Research Article |
Corresponding author: Kévin J. L. Maurin ( kjlm1@students.waikato.ac.nz ) Academic editor: James W. Byng
© 2020 Kévin J. L. Maurin.
This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Citation:
Maurin KJL (2020) A dated phylogeny of the genus Pennantia (Pennantiaceae) based on whole chloroplast genome and nuclear ribosomal 18S–26S repeat region sequences. PhytoKeys 155: 15-32. https://doi.org/10.3897/phytokeys.155.53460
|
Pennantia, which comprises four species distributed in Australasia, was the subject of a monographic taxonomic treatment based on morphological characters in 2002. When this genus has been included in molecular phylogenies, it has usually been represented by a single species, P. corymbosa J.R.Forst. & G.Forst., or occasionally also by P. cunninghamii Miers. This study presents the first dated phylogenetic analysis encompassing all species of the genus Pennantia and using chloroplast DNA. The nuclear ribosomal 18S–26S repeat region is also investigated, using a chimeric reference sequence against which reads not mapping to the chloroplast genome were aligned. This mapping of off-target reads proved valuable in exploiting otherwise discarded data, but with rather variable success. The trees based on chloroplast DNA and the nuclear markers are congruent but the relationships among the members of the latter are less strongly supported overall, certainly due to the presence of ambiguous characters in the alignment resulting from low coverage. The dated chloroplast DNA phylogeny suggests that Pennantia has diversified within the last 20 My, with the lineages represented by P. baylisiana (W.R.B.Oliv.) G.T.S.Baylis, P. endlicheri Reissek and P. corymbosa diversifying within the last 9 My. The analyses presented here also confirm previous molecular work based on the nuclear internal transcribed spacer region showing that P. baylisiana and P. endlicheri, which were sometimes considered synonyms, are not sister taxa and therefore support their recognition as distinct species.
chimeric mapping reference, chloroplast DNA, internal transcribed spacer, Next Generation Sequencing, off-target reads, Torricellia
Pennantia J.R.Forst. & G.Forst. is the sole genus of the family Pennantiaceae J.Agardh, a member of Apiales that comprises four species in Australasia (
General distribution of the four Pennantia species. TKI = Three Kings Islands. Generated in QGIS 3.0.1 from Google Satellite data obtained through the XYZ Tiles tool (https://mt1.google.com/vt/lyrs=s&x={x}&y={y}&z={z}).
The placement of Pennantiaceae within Apiales has been a matter of debate. Their morphology is consistent with Apiales in the inferior position of their ovary and their low number of carpels (
This study has three goals. (1) To propose the first molecular phylogeny that samples all four species of Pennantia for whole plastid DNA sequences, dated using two Apiales fossils and one secondary calibration. (2) To present and evaluate the relevance of a method I used to generate sequence data for nuclear markers at low marginal cost from the shotgun sequencing of genomic DNA: I mapped reads that were unmapped to the chloroplast DNA reference sequence (“off-target reads”) against a chimeric 18S–26S nuclear ribosomal DNA repeat region reference sequence to build the sequences for a nuclear DNA phylogeny. (3) To use both the chloroplast DNA and nuclear DNA phylogenies to further examine proposals made by
Sampling plan of this study, with voucher information and GenBank accession numbers. Samples sorted alphabetically by order name then species name.
Order | Family | Species | Distribution | Source of plant material or sequence | Herbarium accession # | Voucher or publication | Markers | GenBank accession # |
---|---|---|---|---|---|---|---|---|
Apiales | Araliaceae | Cheirodendron bastardianum (Decne.) Frodin | Marquesas Islands | P | P02800554 | Perlman 19764 | Chloroplast | MT385071 |
Apiaceae | Daucus carota L. | Native to temperate Europe and south-west Asia | GenBank | – |
|
Chloroplast | DQ898156 | |
Torricelliaceae | Melanophylla alnifolia Baker | Madagascar | P | P02529054 | Ranirison 966 | Chloroplast | MT385073 | |
Torricelliaceae | Melanophylla modestei G.E. Schatz, Lowry & A.-E. Wolf | Madagascar | P | P06233571 | Bernard 1700 | Chloroplast | MT385074 | |
Pennantiaceae | Pennantia baylisiana (W.R.B.Oliv.) G.T.S.Baylis | Three Kings Islands/Manawatāwhi (Great Island/Manawa Tawhi) | CHR | CHR 655088 | Maurin 87 | Chloroplast | MT385075 | |
Nuclear | MT434778 | |||||||
GenBank | – | Rotherdam et al. (unpubl.) | Nuclear | EF660531 | ||||
Pennantiaceae | Pennantia corymbosa J.R.Forst. & G.Forst. | New Zealand’s main islands and some neighbouring offshore islands | CHR | CHR 649661 | Maurin 45 | Chloroplast | MT385076 | |
Nuclear | MT434779 | |||||||
GenBank | – | Rotherdam et al. (unpubl.) | Nuclear | EF635468 | ||||
Pennantiaceae | Pennantia cunninghamii Miers | East coast of Australia |
|
CANB869762 | Purdie 9229 | Chloroplast | MT385077 | |
Nuclear | MT434780 | |||||||
GenBank | – | Rotherdam et al. (unpubl.) | Nuclear | EF635470 | ||||
