup160M190, and Df(2L)Nup160EP372M69) exhibited hybrid inviability, female sterility, and morphological anomalies.Thus, the previous conclusion that Nup160 sim , not the introgression of Csl4 or RfC38, is responsible for hybrid inviability and female sterility (Tang and Presgraves, 2009;Sawamura et al., 2010) was confirmed.Furthermore, it is now apparent that morphological anomalies were caused by the same gene, which was not conclusive in the previous analysis (Sawamura et al., 2010).The results for Nup160 EP372M85 were the same as for the three nulls, suggesting that this might also be a null mutation (class i).Nup160 EP372M18 and Nup160 EP372M121 exhibited hybrid inviability and female sterility, but not morphological anomaly.These might be partial loss-of-function mutations (class ii).

Nup160
revertants (i.e., Nup160 EP372M133 and Nup160 EP372M142 ) did not exhibit hybrid inviability, female sterility, or morphological anomaly.This, too, is consistent with the conclusion that Nup160 sim is responsible for the three phenotypes.The results for Nup160 EP372M161 , Nup160 EP372M26 , Nup160 EP372M39 , Nup160 EP372M94 , and Nup160 EP372M185 were the same as for the revertants.They are apparently not Nup160 revertants because heterozygotes with Nup160 EP372 or Nup160 e00704 were lethal, but they behave like revertants in terms of hybrid phenotypes (class iii).This is similar to the original Nup160 EP372 .
Interestingly, the remaining two derivatives exhibited different combinations of hybrid phenotypes, although none affected morphology.Nup160 EP372M227 resulted in complete female sterility but not hybrid inviability (class iv).This is similar to Nup160 SH2055 , which leads to incomplete female sterility but has no effect on hybrid inviability (Sawamura et al., 2010).Df(2L)Nup160EP372M219 exhibited the opposite trend, resulting in hybrid inviability but not in female sterility (class v).

Developmental analyses
In crosses between y w; Int(2L)D+S/CyO, y + females and D. simulans Lhr males (Fig. 2), Int(2L)D+S, Nup160 sim carrier males (phenotypically yellow white) were observed in the third instar larval and early pupal stages but not as late pupae and adults (Table 4).Thus, the lethal phase of the Nup160 sim carrier males seems to be during the early pupal stage.The total duration of development was 10.7 hr longer in female introgression homozygotes than in female heterozygotes (t = 6.104, df = 153, P = 8.15 × 10 -9 ), and 13.9 hr longer in male introgression homozygotes than in male heterozygotes (t = 6.430, df = 130, P = 2.23 × 10 -9 ; Table 5).Thus, a recessive gene (or genes) on the introgression chromosome makes development slower in both females and males.

DISCUSSION
We obtained 13 imprecise recessive lethal excisions from Nup160 EP372 and examined their effects on hybrid viability, female fertility, and morphology in the appropriate genotypes.Our results confirm previous observations that the D. simulans allele of the Nup160 gene (Nup160 sim ), not introgression of Csl4 or RfC38, is responsible for hybrid inviability and female sterility in crosses between D. simulans and D. melanogaster (Tang and Presgraves, 2009;Sawamura et al., 2010).In addition, we discovered that morphological anomalies are also caused by Nup160 sim .The Nup160 EP372 derivatives resulted in variable hybrid phenotypes, ranged from class i to class v.For example, the class v derivative affected hybrid viability more severely than female fertility, whereas the class iv derivative had the opposite effect.All the exons of Nup160 were intact in class i-v derivatives; the derivatives differed in the partial transposon remnants found in the 5'UTR or in a deletion of adjacent sequences.Such exogenous sequences might negatively regulate Nup160 expression both temporally and spatially.There is also a possibility that Df(2L)Nup160EP372M219 lacks an upstream regulatory region of Nup160.These mutations partially complement the hybrid phenotypes and will be useful in future analyses to examine the developmental mechanisms of hybrid inviability and female sterility.
In the present analysis, the lethal phase of the Nup160 sim carrier hybrid males was determined to be during the early pupal stage, which is later than that for regular hybrid males from crossing D. melanogaster females with D. simulans males (i.e., those not rescued by Lhr or Hmr; Sturtevant, 1920;Hadorn, 1961;Bolkan et al., 2007).And it has been suggested that Nup160 sim does not directly interact with the rescuing genes Lhr and Hmr, but rather that Nup160 sim results in hybrid inviability through an independent genetic system (Tang and Presgraves, 2009;Sawamura et al., 2010).Also in the present analysis, a recessive gene (or genes) on the Int(2L)D+S introgression was found to slow development of the homozygous carriers.Nup160 sim is a candidate, although we cannot rule out the possibility that other linked genes are responsible.Because homozygous (or hemizygous) Nup160 sim in the D. melanogaster genetic background results in not only female sterility but also morphological anomalies in both sexes, it is not a surprise that the same gene perturbs development in a pleiotropic manner.The Nup160 sim gene is apparently involved in multiple reproductive isolation phenotypes in the cross between D. melanogaster and D. simulans.
We are grateful to the Bloomington, Exelixis, Kyoto, and Szeged Drosophila stock centers for providing fly strains.This work was supported by a Grant-in-Aid for Scientific Research (21570001) from the Japan Society for the Promotion of Science to K. S.

Fig. 1 .
Fig. 1.Transposon insertions in the Nup160 gene and mating schemes to examine the effects of Nup160 EP372 derivatives.A, Positions and directions of three transposon insertions (triangles and arrows, not to scale).Open reading frames (full or partial) of Csl4, Nup160, and RfC38 are indicated (exons are numbered).UTR, untranslated region.B, Cross used to test the effects of mutations on morphology and female fertility.Open chromosome regions are from D. melanogaster; shaded regions are from D. simulans.C, Cross used to test the effects of mutations on hybrid viability.Int, Int(2L)D+S, Nup160 sim ; *, Nup160 EP372 derivative.

Fig. 2 .
Fig. 2. Cross used to examine the phase of hybrid lethality caused by Nup160 sim .Int, Int(2L)D+S, Nup160 sim .Open chromosome regions are from D. melanogaster; shaded regions are from D. simulans.Introgression carrier hybrid males (circled) are phenotypically distinguishable from the others.

Table 3 .
Hybrid viability of crosses between females heterozygous for Nup160 EP372 derivatives and CyO and D. simulans Lhr males Viability of Cy + flies relative to Cy flies was calculated as the number of Cy + flies divided by the number of Cy flies. a

Table 4 .
Cross of y w; Int(2L)D+S/CyO, y + females and D. simulans Lhr males Simultaneously collected; developmental speed difference among phenotypes might be reflected.b Total of In(2L)D+S carrier and CyO carrier females.c Lethal at the early pupal stage. a