The genus Amana Honda (Liliaceae), when it is treated as separate from Tulipa, comprises six perennial herbaceous species that are restricted to China, Japan and the Korean Peninsula. Although all six Amana species have important medicinal and horticultural uses, studies focused on species identification and molecular phylogenetics are few. Here we report the nucleotide sequences of six complete Amana chloroplast (cp) genomes. The cp genomes of Amana range from 150,613 bp to 151,136 bp in length, all including a pair of inverted repeats (25,629–25,859 bp) separated by the large single-copy (81,482–82,218 bp) and small single-copy (17,366–17,465 bp) regions. Each cp genome equivalently contains 112 unique genes consisting of 30 transfer RNA genes, four ribosomal RNA genes, and 78 protein coding genes. Gene content, gene order, AT content, and IR/SC boundary structure are nearly identical among all Amana cp genomes. However, the relative contraction and expansion of the IR/SC borders among the six Amana cp genomes results in length variation among them. Simple sequence repeat (SSR) analyses of these Amana cp genomes indicate that the richest SSRs are A/T mononucleotides. The number of repeats among the six Amana species varies from 54 (A. anhuiensis) to 69 (Amana kuocangshanica) with palindromic (28–35) and forward repeats (23–30) as the most common types. Phylogenomic analyses based on these complete cp genomes and 74 common protein-coding genes strongly support the monophyly of the genus, and a sister relationship between Amana and Erythronium, rather than a shared common ancestor with Tulipa. Nine DNA markers (rps15–ycf1, accD–psaI, petA–psbJ, rpl32–trnL, atpH–atpI, petD–rpoA, trnS–trnG, psbM–trnD, and ycf4–cemA) with number of variable sites greater than 0.9% were identified, and these may be useful for future population genetic and phylogeographic studies of Amana species.

The monocot genus Croomia (Stemonaceae) comprises three herbaceous perennial species that exhibit EA (Eastern Asian)–ENA (Eastern North American) disjunct distribution. However, due to the lack of effective genomic resources, its evolutionary history is still weakly resolved. In the present study, we conducted comparative analysis of the complete chloroplast (cp) genomes of three Croomia species and two Stemona species. These five cp genomes proved highly similar in overall size (154,407–155,261 bp), structure, gene order and content. All five cp genomes contained the same 114 unique genes consisting of 80 protein-coding genes, 30 tRNA genes and 4 rRNA genes. Gene content, gene order, AT content and IR/SC boundary structures were almost the same among the five Stemonaceae cp genomes, except that the Stemona cp genome was found to contain an inversion in cemA and petA. The lengths of five genomes varied due to contraction/expansion of the IR/SC borders. A/T mononucleotides were the richest Simple Sequence Repeats (SSRs). A total of 46, 48, 47, 61 and 60 repeats were identified in C. japonica, C. heterosepala, C. pauciflora, S. japonica and S. mairei, respectively. A comparison of pairwise sequence divergence values across all introns and intergenic spacers revealed that the ndhF–rpl32, psbM–trnD and trnS–trnG regions are the fastest-evolving regions. These regions are therefore likely to be the best choices for molecular evolutionary and systematic studies at low taxonomic levels in Stemonaceae. Phylogenetic analyses of the complete cp genomes and 78 protein-coding genes strongly supported the monophyly of Croomia. Two Asian species were identified as sisters that likely diverged in the Early Pleistocene (1.62 Mya, 95% HPD: 1.125–2.251 Mya), whereas the divergence of C. pauciflora dated back to the Late Miocene (4.77 Mya, 95% HPD: 3.626–6.162 Mya). The availability of these cp genomes will provide valuable genetic resources for further population genetics and phylogeographic studies on Croomia.

