Diversity and endemism of Murinae rodents in Thai limestone karsts: a genetic approach | Diversité et endémisme des rongeurs Murinae dans les massifs karstiques thaïlandais : une approche génétique

Limestone karsts are characterized by an important species richness and high levels of endemic species of plants, vertebrates and invertebrates adapted to this extreme environment. However, in Southeast Asia, karst ecosystems suffer from a considerable lack of scientific data and remain widely unknown despite their high biological importance. Combining field investigations with phylogeographic and phylogenetic analyses based on several kinds of molecular markers, this thesis aims at exploring the diversity and endemism of Murinae rodents in limestone karsts of Thailand. Thai limestone karsts host two endemic Murinae rodent species, Leopoldamys neilli and Niviventer hinpoon. This thesis reveals that L. neilli is more largely distributed in Thailand than indicated by previous records available in the literature. The species has been recorded in numerous limestone karsts of Thailand with the exception of its peninsular area. L. neilli has also been discovered in central Laos. Moreover our niche modeling study indicates that large tracts of suitable habitat for this species may also occur in several regions of Indochina. L. neilli populations are highly fragmented and a deep genealogical divergence among its lineages is observed. Both mitochondrial and nuclear markers support a large-scale population structure of four main groups (west, centre, north and northeast) and a strong finer structure within each of these groups. These results indicate that L. neilli populations are isolated on karsts such as on islands and that migrations among them are reduced. Our findings also suggest that the current phylogeographic pattern of this species results from the fragmentation of a widespread ancestral population and that vicariance has played a significant role in the evolutionary history of L. neilli. These deep vicariant events that occurred during Plio-Pleistocene are related to the formation of the Central Plain of Thailand. Moreover, the western populations of L. neilli are genetically and morphologically highly divergent from the other populations and could represent a separate species. This strong phylogeographic pattern is not observed for other Murinae species with lower levels of ecological specialization such as Leopoldamys edwardsi and Rattus tanezumi. Finally, this thesis provides preliminary information about the diet of L. neilli and indicates that plants of the Solanaceae family constitute an important part of its diet. The phylogeographic structure of N. hinpoon is not similar to the one of L. neilli. In contrast to L. neilli, N. hinpoon is confined to central Thailand and mitochondrial markers used in this study indicated that this species is genetically homogenous and characterized by a single mitochondrial lineage. Valuable data to refine the conservation status of L. neilli and N. hinpoon, two species currently listed as Data Deficient on the IUCN Red List have been gathered during this study. Three main threats to the long-term subsistence of L. neilli have been identified: (1) the high fragmentation of its population, (2) the large-scale destruction of limestone karsts in Southeast Asia, and (3) the intense trapping of this species for human consumption in northeastern Thailand. Therefore we propose to consider L. neilli as “Near threatened” on the IUCN Red List. However, if the western lineage of L. neilli represents a separate species, it should be listed as “Vulnerable”. Due to its small distribution range and the high threats that its habitat is facing in central Thailand, N. hinpoon would also be qualified as “Vulnerable” on the IUCN Red List. In addition to karst endemic species, this thesis reveals that Thai limestone karsts host high levels of Murinae rodent diversity. A sequence-based species delimitation method completed by the analysis of the level of genetic divergence was used to define species boundaries within our rodent samples collected in limestone karsts. A total of 12 described Murinae species, corresponding to 17 putative species based on our genetic criteria, were encountered in limestone karsts of Thailand. Most of these species are associated to forest ecosystems. Therefore this study suggests that limestone karsts could play a key role in the preservation of the rodent biodiversity by providing refuges for the forest-dwelling Murinae rodents in deforested regions. An important cryptic diversity has been detected within the traditionally recognized species Maxomys surifer and Berylmys bowersi. They could be considered as species complex and require further taxonomic work. The potential distribution of Leopoldamys edwardsi and Leopoldamys sabanus, two species also distributed in Thailand, has been investigated using niche modeling techniques. The predicted distribution ranges of these two species suggest a clear geographical separation between them, with the potential distribution of L. edwardsi being limited to the northern part of Indochina while L. sabanus is mainly distributed in the Sundaic region. Our findings also suggest that these two species could have survived in large areas of Southeast Asia during Quaternary ice ages without large scale extinction and that no drastic modification of the distribution of these species will occur in the future due to climate changes. In conclusion, using various genetic approaches, this work gains important insights into the Murinae rodent diversity of Thai limestone karsts and represents the first detailed study of karst endemic rodent species in Southeast Asia.