Pennantiaceae | Pennantia endlicheri Reissek | Norfolk Island |
|
CBG8703383 | Telford 10450 | Chloroplast | MT385078 | |
Nuclear | MT434781 | |||||||
GenBank | – | Rotherdam et al. (unpubl.) | Nuclear | EF635469 | ||||
Pittosporaceae | Pittosporum eugenioides A.Cunn. | North and South Islands of New Zealand | CHR | CHR 553618 | Courtney, s.n. | Chloroplast | MT385079 | |
Araliaceae | Raukaua anomalus (Hook.) A.D.Mitch., Frodin & Heads | New Zealand’s main islands | CHR | CHR 649673 | Maurin 57 | Chloroplast | MT385080 | |
Araliaceae | Raukaua edgerleyi (Hook.f.) Seem. | New Zealand’s main islands | CHR | CHR 655508 | Maurin 103 | Chloroplast | MT385081 | |
Araliaceae | Raukaua simplex (G.Forst.) A.D.Mitch., Frodin & Heads | New Zealand’s main islands, Auckland Islands | CHR | CHR 437312 | Sykes 42/87 | Chloroplast | MT385082 | |
Araliaceae | Schefflera actinophylla (Endl.) Harms | Northern and north-eastern coast of Australia |
|
CANB874342 | Lepschi 7083 | Chloroplast | MT385083 | |
Araliaceae | Schefflera digitata J.R.Forst. & G.Forst. | New Zealand’s main islands | CHR | CHR 649676 | Maurin 60 | Chloroplast | MT385084 | |
Torricelliaceae | Torricellia tiliifolia DC. | China, eastern Himalaya | GenBank | – |
|
Chloroplast | NC040944 | |
Aquifoliales | Aquifoliaceae | Ilex paraguariensis A.St.-Hil. | South America | GenBank | – |
|
Chloroplast | KP016928 |
Asterales | Argophyllaceae | Corokia cotoneaster Raoul | New Zealand’s main islands | CHR | CHR 655097 | Maurin 96 | Chloroplast | MT385072 |
Dipsacales | Caprifoliaceae | Dipsacus asper Wallich ex Candolle | China, south-east Asia | GenBank | – |
|
Chloroplast | MH074864 |
Paracryphiales | Paracryphiaceae | Quintinia verdonii F.Muell. | Eastern Australia | GenBank | – |
|
Chloroplast | MK397891 |
DNA from the samples of Pennantia corymbosa, Raukaua anomalus (Hook.) A.D.Mitch., Frodin & Heads and Schefflera digitata J.R.Forst. & G.Forst. was extracted using a CTAB-based protocol (
Genomic DNA libraries of Pennantia corymbosa, Raukaua anomalus and Schefflera digitata were prepared using Illumina Nextera DNA Library Prep kits, following the manufacturer’s instructions (Reference Guide, #15027987 v01, January 2016) except that I halved the quantities of reagents and the target amount of input DNA. Libraries of the other samples were prepared using Illumina TruSeq Nano DNA Library Prep kits, according to the manufacturer’s instructions (Reference Guide, # 15041110 Rev. D, June 2015), again using halved reagent quantities and target input DNA; genomic DNA was fragmented using a Covaris ME220 Focused-ultrasonicator (settings: 75 s duration – 40 W peak power – 25% duty factor – 50 cycles per burst). The concentration and size range of libraries were measured with a LabChip GX Touch HT (Perkin Elmer). Libraries were enriched for chloroplast DNA using a custom MYBaits kit (Arbor Biosciences, Ann Arbor) modified from
Reads were first trimmed using Trimmomatic v. 0.38 (
The resulting sequences, except Melanophylla modestei G.E. Schatz, Lowry & A.-E. Wolf, were of good overall quality (Suppl. material
In the absence of a complete 18S–26S nuclear ribosomal DNA repeat region for Apiales, I built a chimeric 18S–26S nuclear DNA repeat region from several GenBank sequences. I concatenated the 18S rRNA sequence of Melanophylla alnifolia Baker (AJ236002), the ITS1, 5.8 S RNA, and ITS2 sequences of Pennantia cunninghamii (EF635470), and the 26S rRNA sequence of Pittosporum fairchildii Cheeseman (AF479192), in that order. The structure of the resulting chimeric 18S–26S nuclear DNA repeat region is provided in Suppl. material
The quality of the resulting assemblies was rather variable. There was no clear relationship between the number of reads available to map and the number of reads actually mapped to the chimeric reference (Suppl. material
Sixty protein-coding sequences (CDS, 46,051 sites) from the long and short single copy regions were used for the chloroplast DNA analyses (see list in Suppl. material
Phylogenetic analyses were conducted with the BEAST suite v. 2.5.2 (
The chloroplast DNA phylogeny was calibrated using two fossils and one secondary calibration. Firstly, I assigned the age of the earliest confirmed fossils of Torricellia, which are ca. 48 My old (
The robustness of the Bayesian inference of tree topology for the phylogenies resulting from both the chloroplast DNA and the nuclear DNA sequence data was assessed with a maximum likelihood approach. RAxML v. 8.2.12 (
Finally, the six resulting trees (chloroplast DNA or nuclear DNA, with BEAST2/Birth-Death model, BEAST2/Yule model or RAxML) were first formatted in FigTree v. 1.4.4 (
The combination of the chains run under the Birth-Death model or the Yule model resulted in an Effective Sample Size (ESS) > 200 for all their parameters. The tree had the same topology and was very well supported within the ingroup Apiales under both models, all the node posterior probabilities (PP) being equal to 1. Moreover, the same topology was obtained for the chloroplast DNA tree built with RAxML, with 100% bootstrap support within Apiales. The tree resulting from the Birth-Death model is shown in Fig.
