Abstract
Dipelta (Caprifoliaceae) is a Tertiary relict genus endemic to China, comprising three species with horticultural and medicinal values. For the lack of genomic information, interspecific relationships and divergence times in the genus remain unresolved. In the present study, we assembled and characterized the complete plastomes, the partial mitogenomes, and nuclear internal transcribed spacer (ITS) fragments from genome skimming datasets of 14 Dipelta individuals. The plastomes were conserved in genomic structure, gene order, and gene content, but with highly variable repeat sequences. Three genes (rpl23, ycf1, ycf2) were examined with positive selection, and nine divergent hotpot regions (psbL, accD, rpl23, ycf2, ycf3, rbcL-accD, trnI-CAU-ycf2, ndhH-rps15, and rps18 intron) were potentially valuable for DNA barcodes. Contrasted to the variability in plastome sequences, mitogenomes contained 12 protein-coding genes with limited indels and nucleotide substitutions, and no gene was found under positive selection. Genes in organellar genomes tended to have a similar pattern of codon usage bias, with a preference of A/U ending codons. Phylogenetic trees constructed with plastomes, mitogenomes, and ITS sequences consistently supported that Dipelta was monophyletic, and Dipelta elegans was sister to the other two taxa. Interspecific divergences were estimated at about 33–37 Ma in the Eocene/Oligocene boundary, suggesting the paleo-endemism of the extant species as “living fossils” of the East Asian Flora. Our study well-exhibited that genome skimming could provide valuable genomic information to elucidate the evolutionary history of the complex group in a cost-efficient way.
Introduction
Dipelta (Caprifoliaceae), mainly distributed in southwest and central China, is a Tertiary relict woody genus with three extant species, Dipelta floribunda, Dipelta yunnanensis, and Dipelta elegans (Figure 1) (). All species in the genus are of great value in landscape design for their peculiar bell-shaped flowers. D. yunnanensis and D. floribunda are also used as traditional Chinese medicines to treat pruritus, measles, and damp-heat syndrome (). Due to the excessive deforestation and continuous deterioration of habitats, D. yunnanensis has been listed as a vulnerable species by the International Union for Conservation of Nature. D. elegans is also classified as a rare and endangered plant in the China Species Red List for its extremely small population size (). Studies on Dipelta are generally involved in the phylogenetic analysis of Caprifoliaceae or Dipsacales (; ; ; ; Xiang et al., 2020). However, interspecific relationships and diversification dating of the genus have not been clearly clarified. Either D. yunnanensis or D. elegans was possibly a sister to the other two taxa (; ). The low resolution of chloroplast fragments even produced a completely polytomous phylogenetic tree of the genus (). According to fossil records in the middle or late Eocene and its historical distribution, Dipelta was considered as a paleo-endemic genus in the north temperate flora (). However, when molecular dating was used to estimate the divergence times among the extant species, controversial conclusions were drawn in previous works. Chloroplast DNA fragments suggested that infrageneric taxa diverged in the Miocene (9.92 Ma) (), indicating the paleo-endemism of the living species. By contrast, argued that the genus was possibly neoendemic, and speciation occurred in the Pleistocene (0.66 Ma). To elucidate the phylogenetic incongruence and divergence issues, additional genomic data are indispensably needed.
FIGURE 1
Among the three genomes in plants, plastome is a highly conserved circular DNA molecule with a size of 120–170 kb () and has been widely used in studies of species authentication, population genetics, and genetic engineering (). The internal transcribed spacer (ITS) of nuclear ribosomal DNA (nrDNA) is a well-known marker for phylogenetic works due to its high discrimination power at the species level (). Mitogenome has been neglected in plant phylogenetic analysis for its low nucleotide substitution rate, flexible genome structure, and potential gene transfer among intracellular organelles (Zhao et al., 2018; ). Recent works revealed that gene regions of mitogenome could also offer valuable phylogenetic information (; ). Furthermore, comparative analysis of plastomes and mitogenomes would provide an illuminating insight into the architectures and evolution of two genomes. Traditionally, PCR-based methods are involved in obtaining DNA sequences. For the lack of genome information, it possibly fails to get genetic data for non-optimal primers or poor quality of DNA. With the rapid advances in high-throughput sequencing technologies, sequence information of three genomes might be gained simultaneously through the genome skimming approach (; ). This low-pass genome sequencing method generates high-copy fractions of total genomic DNA, including the ribosomal DNA, plastome, mitogenome, and other multicopy elements, becoming an efficient tool for phylogenomics and biodiversity researches (). In the present study, the complete plastomes, partial mitogenomes, and nuclear ITS sequences from 14 individuals of three Dipelta species (Figure 1 and Supplementary Table A1) were assembled and annotated through the genome skimming approach. The genomes were compared and subsequently used for phylogenetic analysis. Based on these data, we aim to characterize the variability of organellar genomes, elucidate interspecific phylogenetic relationships, and estimate divergence times of the three Dipelta species.
