A Lethal Genetic Incompatibility between Naturally Hybridizing Species in Mitochondrial Complex I

The evolution of reproductive barriers is the first step in the formation of new species and can help us understand the diversification of life on Earth. These reproductive barriers often take the form of “hybrid incompatibilities,” where alleles derived from two different species no longer interact properly in hybrids. Theory predicts that hybrid incompatibilities may be more likely to arise at rapidly evolving genes and that incompatibilities involving multiple genes should be common, but there has been sparse empirical data to evaluate these predictions. Here, we describe a mitonuclear incompatibility involving three genes in physical contact within respiratory Complex I in naturally hybridizing swordtail fish species. Individuals homozygous for specific mismatched protein combinations fail to complete embryonic development or die as juveniles, while those heterozygous for the incompatibility have reduced function of Complex I and unbalanced representation of parental alleles in the mitochondrial proteome. We find that the impacts of different genetic interactions on survival are non-additive, highlighting subtle complexity in the genetic architecture of hybrid incompatibilities. We document the evolutionary history of the genes involved, showing for the first time that an incompatibility has been transferred between species via hybridization. This work thus provides the first glimpse into the genetic architecture, physiological impacts, and evolutionary origin of a complex incompatibility impacting naturally hybridizing species.

sensitivity of multi-protein interactions to changes in any of their components 6 . However, genetic 63 interactions are notoriously difficult to detect empirically except in systems with especially 64 powerful genetic tools 8 , and this problem is exacerbated with complex genetic interactions 9,10 . from their relatives, and in many cases are a patchwork of ancestry from several genomes due to 91 historical hybridization [37][38][39] . The impact of these dynamics on the evolution of hybrid 92 incompatibilities have been poorly investigated 40 since the foundational theory in this area was 93 developed before the ubiquity of hybridization was fully appreciated 7 . 94 Here, we use an integrative approach to precisely map the genetic basis and physiological 95 impacts of a lethal mitonuclear hybrid incompatibility in swordtail fish and uncover its

144
The degree of segregation distortion observed in F2 individuals on chromosome 6 is also 145 surprising (Fig. 1D). Only 3% of individuals harbor homozygous X. birchmanni ancestry in this 146 region (compared to 0.1% in the chromosome 13 region and 25% on average at other loci across 147 the genome; Fig 1F), which is again lower than expected for a nuclear-nuclear hybrid  However, a few populations segregate for the mitochondrial genomes of both parental types, and 159 8 we focused on one such population (the "Calnali Low" population, hereafter the admixture 160 mapping population). Admixture mapping for associations between nuclear genotype and 161 mitochondrial ancestry (after adjusting for expected covariance due to genome-wide ancestry 44 ) 162 revealed two genome-wide significant peaks and one peak that approached genome-wide 163 significance (Fig. 1G, Table S1-S3). The strongest peak of association spanned approximately 164 77 kb and fell within the region of chromosome 13 identified using F2 crosses (Fig. 1G). This  We also identified a peak on chromosome 6 that approached genome-wide significance  proteins and assembly factors that form respiratory Complex I 51 (see Table S1 for locations of  that it has a distinct genetic architecture from the incompatibility involving the X. malinche 205 mitochondria and X. birchmanni ndufs5 and ndufa13. Specifically, analysis of genotypes at the 206 admixture mapping peak indicates that the X. birchmanni mitochondria is incompatible with 207 homozygous X. malinche ancestry on chromosome 15 (Fig. 1C, Fig. S9). This region did not 208 contain any members of Complex I, but dozens of genes in this interval interact with known 209 mitonuclear genes (see Table S3; Supplementary Information 1.1.10). The fact that we detect 210 incompatible interactions with both the X. malinche mitochondria (at ndufs5 and ndufa13) and revealed that embryos with homozygous X. birchmanni ancestry at ndufs5 and X. malinche 229 mitochondria are present at early developmental stages, but that these embryos failed to develop 230 beyond a phenotype typical of the first seven days of gestation (the full length of gestation is 21-231 28 days in Xiphophorus; Fig. 2A-B, Fig. 2D-E). Individuals with mismatched ancestry at ndufs5 232 whose siblings were fully developed still had a detectable heartbeat but had consumed less yolk 233 than their siblings and remained morphologically underdeveloped (Fig. 2D, Fig. S14

253
Our analysis of developing embryos indicates that individuals with the ndufs5 254 incompatibility exhibit delayed or arrested embryonic development, whereas those with the 255 ndufa13 incompatibility did not. This suggests that these genes may drive lethality through 256 partially distinct mechanisms. As such, we chose to further investigate the effects of ndufs5 and  Table S6). We find initial evidence that cardiac defects persist into adulthood in   Xiphophorus embryos prevents us from using assays that directly target Complex I. To further 291 explore the effects of the hybrid incompatibility on Complex I function in vivo, we turned to 292 adult F1 hybrids between X. birchmanni and X. malinche (Fig. 3A). Since F1 hybrids that derive 293 their mitochondria from X. malinche and are heterozygous for ancestry at ndufs5 and ndufa13 are 294 fully viable, we asked whether there was evidence for compensatory nuclear or mitochondrial 295 regulation that might be protective in F1 hybrids. We found no evidence for significant With no indication of a compensatory regulatory response, we reasoned that we might be 300 able to detect reduced mitochondrial Complex I function in hybrids heterozygous for ancestry at 301 ndufs5 and ndufa13. We quantified respiratory phenotypes in isolated mitochondria using an  While we can leverage high resolution admixture mapping to pinpoint the nuclear 332 components of the hybrid incompatibility, we cannot take this approach to distinguish among the 333 37 genes in the swordtail mitochondrial genome because they do not undergo meiotic 334 recombination. To begin to explore the possible mitochondrial partners of ndufs5 and ndufa13, 335 we therefore turned to protein modeling, relying on high quality cryo-EM based structures 59-61 .

