Method to obtain homozygous introgressing segments in Drosophila to 1 identify the presence of hybrid sterility genes of the reproductive isolation

Here, we present for the first time, a method to generate homozygous segmental 5 introgressions, by means of crosses between a pair of synmorphic species. The 6 introgressions were monitored by the cytogenetic method of polygenic chromosome 7 asynapses. Later the introgressions were evaluated in their capacity to produce sterility 8 in segmental males. Also, the smallest segment with the capacity to produce sterility in 9 segmental males was mapped by in situ hybridization of polythene chromosomes, using 10 8 sequences of BACs clones as probes. Finally, a bioinformatic analysis was carried out 11 to identify the presence of particular genes. From 2 parental strains, D. buzzatii and D. 12 koepferae , 6 simple segmental hybrid lines were generated, whose introgressing 13 segments are distributed along chromosome 4 of these species. From the 6 simple 14 segmental lines and by means of a new crossing strategy, the 6 respective homozygous 15 segmental hybrid offspring were obtained, each of them carrying a specific homozygous 16 introgression. None of the 6 heterozygous introgressions was capable of producing 17 sterility in segmental males, while 4 of the same homozygous introgressions produced 18 total sterility in segmental males, including in this group the two smallest introgressive 19 segments, one of 5.03 % and the other 7.87% with respect to the total length of 20 chromosome 4, which are located in the region F2 to F4 of the standard cytological map 21 based on polythene chromosomes of the Drosophila Repleta group. In situ hybridization, 22 using 8 clones from contig 1065 located along the F2 to F4 region of the physical map 23 of D. buzzattii constructed in BACs, confirmed the precise location of the 6 clones in the 24 chromosomal region F2 to F4 of chromosome 4 of the polygenic chromosomes of both 25 D. buzzatii and D. mojavensis. The bioinformatic analysis of the F2 to F4 region, using 26 the complete genetic sequence of the contig 1065 of D. buzzatti shows the presence of 27 two predicted genes in the genomic map of D. buzzatii (g.1313.t1 and g.1314.t1), and 28 the orthologous association of these 2 genes both with the D. moj_GI22766 gene of D. 29 mojavensis and with the Trivet gene of D. melanogaster.


32
One long-standing issue of debate in speciation theory concerns the number of genes that 33 are necessary for speciation to occur. 34 In Drosophila, it is suggested that the density of sterility factors may be of the order of 35 one per 1 percent of the genome and, by extrapolation, a minimum of 40 of these loci 36 has been estimated on the X chromosome, which gives an estimate of the sterility genes 37 between D. simulans and D. mauritiana of the order of 100 ) . Genes responsible for intraspecific gene sterility (Rieseberg and Blackman, 2010) show 50 an easily recognizable, clear-cut segregation, and are generally recessive. Experiments 51 on genomic fragment introgression from one species into another suggest two possible 52 types of action: 1) additive with a threshold effect (Naveira, et al., 1984) (Naveira &  53 Fontdevila, 1998) and 2) epistatic effects between conspecific genes in a heterospecific 54 background ; (Presgraves, 2007). Both architectures are polygenic, they 55 do not exclude the presence of genes with greater effects with strong or additive effects, 56 and that when identified can be cloned (Ting, et al., 1998), supporting the view that alleles 57 of large effect that distinguish species may sometimes reflect the accumulation of 58 multiple mutations of small effect at select genes (McGregor, et al., 2007). And Feller et 59 al, 2020, found that several of the characters selected for reproductive isolation are 60 assigned to a region of significant QTLs, supporting the theory of the presence of major 61 effect loci. 62 Many researchers seek to do identify individual genes from reproductive isolation scattered 63 throughout the Drosophila genome, and they do so by examining F2 backcross hybrids and 64 high resolution analysis between closely related species (Perez and Wu, 1995) (Coyne 65 and Orr, 1998) . In addition to genomic studies compared 66 between the sequenced species of the genus melanogaster (Song et al., 2011) 67 (Drosopgila 12 genome consortium, 207). 68 The Drosophila repleta group is an array of about 100 cactophilic species endemic to the 69 New World, many of them cactophilic species living in the deserts and arid zones of the 70 American continent and has become a model group for studies of ecological adaptation 71 and speciation for over 70 years (Ruiz et al., 1982 koepferae, forming twoo different clades.The radiation of the remaining six species 87 seems to be extremely recent, < 0.5 -1 Myr ago. 88 Considering the replacement rates of 6. 85  The cytological physic map for both D. koepferae and D. buzzati have the same polytene 110 karyotype as species of the Drosophila repleta group consisting of six chromosomes. 111 Each chromosome is divided into cytological intervals, identified by capital letters (A to 112 H). Each interval contains a specific amount of subintervals identified by numbers (1 to 113 5). Each subinterval is divided by a series of bands, identified by lowercase letters (a to 114 h), in alphabetical order from telomere to centromere (Wharton, 1942), (Ruiz et al., 1982