In the phylogeny presented in Fig.
Dated chloroplast DNA BEAST 2 phylogeny of Pennantia, under the Birth-Death model. Mean node age and 95% HPD (in My) is given in the table embedded in the figure under the corresponding letter code. 95% HPD is also represented by blue bars. All node posterior probabilities are equal to 1 except if indicated otherwise. The calibrated nodes (see text) are indicated by red dots.
The chains yielded an ESS far greater than 200 even before they were combined under both the Birth-Death model and the Yule model. The resulting tree showed the same topology with comparable PP under both models, although the PP under the Yule model tended to be slightly lower than under the Birth-Death model. The topology of the tree produced from the RAxML analysis was congruent with the topology of the BEAST2 trees, with bootstrap values of 100% except for the node placing the two samples of P. corymbosa and P. endlicheri as sister to each other (bootstrap = 88%). For consistency with the chloroplast DNA phylogeny, I draw conclusions regarding the nuclear DNA phylogeny primarily by examining the Birth-Death model tree (Fig.
Undated 18S–26S nuclear DNA repeat region BEAST 2 phylogeny of Pennantia, under the Birth-Death model. The tree was rooted to make P. cunninghamii sister to the other species of Pennantia, in accordance with the chloroplast DNA tree and the ITS tree of
The percentage of identical sites between the two samples of each species was ≥ 98.7%. There were relatively few parsimony-informative sites in the nuclear DNA alignment: only 35 out of 538 (6.5%) sites in the ITS1/ITS2 partition and 0 out of 272 in the rRNA partition. The two samples of each species were recovered as sisters, usually with strong support: PP = 1 for P. cunninghamii and P. baylisiana, PP = 0.97 for P. corymbosa, but PP = 0.75 only for P. endlicheri. Moreover, the topology of this tree was congruent with that of the tree based on chloroplast DNA (Fig.
Phylogenies based on chloroplast DNA markers and the 18S–26S nuclear DNA repeat region indicate the same relationships among the four species of Pennantia. They are also congruent with the ITS phylogeny of
The age of the MRCA of Pennantiaceae and Torricelliaceae (which is the crown age of Apiales) was estimated about 86.7 My, with an HPD of [73.3,100.9] My. This mean estimate is consistent with some of the previous dated phylogenies that include this MRCA: 73.6 My (
The mean crown age of Pennantiaceae was estimated to be 9.5 My with an HPD of [2.6,19.5] My, which is slightly older than the previous estimate for Pennantia of 6.6 My with an HPD of ca. [1.6,15.8] My suggested by
The phylogenies presented here significantly supported Pennantia baylisiana being a distinct species to Pennantia endlicheri, corroborating the conclusions
The present phylogenies also supported the placement by
The analysis of chloroplast genome sequences supports previous phylogenetic results based on nuclear DNA in suggesting that Pennantia cunninghamii is sister to the rest of the genus. Moreover, it strongly supports previous nuclear DNA analyses in placing P. baylisiana as sister to the clade P. endlicheri + P. corymbosa rather than sister to P. endlicheri alone, with which it has sometimes been considered conspecific (e.g.
I would like to thank Chris Lusk and Rob Smissen for commenting on the draft manuscripts; Peter de Lange and Porter Lowry II for commenting on the submitted manuscript; Otari Native Botanic Garden, the Pukemokemoke Bush Trust, the Department of Conservation, the Maungatautari Ecological Island Trust and the iwi Waikato Tainui, Ngāti Rangi and Mōkai Pātea for sampling permits and agreements in New Zealand; the herbaria CHR,
Figs S1–S5; Tables S1–S3
Data type: figures and tables
BEAST2 and RAxML files
Data type: molecular data