Aim: The Massenerhebung effect (Mass elevation effect) refers to heat or wind-driven altitudinal distribution patterns of temperature-dependent parameters among massifs with narrower range and lower elevation around peripheral and isolated mountains compared to core and continuous ones. Although common in ecology, this effect is rarely discussed in population genetics. Here, we use genetic markers to reveal population genetic patterns and also test the mountain- and sky-barrier hypotheses relevant to the Massenerhebung distribution pattern of Acer morrisonense in Taiwan's rugged topography and varied local climates.

Location: The alpine and cloud forest of Taiwan.

Taxon: Acer morrisonense Hayata

Methods: Two chloroplast DNA (cpDNA) fragments and 17 expressed sequence tag-simple sequence repeat (EST-SSR) loci respectively from 200 and 286 individuals were used to elucidate the phylogeographic pattern of pollen and seed dispersal of A. morrisonense. These data were combined with ecological niche modeling (ENM) to infer distribution range shifts and refugia. We also correlated the genetic-divergence indices with spatial factors to clarify latitudinal and altitudinal effects on genetic diversity.

Results: The incongruent phylogeographic patterns of genetic distributions between nuclear and cpDNA markers indicate unhindered pollen flow but spatially constrained seed dispersal. Taken together with ENM, the genetic pattern further reflects historical colonization from central-mountain refugia to edges since the Holocene. The Massenerhebung reduces the gene flow by the surrounding mountains and also causes lower genetic diversity compared to central alpine populations.

Main conclusions: This study is the first to reveal the influence of Massenerhebung effect on cpDNA genetic structure of montane trees and reflect the spatial trends of seed dispersal. This population genetic pattern can also be attributed to the demography-related range shifts with paleoclimate fluctuations under complex mountain topography, supporting the mountain-barrier hypothesis. The results have important implications for conserving the genetic diversity of species with a wide altitudinal distribution range.

The sampling range mainly focused on CMR following Chiou et al. (2010), and the SMR was also considered for its second-largest mountain range in Taiwan. Thirteen populations with 286 individuals (Table 1) and 12 populations with 200 individuals of A. morrisonense (Table 2) surrounding the SMR and CMR in Taiwan were sampled for nuclear SSR genotyping and cpDNA sequencing, respectively (Fig. 2). The spacing between sampled individuals was at least 20 m. The altitudes of the sampling sites ranged from 1,280 m to 2,700 m a.s.l. Based on the altitudinal distribution pattern of A. morrisonense (Fig. 1), the populations SKR, LLS, and TPS in the north and JBS, LD, and STC in the south were respectively defined as northern and southern peripheral populations compared to the rest of the core populations (Fig. 2). Fresh leaves of each individual were collected and dried immediately in silica gel to prevent DNA degradation. Voucher specimens of each sample were deposited at National Taiwan Normal University.

Total genomic DNA was extracted from dried leaf tissue following a modified cetyltrimethylammonium bromide (CTAB) method (Doyle & Doyle, 1987). The extracted DNA was dissolved in 1× TE buffer and stored at −20 °C. The expressed sequence tag-simple sequence repeats (EST-SSRs) used for genotyping for the genetic diversity analyses were developed from transcriptomic assemblies of Acer saccharum Marshall (Accession: Acer saccharum 010515, Hardwood Genomics Project, http://www.hardwoodgenomics.org/) and Acer negundo L. (Accession: VFFP, One Thousand Plants (1KP) Consortium, https://sites.google.com/a/ualberta.ca/onekp/). A total of 17 polymorphic EST-SSR loci were developed and used in this study (Table S1). For genotyping, the PCR products were analyzed by capillary electrophoresis on an Applied Biosystems 3730 DNA Analyzer (Applied Biosystem, USA). The fragment size was analyzed by Peak Scanner version 1.0 (Applied Biosystem, USA) at the National Center for Genome Medicine, Academia Sinica, Taiwan, and determined by reference to size standard ABI GS500 LIZ (Applied Biosystems, USA). For peak picking and noise reduction, a minimum peak height of 100 was adopted for allelic calling. Those peaks with sizes falling into the expected range were manually checked and adjusted.

The dataset is integrated in GenALEx format.