Method for genetic markers by chromosomal asynapsis 129
Each Drosophila culture produced from cross and backcrosses, was analyzed from a 130 sample of six to ten third instar larvae. Each offspring were genotyped based on the 131 presence of genetic markers known as chromosomal asynapsis, each chromosomal 132 segments introgressed produce one asynapsis is formed by an incomplete pairing of 133 chromatin fibers in the area of union between a pair of homologous chromosomes, 134 exclusively at sites where introgression has been successful. The procedure to identify 135 chromosomal asynapsis of polytene chromosomes is performed by the standard 136 squashing method: the salivary glands from a third instar Drosophila larva are extracted 137 in 45 percent acetic acid and placed on a slide with a drop of standard dye solution lacto-138 aceto-orcein. This preparation is sandwiched between a cover glass, placed between a 139 paper-towel and crushed with the thumb tip to release the gland. 140 The length and location of the introgressing segment in the cytological map indicate each 141 specific asynapsis.  229  230  231  232  233  234  235  236  237  238  239  240  241  242  243  244  245  246  247  248  249  250  251  252  253  254  255  256  257  258  The series of crosses to produce homozygous segmental offspring are summarized in 279 figure 3 and described below in a scheme comprising 7 general stages. 280 According to this crossing scheme, the first obtaining of homozygous segmental offspring 281 corresponding to stage 4 is obtained from the crossing between heterozygous segmental 282 males and females, which for this pair of sibling species D. buzzatii-D and Koepferae are 283 fertile (Naveira and Fontdevila 1986) (García-Franco et al, 2020 To fertility evaluation one To fertility evaluation one

BC 3
To fertility evaluation two  374 The maping and obtantion the genetic sequences, of the small chromosomal

397
Of the 31 contigs of chromosome four ( fig. 4), the contig selected for our research is 398 number 1065, which is made up of 63 BAC clones, each registered with a consecutive 399 number that goes from number 28 to number 90 (table 1). 400 The chromosomal region encompassing contig 1065 includes the chromosomal segment 401 of interest F2 (F2a-F3d) ( Table 3 and Fig. 4)  The DNA of BAC clones was used as an in situ hybridization probe, to confirm  The BAC clones were labeled with biotin by the "Nick Translation" method, using 427 the protocol of the commercial kit (BioNick kit, Gibco-BRL).

428
After labeling the probe, the next steps of the in situ hybridization process consist

Obtaining single segmental lines in homozygosity 451
By strictly following the crossing protocol, described in figure 3, it was possible to 452 obtain six segmental lines in the homozygous state (Table 3).

542
From the data obtained in the results of the test crosses 543 (Table 4) it was possible to observe the effect that each segment produces on 544 the fertility of males and segmental females.

545
The effect on fertility is shown in three levels: 1) total sterility; 2) semi-sterility and 546 3) fertility, and it turns out to be complex as it depends on several factors: 547 segment size, position on the chromosome, sex and genotype, as shown in Table   548 5.   also found, the de Naveira and Fontdevila 1986, 1991a, 1991b, as well as the 564 segments proposed by I. Marin, 1996.

565
All those chromosomal segments mentioned above, with the exception of the 6 566 segments analyzed in this work, were evaluated in their respective publications, 567 only in the heterozygous state for their ability to possess hybrid sterility genes.

568
In the following diagram, each of these segments are located in their   598 Eight BACS clones out of a total of 63 that make up contig 1065 (Tables 2 and   599 6), was acquired from the Oakland Children's Hospital Research Institute 600 (CHORI).