Materials and Methods
Sample Collection, Sequencing, and Annotation
Fourteen Dipelta individuals (two of D. elegans, six of D. yunnanensis, and D. floribunda, respectively) were collected throughout their natural distribution ranges (Supplementary Table A1 and Figure 1). All voucher specimens were deposited in the Southwest Forestry University Herbarium. Genomic DNA was isolated with the CTAB method from silica gel dried leaves and then sequenced using the Illumina Hiseq 2500 platform in Novogene, Co. Ltd. Raw reads were quality trimmed using the NGS QC Toolkit v2.3.3 () with default cut-off values. The clean reads were mapped to the references using MITObim v1.8 () and annotated with DOGMA (Wyman et al., 2004). Genome annotation followed the workflow in our previous work (). The reference sequences were from D. floribunda for plastome (MG738670) () and ITS (FJ745389) (). There is no mitogenome information available in Caprifoliaceae. Chrysanthemum indicum (MH716014) () was treated as the reference for mitogenome analysis. All sequences were deposited in GenBank with the accession numbers (Supplementary Table A1).
Identification of Repeat Sequences
A perl script MISA () was used to detect simple sequence repeats (SSRs) loci in the organelle genomes. In addition, the minimum numbers of repeats were set to 10, 5, 4, 3, 3, and 3 for mono-, di-, tri-, tetra-, penta-, and hexa-nucleotides, respectively. The online REPuter program () was employed to identify dispersed and palindromic repeats by setting parameters as (1) Hamming distance of 3; (2) 90% or greater sequence identity; and (3) a minimum repeat size of 30 bp. Additionally, the tandem repeats were screened using the online Tandem Repeats Finder program () with the alignment match, mismatch, and indels of two, seven, and seven, respectively. The minimum alignment score and maximum period size were constrained to 80 and 500.
Comparative Genome and Positively Selected Gene Analysis
Mauve Alignment () was performed to reveal the rearrangement among plastomes. Sequences were aligned and compared with mVISTA () to identify the variation of coding and non-coding regions in plastomes by using one plastome of D. floribunda as a reference. The protein-coding genes of plastomes and mitogenomes were used for determining the codon usage pattern. Relative synonymous codon usage (RSCU) was estimated with CodonW (). In the absence of codon usage bias, the RSCU values would be 1.00. To test the positively selected genes, a maximum likelihood phylogenetic tree was constructed by using RAxML v 7.2.8 ().
The variations of non-synonymous (dN) and synonymous (dS) nucleotide sites and their ratio (dN/dS or ω) are valuable indicators for the estimation of evolutionary rates and natural selection. We used the codon substitution models to estimate selection pressures, which allowed the ω ratio to vary among sites in all phylogenetic branches. The values of dN, dS, and ω were calculated using the Codeml program in PAML v4.7 (Yang et al., 2005), and the selection pressure was estimated by the site-specific model (seqtype = 1, model = 0, NS sites = 0, 1, 2, 3, 7, 8). Likelihood ratio test (LTR) was performed to estimate the candidate sites of positive selection.