The figwort genus Scrophularia L. (Scrophulariaceae) comprises 200–300 species and is widespread throughout the temperate Northern Hemisphere. Due to reticulate evolution resulting from hybridization and polyploidization, the taxonomy and phylogeny of Scrophularia is notoriously challenging. Here we report the complete chloroplast (cp) genome sequences of S. henryi Hemsl. and S. dentata Royle ex Benth. and compare them with those of S. takesimensis Nakai and S. buergeriana Miq. The Scrophularia cp genomes ranged from 152 425 to 153 631 bp in length. Each cp genome contained 113 unigenes, consisting of 78 protein‐coding genes, 31 transfer RNA genes, and 4 ribosomal RNA genes. Gene order, gene content, AT content and IR/SC boundary structure were nearly identical among them. Nine cpDNA markers (trnH‐psbA, rps15, rps18‐rpl20, rpl32‐trnL, trnS‐trnG, ycf15‐trnL, rps4‐trnT, ndhF‐rpl32, and rps16‐trnQ) with more than 2% variable sites were identified. Our phylogenetic analyses including 55 genera from Lamiales strongly supported a sister relationship between ((Bignoniaceae + Verbenaceae) + Pedaliaceae) and (Acanthaceae + Lentibulariaceae). Within Scrophulariaceae, a topology of (S. dentata + (S. takesimensis + (S. buergeriana + S. henryi))) was strongly supported. The crown age of Lamiales was estimated to be 85.1 Ma (95% highest posterior density, 70.6–99.8 Ma). The higher core Lamiales originated at 65.6 Ma (95% highest posterior density, 51.4–79.4 Ma), with a subsequent radiation that occurred in the Paleocene (between 55.4 and 62.3 Ma) and gave birth to the diversified families. Our study provides a robust phylogeny and a temporal framework for further investigation of the evolution of Lamiales.

Background Phoebe (Lauraceae) comprises of evergreen trees or shrubs with approximately 100 species, distributed in tropical and subtropical Asia and Neotropical America. A total of 34 species and three varieties occur in China. Despite of economic and ecological value, only limited genomic resources are available for this genus.ResultsWe sequenced the two complete chloroplast (cp) genomes of Phoebe chekiangensis and P. bournei using Illumina sequencing technology via a combined strategy of de novo and reference-guided assembly. We also performed comparative analyses with the cp genomes of P. sheareri and P. sheareri var. oineiensis previously reported. The chloroplast genomes of P. chekiangensis and P. bournei identically contain 112 genes consisting of 78 protein coding genes, 30 tRNA genes, and 4 rRNA genes, with the size of 152,849 and 152,853 bp, respectively. From the two chloroplast genomes, 131 SSRs were identified and 12 different SSRs located in five protein coding genes. The analysis showed the extremely conserved structure of chloroplast genomes with surprisingly little variations at the LSC/IR and SSC/IR boundaries. Moreover, the mean nucleotide diversity was found to be 0.162% for 77 regions, suggesting an extraordinarily low level of sequence divergence. Four highest divergent regions (trnH-psbA, rps14-trnT, petA-psbJ, ccsA-ndhD) with the percentage of nucleotide diversity higher than 0.50% were identified, which had potential use for species identification and phylogenetic studies.ConclusionThis study will facilitate our understanding of population genetics, phylogenetic relationship and plant evolution of Phoebe species.Electronic supplementary materialThe online version of this article (doi:10.1186/s40529-017-0192-8) contains supplementary material, which is available to authorized users.

Summary ‘Living fossils’ are testimonies of long‐term sustained ecological success, but how demographic history and natural selection contributed to their survival, resilience, and persistence in the face of Quaternary climate fluctuations remains unclear. To better understand the interplay between demographic history and selection in shaping genomic diversity and evolution of such organisms, we assembled the whole genome of Cercidiphyllum japonicum, a widespread East Asian Tertiary relict tree, and resequenced 99 individuals of C. japonicum and its sister species, Cercidiphyllum magnificum (Central Japan). We dated this speciation event to the mid‐Miocene, and the intraspecific lineage divergence of C. japonicum (China vs Japan) to the Early Pliocene. Throughout climatic upheavals of the late Tertiary/Quaternary, population bottlenecks greatly reduced the genetic diversity of C. japonicum. However, this polymorphism loss was likely counteracted by, first, long‐term balancing selection at multiple chromosomal and heterozygous gene regions, potentially reflecting overdominance, and, second, selective sweeps at stress response and growth‐related genes likely involved in local adaptation. Our findings contribute to a better understanding of how living fossils have survived climatic upheaval and maintained an extensive geographic range; that is, both types of selection could be major factors contributing to the species' survival, resilience, and persistence.