601
The Contig 1065 encompasses almost the entire F2 toF4 region (Fig. 4)   According to the proposed cytogenetic relationships between D. buzzatii and D.  The BLAST procedure with the megablast algorithm for highly similar sequences, 668 identified the region with the highest homology: 669 Is a zone of 3428 nucleotides in the R arm of chromosomal 3 of Drosophila 670 melanogaster, with 11 different segments homologous, which is located within the trivet 671 Previous studies on reproductive isolation between Drosophila buzzatti and Drosophila 751 koepferae were always carried out with introgressions of heterozygous chromosomal 752 segments (Naveira and Fontdevila 1986) (García Franco et al., 2020) and it was never 753 possible to identify individual genes with strong effect, although it was able to 754 demonstrate the effect of sterility in segmental males due to a threshold size when it was 755 possible to introgress a segment with a minimum total size of 31.15% with respect to the 756 total size of chromosome 4, either individually, in pairs or trios, or, both in cis action and 757 in trans action, but always in heterozygosity, as shown in Figure 7. Therefore, with this 758 methodology for obtaining homozygous introgressing segments, we now have the ability 759 to produce an enormous range of homozygous introgressive segments and be able to 760 evaluate them, not only for these species but for any group of species that present e 761 fertile segmental males in heterozygosis, and importantly, being able to isolate groups 762 of small individual introgressing segments that could be associated with one or a few 763 hybrid sterility factors and be able to differentiate them from those that act epistatically 764 between genes of different loci. 765 Obtaining 6 introgressive segments in a heterozygous state ( fig. 2) with sizes between 766 6.25% the smallest and up to 21.87% the largest (table 3), did not produce any effect of 767 sterility in males or segmental females, such as can be seen in table 5. 768 But when they were obtained in a homozygous state ( figure 3), they now showed that 769 the 6 segments were capable of producing some effect on sterility (table 5), and not only 770 in males but also in segmental females, depending not only on size of the segment, but 771 of its position within chromosome 4.  0K 10k 20k 30k 40k 50k 60k 70k 80k 90k 100k 110k 120k 130k 140k 150k 160k 170k 180k 190k 200k 210k 220k 230k 240k 250k 260k  In such a way that, the segments of size 6.25% (F2a-F3d), 7.87% (F3a-F4j) and 12.25% 773 (B1a-B4f) located in the F region and the first two and the third in the D region 774 respectively, they were able to produce sterility in segmental males, although in females, 775 its effect was null. While an intermediate size segment (12.25%) located in region B 776 produced semi-sterility in males and no effect in segmental females. Unlike the larger 777 segments, 15.62% (D1a-D5d) and 21.87% (F3a-G5c), located in the D and F regions 778 respectively, produced sterility in both females and segmental males. 779 The above indicates the potential presence of a single strong effect factor with the 780 capacity to produce sterility in segmental males carrying the smallest segments F2a-F3d 781 (6.25%) and F3a-F4j (7.87%), although we cannot rule out the possibility of epistasis 782 with other factors. 783 In addition, three of the segments (B4c-B4e, D5b-E1a and F2a-F3d) (fig 7D), proposed 784 by Marin, 1996 as having epistatic sterility factors, coincide in their position with 785 segments of this work (B1a-B4f, D3a-E1h, F2a-F3d) (Fig. 7B), especially the region that 786 overlaps between segments F2a-F3d and F3a-F4j, and that indicates an even smaller 787 region (F3a-F3e), with the possibility of the presence of a single strong effect factor, 788 which could act independently, the same happens with the overlapping segment 789 between the segments D1a-D5d and D3a-E1h, the small region D1a-D2e. 790 Also in relation to size, it is very revealing that the segments when they are small only 791 affect segmental males but when a larger region is considered (F3a-G5) they also affect 792 segmental females, which suggests additive or epistatic action. . Including the different 793 effect observed between two segments of the same size (12.25%), B1a-B4f and D3a-794 E1h, the first produce semi-sterility and the second sterility, in males (table 5 and figure  795  7). 796 On the other hand, the effect observed in the 6 segments regardless of their position 797 within chromosome 4 indicates the presence of factors distributed throughout the entire 798 chromosomes, which supports Naveira's original proposal (Naveira and Fontdevila, 799 1986) (Naveira et al., 1986) Naveira and Fontdevila, 1991a) (Naveira and Fontdevila 800 1991b) in the extensive study that he has carried out with this pair of species, where he 801 has demonstrated the effect of segment size throughout the genome. 802 Although each of the 6 homozygous segments is of comprehensive interest, we decided 803 to focus on the smaller F2a-F3d and F3a-F4j segments ( CHORI) the 8 clones, which are located along the small region F2 to F4 of the chromosm 807 map of D. buzzatii ( Figure 4 and Table 2, and successfully hybridize each one of them 808 when using them as probes, in the polygenic chromosomes of D. buzzatii and D. 809 mojavensis (Figures 8 and 9). With which we show that we are within region F2 to F4. 810 The bioinformatic study, using BLAST processes and identification of genes and 811 annotated genetic structures of the genomes of D. buzzatii (Drosophila buzzatii Genome 812 Projet), D. mojavensis and D. melanogaster (GenBank and Flybase) allowed us to focus 813 the genetic region of D. . buzzatii where the contig 1065 sequence is located within the 814 genome of D. buzzatii, turning out to be an interesting region, which although in D.