Phylogenetic Analysis and Divergence Time Estimation
Phylogenetic analysis of the Dipelta species was separately conducted using three datasets (the plastomes, 12 mitochondrial genes, and nuclear ITS sequences). Besides Dipelta, six Dipsacales species (three Caprifoliaceae and three Adoxaceae species) were selected to construct plastome and ITS phylogenetic trees with an outgroup of Guizotia abyssinica (Supplementary Table A1). For the lack of mitogenome information, no other Dipsacales species was included in the present study. The sequences were aligned using MAFFT with the default parameter sets (). The maximum likelihood (ML) analysis was performed using RAxML v 7.2.8 () with 1,000 bootstrap replicates. The best-fit model (GTR + G) was determined with Modeltestv3.7 (). We estimated the divergence time of Dipelta species using a Yule process speciation prior and GTR substitution model with the uncorrelated relaxed clock in BEAST program (). We set the fossil calibrations of Dipelta and Kolkwitzia with mean = 40 Ma, SD = 1.0 (; ), and the Dipsacales constraint prior was set as a normal distribution, mean = 80 Ma, SD = 1.0 (). The chain length of MCMC was set to 100,000,000 generations, and the sampling frequency was set to 1,000. The effective sample size (>200) was evaluated in Tracer v1.6. The posterior sample of trees was summarized using TreeAnnotator v1.8.0 with the first 10% of the samples discarded as burn-ins ().
Results
Features of ITS and Organellar Genomes
Fourteen Dipelta individuals (two for D. elegans, six for D. yunnanensis, and D. floribunda, respectively) were sequenced with genome skimming technology, generating about 2G shallow-coverage genomic data for each individual. The sequencing depths were at least 320, 250, and 2,449 × for plastomes, mitogenome, and ITS, respectively (Supplementary Table A2). Nuclear ITS fragments of D. yunnanensis and D. floribunda were in similar sizes (613 bp or 61 bp), while D. elegant counterparts were longer (623–639 bp) with higher GC contents (Supplementary Table A2). The aligned ITS fragments, containing highly conserved 5.8S rDNA of 160 bp, were totally 639 bp in length with 12 variable sites and 10 parsimony informative sites. Partial mitogenomes of Dipelta were also determined in this study, with the sizes of 218,040 bp (D. floribunda_GSWX)–242,194 bp (D. yunnanensis_YNKM). Twelve intact mitochondrial genes (atp1, atp9, mttb, atp8, cox 3, atp 6, rpl5, rps13, ccmb, rpl10, matR, and cob) were annotated (Supplementary Table A1). The aligned full-length sequences ranged from 351 bp (rps13) to 1,533 bp (atp1). Among these genes, the atp8 gene possessed several indels among individuals, showing the highest variable sites (50.6%), but no nucleotide substitution was found in mttB, atp8, cox3, and ccmB genes (Supplementary Figure A2). The GC contents of the mitochondrial genes ranged from 37.1% (atp6) to 52.0% (matR). When the 12 annotated genes were combined, the concatenated sequence was 9,855 bp in length with 199 variable sites (Supplementary Table A1).
The plastome size ranged from 155,114 bp (D. floribunda_GSWX) to 155,948 bp (D. yunnanensis_SCYX) (Supplementary Figure A1 and Supplementary Table A2). Each plastome exhibited a typical quadripartite structure with a pair of inverted repeat (IR) regions (23,201–23,496 bp) separated by a small single-copy region (SSC, 18,952–19,117 bp) and a large single-copy region (LSC, 83,733–89,975 bp) (Supplementary Table A2). The GC contents in the single-copy (SC) regions of Dipelta were lower than IR counterparts, which was ascribed to the high GC contents of rRNA genes in IRs (). There were 128 genes in each plastome, consisting of 82 protein-coding genes, 37 tRNA genes, and eight rRNA genes (Table 1 and Supplementary Table A2). Fifteen genes situated in the IR regions (rrn4.5, rrn5, rrn16, rrn23, trnA-UGC, trnI-CAU, trnI-GAU, trnL-CAA, trnN-GUU, trnR-ACG, trnV-GAC, rps7, rps12, ndhB, ycf2), while the rpl23 gene located in the junction of the IRb and LSC region. Moreover, 15 intron-containing genes were determined, 12 of which had single intron and three (rps12, rps18, ycf3) had two introns. The rps12 gene was trans-spliced into three exons, two in IR and one in the LSC region, as reported in other land plants (; ).