Parrotia subaequalis is an endangered Tertiary relict tree from eastern China. Despite its important ecological and horticultural value, no transcriptomic data and limited molecular markers are currently available in this species. In this study, we first performed high-throughput transcriptome sequencing of two individuals representing the northernmost (TX) and southernmost (SJD) population of P . subaequalis on the Illumina HiSeq 2500 platform. We gathered a total of 69,135 unigenes for P . subaequalis (TX) and 84,009 unigenes for P . subaequalis (SJD). From two unigenes datasets, 497 candidate polymorphic novel expressed sequence tag-simple sequence repeats (EST-SSRs) were identified using CandiSSR. Among these repeats, di-nucleotide repeats were the most abundant repeat type (62.78%) followed by tri-, tetra- and hexa-nucleotide repeats. We then randomly selected 54 primer pairs for polymorphism validation, of which 27 (50%) were successfully amplified and showed polymorphisms in 96 individuals from six natural populations of P . subaequalis . The average number of alleles per locus and the polymorphism information content values were 3.70 and 0.343; the average observed and expected heterozygosity were 0.378 and 0.394. A relatively high level of genetic diversity ( H T = 0.393) and genetic differentiation level ( F ST = 0.171) were surveyed, indicating P . subaequalis maintained high levels of species diversity in the long-term evolutionary history. Additionally, a high level of cross-transferability (92.59%) was displayed in five congeneric Hamamelidaceae species. Therefore, these new transcriptomic data and novel polymorphic EST-SSR markers will pinpoint genetic resources and facilitate future studies on population genetics and molecular breeding of P . subaequalis and other Hamamelidaceae species.

The genus Croomia (Stemonaceae) is an excellent model for studying the evolution of the Eastern Asia (EA)–Eastern North America (ENA) floristic disjunction and the genetic mechanisms of floral zygomorphy formation. In addition to the presence of both actinomorphic and zygomorphic flowers within the genus, species are disjunctively distributed between EA and ENA. However, due to the limited availability of genomic resources, few studies of Croomia have examined these questions. In this study, we sequenced the floral and leaf transcriptomes of the zygomorphic flowered Croomiaheterosepala and the actinomorphic flowered Croomia japonica, and used comparative genomic approaches to investigate the transcriptome evolution of the two closely related species. The sequencing and de novo assembly of transcriptomes from flowers of C. heterosepala (ChFlower), flowers of C. japonica (CjFlower), and leaves of C. japonica (CjLeaf) yielded 57,193, 62,131 and 64,448 unigenes, respectively. In addition, estimation of Ka/Ks ratios for 11,566 potential orthologous groups between ChFlower and CjFlower revealed that only six pairs had Ka/Ks ratios significantly greater than 1 and are likely under positive selection. A total of 429 single copy nuclear genes (SCNGs) and 21,460 expression sequence tags-simple sequence repeats (EST-SSRs) were identified in this study. Specifically, we identified seven CYC/TB1-like genes from Stemonaceae. Phylogenetic and molecular evolution analyses indicated that these CYC/TB1-like genes formed a monophyletic clade (SteTBL1) and were subject to strong purifying selection. The shifts of floral symmetry in Stemonaceae do not appear to be correlated with TBL copy number.

Aim Several hypotheses are available to predict change in genetic diversity at expanding peripheral ranges. However, empirical evidence to test predictions of the centre–periphery hypothesis (CPH) at contracting range limits is scarce. To address this issue, we assessed spatial patterns of genetic variation, effective population size, and contemporary and historical gene flow in a widespread, Tertiary relict tree species from subtropical China. Location Warm‐temperate deciduous forests of subtropical China. Taxon Emmenopterys henryi (Rubiaceae) Methods We applied kernel density estimation to determine the centre of the species’ geographical range. Using microsatellite markers, we assessed genetic structure and diversity in 36 populations (503 individuals) sampled in the centre and periphery across the species’ range. We further examined both historical and contemporary gene flow. Finally, we applied coalescent methods to simulate population demography. Results In support of CPH predictions, the highest density of E. henryi coincided with the geographical centre of the species’ distribution range, and genetic diversity significantly declined with distance from this range centre. Historical migration from the core to the edge was significantly higher than in the opposite direction, whereas contemporary migration followed an opposite pattern. Central and peripheral populations had similar levels of genetic differentiation. Main conclusions Core‐to‐edge patterns of genetic diversity, but not genetic differentiation, were consistent with the CPH in E. henryi. Also in line with the CPH, historical (but not contemporary) migration from the core to the edge was significantly higher than in the opposite direction. Results suggest that the complex topography in subtropical China and population demographic processes may strongly influence whether CPH predictions are met in different population genetic parameters.