Name: FERNANDA COUTO ZAIDAN
Type: PhD thesis
Publication date: 12/12/2019
Advisor:
Name |
Role |
|---|---|
| Valéria Fagundes | Advisor * |
Examining board:
| Name |
Role |
|---|---|
| Ana Carolina Capellini Rigoni | External Alternate * |
| Celso Oliveira Azevedo | Internal Examiner * |
| JOÃO FILIPE RIVA TONINI | External Examiner * |
| Roberta Paresque | Internal Examiner * |
| Valéria Fagundes | Advisor * |
| Yatiyo Yonenaga-Yassuda | External Examiner * |
| Yuri Luiz Reis Leite | Internal Examiner * |
Summary: The terrestrial rodent Akodon cursor is endemic to the Brazilian Atlantic Forest, with broad range, easily found in open grasslands and disturbed areas from Paraíba in the north to the state of Paraná in the South of Brazil. This species has been under scientific investigation since the 70s and has passed through various taxonomic alterations until all its range and all its karyotypic diversity was known. One of its most remarkable
characteristics is its chromosomal diversity, presenting three diploid numbers, 2n= 14, 15 and 16 and number of autosomal arms varying from 18 to 26, what makes it have nearly 30 described karyotypes. Despite ample chromosomal variation, it is also observed structuring of genetic diversity of the species in two main geographic groups (north and south), which are coincident with the Jequitinhonha river, in the east of Brazil. Due to its broad geographic distribution, macro and micro variation in its DNA and for being a young species (with possibly less than 2 million years old), makes A. cursor an excellent model for investigating the relative roles of geographic constraints and structural genomic variants (i.e. chromosomes) in the early stages of speciation. During this PhD project we collected individuals from two populations in the north and one in the south of the species range, what made possible carrying out more than 400 experimental crosses, which in turn generated more than 500 animals that were analysed though ytogenetics and a sub sample had its fertility estimated using histological analyses. Besides that, genomic analyses were performed generating more than 10,000 SNPs of A. cursor, A. montensis and natural interspecific hybrids. Chapter 1 was dedicated to introducing an
overview of the diversity of the species by integrating chromosome information with geographic distribution and genomic data obtained through ddRAD-sequencing. In Chapter 2 we emphasized investigating fertility estimates regarding the different karyotypes and levels of geographic isolation between populations. In Chapter 1 I presented an overview about A. cursor by integrating genomic data, obtained through
ddRAD sequencing, with cytogenetic information and geographic distribution. In Chapter 2 I put emphasis in estimates of fertility regarding the different karyotypes and levels of geographic isolation between populations. In Chapter 1 it was observed that geographic
structure outstands in the genetic distribution of lineages of A. cursor, recovering the known north south clades with representatives of the three karyotypes in both regions. In addition to that, I suggest that specifically the region of the source of the Jequitinhonha river is related to the phylogeographic break of this rodent, once it is reported for that
region recent geological activity that overlaps with the putative period when A. cursor split into two main lineages. The different karyotypes seem to have appeared more than once during the evolution of A. cursor, since representatives of each karyotype do not share a most recent common ancestor. This assumption has implications on the theories of chromosomal evolution of A. cursor and the putative region of origin of the species, suggesting it was actually in its central portion and not in southern Brazil. Our data showed that in a particular population, WHERE indidivuals of the three karyotypes live in simpatry, the karyotypes do not form distinct genomic clusters, as previously
hypothesized, with the invidividuals 2n= 14 as a monophyletic clade, without signs of admixture with the clade of individuals 2n= 15+16. The heterokaryotype 2n= 15 presents rearrangements in pais 1 and 3 in heteorzigosis and the 2n= 14 has its rearrangement in homozygosis, therefore, it was believed that the heterokaryotype individuals of A. cursor would be a result of crosses between 2n= 14 and 2n= 16. However, genomic data pointed that, at least in this particular population, 2n= 15 is a variation of 2n= 16. This information raised the possibility that there could be reproductive isolation between karyotypic lineages of A. cursor. In Chapter 2 we refuted this suggestion since
experimental crosses between 2n= 14 and 2n= 16 from this population produced viable litter (though subfertile), indicating that there is no potential prezygotic isolation between such homokaryotype individuals. Thus, it was suggested that reinforcement could be happening in this population, with an increase of prezygotic barriers due to a putative
secondary contact of the lineages 2n= 14 and 2n= 15+16. Moreover, in the present work it was estimated rates of reproductive success (RS) and average litter size of samehomokaryotype crosses, which can be used as references for the species. Crosse between heterokaryotype individuals from the same population presented significantly inferios
rates of RS compared to homokaryotyoes, but it was recovered a high RS when these individuals were backcrossed with honokaryotypes. Though, backcrossings of 2n= 15 with homokaryotypes could be the principal means that this form is maintained in nature. Crosses between individuals from the same population and from adjacent populations
(even the ones from distinct clades) pointed that there is no complete reproductive isolation between lineages north and south of A. cursor, maybe due to the recency of the geological process that triggered the genetic differentiation of lineages. Nonetheless, crosses between individuals from geographically distant and allopatric popultions (Espírito Santo and Pernambuco, more than 2,000 km apart) produced litter 2n= 15 (15NS) that was sterile in crosses and backcrosses. These results evidenced that geographic isolation might be taking these lineages to accumulate epistatic incompatibilities. The observed sterility was corroborated by histological analyses which demonstrated that the 15NS males do not present lumen in their seminiferous tubules nor spermatozoa, such as interspecific hybrids. Hence, we could verify that A. cursor is in a situation analogous to a ring species, once its adjacent populations have the potential to breed, but representatives of the tips of the distribution are reproductively incompatible. By integrating genomic results with estimates of fertility between populations we have a scenario with A. cursor split into two main groups which are not correlated to specific karyotypes and that form sterile intraspecific hybrids when are put together. The cursor grass mouse can still be considered a biological species having signs of being in the initial stages of divergence and one can not predict if evolution will follow towards these main lineages to become two species. With very few works that integrated such approaches the present work adds to our knowledge showing that geography is the main stage for the karyotypes of A. cursor to emerge and interact.
Keywords: Akodon cursor, speciation, chromosomes, experimental crosses, genomics,
lineages