336
Although these structures are only available for distant relatives of swordtails, the presence of 337 the same set of supernumerary Complex I subunits and high sequence similarity suggest that 338 using these structures is appropriate (Table S12-   Theory predicts that hybrid incompatibilities are more likely to arise in rapidly evolving 363 genes 4,5,7,15 . Consistent with this hypothesis, ndufs5 is among the most rapidly evolving genes 364 genome-wide between X. birchmanni and X. malinche (Fig. 4C-D). Aligning the ndufs5 coding 365 sequences of X. birchmanni, X. malinche, and twelve other swordtail species revealed that all 366 four amino acid substitutions that differentiate X. birchmanni and X. malinche at ndufs5 were 367 derived on the X. birchmanni branch (Fig. 4C). Phylogenetic tests indicate that there has been 368 accelerated evolution of ndufs5 on this branch (dN/dS > 99, N = 4, S = 0, codeml branch test P = 369 0.005, Fig. 4C). Similar patterns of rapid evolution are observed at ndufa13, which also showed 370 evidence for accelerated evolution in X. birchmanni (Fig. 4E; dN/ Table S16), including ones 382 predicted to be in contact between ndufs5, ndufa13, nd2, and nd6 (Fig. 4A, 4B; Fig. S41).

384
Introgression of genes underlying mitonuclear incompatibility 385 The presence of a mitonuclear incompatibility in Xiphophorus is especially intriguing, 386 given previous reports that mitochondrial genomes may have introgressed between species 37 . 387 While X. malinche and X. birchmanni are sister species based on the nuclear genome, they are 388 mitochondrially divergent, with X. malinche and X. cortezi grouped as sister species based on the 389 18 mitochondrial phylogeny 37 (Fig. 5A-B). As we show, all X. cortezi mitochondria sequenced to 390 date are nested within X. malinche mitochondrial diversity (Fig. 5B, Fig. S42, S46; 391 Supplementary Information 1.5.3-1.5.4). Simulations indicate that gene flow, rather than 392 incomplete lineage sorting, drove replacement of the X. cortezi mitochondria with the X. 393 malinche sequence (P < 0.002 by simulation; Fig. 5C; Supplementary Information 1.5.4).

394
The introgression of the mitochondrial genome from X. malinche into X. cortezi raises the 395 possibility that other Complex I genes may have co-introgressed 66 . Indeed, the nucleotide 396 sequence for ndufs5 is identical between X. malinche and X. cortezi, and the sequence of ndufa13 397 differs by a single synonymous mutation (although conservation of both genes is high throughout 398 Xiphophorus; Fig. S47-S48). These identical amino acid sequences suggest that hybrids between 399 X. cortezi and X. birchmanni are likely to harbor the same mitonuclear incompatibility we 400 observe between X. malinche and X. birchmanni, as a result of ancient introgression between X. indicating that the hybridization events were independent (Fig. S49). We find that all known X. What genetic and evolutionary forces drive the emergence of hybrid incompatibilities, 416 especially between closely related species? Theory predicts that hybrid incompatibilities 417 involving multiple genes should be common 6,7 , but with few exceptions 8,11-13 , they remain 418 virtually uncharacterized at the genic level 6 . Here, we identify incompatible interactions in 419 mitochondrial Complex I that causes hybrid lethality in lab and wild populations. Our findings in 420 naturally hybridizing species echoes predictions from theory and studies in lab models 8,11-13 that 421 protein complexes may be a critical site of hybrid breakdown.

638
We tested for significant differences in developmental stage between siblings with compatible 639 and incompatible genotype combinations using a two-sided two-sample t-test (Supplementary 640 Information 1.3.1) and examined differences in ancestry between large groups of siblings that 641 varied in their developmental stages ( Supplementary Information 1.1.7). We also collected data 642 on embryonic stage and variability between siblings in embryonic stage from both pure parental 643 species ( Supplementary Information 1.3.1). We used a different approach to pin-point the timing 644 of ndufa13 lethality given that it appeared to act postnatally ( Supplementary Information 1.3.3).    Mapping results allowed us to identify ndufs5 and ndufa13 as X. birchmanni genes that 692 interact negatively with X. malinche mitochondrial genes. We used a protein-modeling based 693 approach with RaptorX (http://raptorx.uchicago.edu) to identify the mitochondrial genes most 694 likely to interact with ndufs5 and ndufa13 (see Supplementary Information 1.4.6). Using the 695 mouse Cryo-EM structure (PDB ID 6G2J) of Complex I, we identified proteins in contact with 696 ndufs5 and ndufa13, which included several mitochondrial (nd2, nd3, nd4l, nd6) and nuclear 697 (ndufa1, ndufa8, ndufb5, ndufc2) genes. We then used RaptorX to predict structures for both the 698