The Himalayan uplift and evolution of aquatic biodiversity across Asia: Snowtrout (Cyprininae: Schizothorax) as a test case

then to the Bay of Bengal; and lastly, an earlier capture by the Ganges of four western Himalaya rivers (i.e., Jhelum, Chenab, Ravi, Sutlej) which are now incorporated within the contemporary Indus Basin. The latter hypothesis is supported by isotopic and seismic data extracted from the detrital fan of the Indus River, as deposited for >30 Ma in the Arabian Sea. These data document a major shift at ~5 Ma, with earlier Himalayan deposition superseding those from the Indus suture zone to the north [91]. This documents the capture of several large Punjabi rivers (i.e., Jhelum, Chenab, Ravi, Sutlej) by the Indus, as represented in the current basin topography. However, this interpretation is not without controversy. A subsequent comparison of sedi- ments from northern Pakistan with those of the Indus Fan suggests the transition was a poten- tial response to differential erosion rates in the source region (the Indus Basin) rather than drainage rearrangements [90]. The east-west divergence among Himalayan clades provides a framework by which the geo- morphic evolution of the drainage system can be interpreted. In this sense, our phylogeo- graphic results can be inferred as biologically corroborating the capture by the Indus of those drainages previously allocated to the Ganges Basin. For example, Schizothorax from the Lower Ganges River diverged from a lineage representing the Upper Ganges and the Indus River. Similar patterns have been noted for Indus and Gangetic dolphin [92]. These data also suggest

Schizothorax in the western Himalaya (i.e., Koshi, Gandaki, and Karnali rivers) were isolated prior to the capture of the Jhelum by the Indus River. Overall, our analyses reflect these unre- solved questions, as indicated by placement of individual haplotypes apart from others with the same geographic origin (Figs 4 and 5, S1 Fig). A more detailed study is needed as our anal- yses only include a limited number of samples representing this biogeographic region. Taxonomic uncertainty in Schizothorax and its implications Several species in our analyses were polyphyletic, with individuals forming biogeographically defined clusters that differed significantly one from another. In addition, multiple cases also emerged where samples within the same region failed to cluster with conspecifics (e.g., Central Asia; Eastern QTP/Southeast Asia; Koshi; Gandaki; Karnali; Ganges). Incongruence between molecular and morphological phylogenies represents a challenge for the current study, as reflected by morphologically-identified species from different drainages failing to cluster together. In contrast, the converse is true for different species in the same drainage [16, 78]. This can reflect incomplete lineage sorting [21, 77]; hybridization (interge- neric: [5, 93], intrageneric: [94]); rapid diversification [5]; and/or convergent evolution (e.g., juxtaposition of trophic and spatial niches across species; [95, 96]). None can be resolved using mtDNA [22]. Unfortunately, these issues are amplified within online databases by an incorrect association of DNA sequences with taxonomy [97].

Conclusions Schizothorax in the Himalaya is a monophyletic genus hypothesized as originating in Central Asia and subsequently dispersing southward into Bhutan via down-cutting rivers, as well as westward across the Eastern QTP and into Southeast Asia. The subsequent capture of the YLTR by the Brahmaputra allowed for an east-to-west dispersal south of the Himalaya, facili- tating sequential movements northward into Himalaya drainages, first in Bhutan, then Nepal, and eventually into the western Himalaya. Our molecular data corroborate western drainages as being previously affiliated with the Ganges River but subsequently captured by the Indus River. PLOS ONE Himalayan uplift and aquatic evolution PLOS ONE | https://doi.org/10.1371/journal.pone.0289736 October 24, 2023 20 / 27