815
buzzatti only shows the presence of two predicted genes, g1313.t1 and g1314.t1, (figure 816 11A). An orthologous gene of the g1313.t1 gene is also shown, which is located in the 817 genome of D. mojavensis and is known as the Dmoj_GI22766 gene (Fig. B), which has 818 a biological function of participating in the regulation of splicing processes, in turn. Once 819 this gene GI22766 is orthologous with its counterpart in D. melanogaster known as the 820 Trivet gene (Fig. 11 C) and which has the same biological function. And these regions, 821 both of D. mojavensis and D. melanogaster, represent regions of high genetic mobility, 822 since at least one mobile element and several insertion sequences of mobile elements 823 are associated with these genes (Fig. 11C), and Interestingly, this entire region is located 824 in a space of only 5 cytogenetic bands of the D. melanogaster genome (Fig. 11 C). And 825 although the diagram in figure 11 only shows a small region of 280 kilobases, when 826 comparing the regions of adjacent to this region it is possible to identify that the groups 827 of synthenic genes are maintained with great consistency between D. buzzatti and D. 828 melanogaster, and they are also maintained, although to a lesser degree, between D. 829 mojavensis and D. Buzzatii. 830 With all of the above, we conclude that the method to obtain homozygous segmental 831 offspring allows for the first time homozygous introgressions with the ability to 832 independently evaluate introgressive segments, or it is possible to extend the scheme 833 for joint-segment analysis. 834 None of the 6 introgressing segments in heterozygosis produce effects on sterility. In 835 contrast, in the homozygous state, the 6 segments produce different effects on sterility, 836 when they are small segments (5 to 7%) they produce effects on the sterility of segmental 837 males, and when they are larger (12.25% to 21.87%) they produce semi-sterility. or 838 sterility in segmental males or sterility in both females and segmental males. That the 839 sterility factors extend throughout the entire chromosome 4 considering that they have 840 evaluated in percentage 57% with the 6 homozygous segments. And that the genes 841 g.1314.t1 and g1313.t1 of D. buzzatti, both still in a state of prediction, represent being 842 candidate genes involved in the reproductive isolation of these symmorphic species.

843
Although under the classical Bateson-Dobzhansky-Muller model, we are observing 844 changes in at least two genes, due to the differential interaction and in size and position 845 of large segments or in different positions of the genome, now we also observe the 846 possibility of the action of a single allele with strong effects on non-heterozygous sterility 847 and without epistasis to other genes, even considering the decreased aptitude of that 848 allele that makes it difficult to fix this allele in the population, therefore, it would then be 849 necessary to consider both possibilities, appealing to both models and using new 850 strategies to understand the relationship between both possibilities, especially in small 851 populations and in high-mobility genes. On the other hand, the dominance models show 852 us an additional possibility to address the situations revealed in this work, although it is 853 necessary to know the type of dominance of the allele in question under the current 854 models in relation to the genotypic and phenotypic values and their position within the 855 metabolic flow, and additionally it is relevant to consider the type of inheritance especially 856 if it turns out to be of maternal origin. Finally, It is now feasible to study allele fixation 857 processes as one of the fundamental processes of the hybrid sterility effect. 858 859