TABLE 1
| Gene group | Gene name |
| Ribosomal RNA genes | rrn4.5c, rrn5c, rrn16c, rrn23c |
| Transfer RNA genes | trnA-UGCac, trnC-GCA, trnD-GUC, trnE-UUC, trnF-GAA, trnG-UCC, trnH-GUG, trnI-CAUc, trnI-GAUac, trnK-UUUa, trnL-CAAc, trnL-UAG, trnL-UAAa, trnM-CAU, trnfM-CAU, trnN-GUUc, trnP-UGG, trnQ-UUG, trnR-UCU, trnR-ACGc, trnS-UGA, trnS-GGA, trnS-GCU, trnT-GGU, trnT-UGU, trnV-GACc, trnV-UACa, trnW-CCA, trnY-GUA |
| Small subunit of ribosome | rps2, rps3, rps4, rps7c, rps8, rps11, rps12bc, rps14, rps15, rps16a, rps18b, rps19 |
| Large subunit of ribosome | rpl2a, rp114, rp116, rp120, rp122, rp123, rp132, rp133, rp136 |
| DNA-dependent RNA polymerase | rpoA, rpoB, rpoC1a, rpoC2 |
| Translational initiation factor | infA |
| Subunits of photosystem I | psaA, psaB, psaC, psaI, psaJ, ycf3b, ycf4 |
| Subunits of photosystem II | psbA, psbB, psbC, psbD, psbE, psbF, psbH, psbI, psbJ, psbK, psbL, psbM, psbN, psbT |
| NADH oxidoreductase | ndhAa, ndhBac, ndhC, ndhD, ndhE, ndhF, ndhG, ndhH, ndhI, ndhJ, ndhK |
| Subunits of cytochrome | petA, petB, petD, petG, petL, petN |
| Subunits of ATP synthase | atpA, atpB, atpE, atpFa, atpH, atpI |
| Large subunit of Rubisco | rbcL |
| Maturase | matK |
| Envelope membrane protein | cemA |
| Subunit of acetyl-CoA | accD |
| C-type cytochrome synthesis gene | ccsA |
| Protease | clpP |
| Conserved open reading frames | ycf1, ycf2ac |
Genes in the 14 Dipelta chloroplast genomes.
aGene with one intron.bGene with two introns.cGenes located in the inverted repeats.
Repetitive Element and SSR Analysis
Repetitive sequences have great values in the study of plant evolution and population genetics. We investigated five category repeats (palindromic, forward, tandem, complement, and reverse repeats) in Dipelta species. Plastome analysis showed that D. floribunda possessed the greatest number of repeats with the average number of 264, and D. yunnanensis had the lowest (256). Forward repetitive sequence was the most frequent, followed by palindromic and tandem repeats in all species (Figure 2). Complement repeats were only found in D. elegans, and reverse repeats were only presented in some individuals of D. floribunda, both of which were not detected in D. yunnanensis (Figure 2). The majority of the repeats were 30–49 bp in length, distributed in intergenic (IGS) or intron regions (Figure 2 and Supplementary Table A4). We also identified six types of perfect SSRs (mono-, di-, tri-, tetra-, penta-, and hexa-nucleotide repeats) in 14 Dipelta plastomes. The number of mononucleotides accounted for the largest proportion (58%), while the hexanucleotide was the least (5%), and the pentanucleotide only occurred in D. floribunda (Figures 3A,C). Most of SSRs were located in the LSC and IGS regions, which were twice as much as protein-coding regions (Figures 3B,D). Among the repeat units, mononucleotide A/T took up the largest proportion, and all of the dinucleotide repeats were comprised of A and T bases (Figure 3E). Moreover, the ATATTA repeat units were only found in D. yunnanensis; the repeat sequences of TATAC, TAT, AAT only appeared in some individuals of D. floribunda. In contrast, the mononucleotide repeats of cytosines were not detected in D. elegans (Figure 3E).
FIGURE 2
FIGURE 3
For mitogenomes, only two forward tandem repeats were found in the atp9 gene of D. floribunda. In addition, penta-nucleotide repeats of (ACTAG)3 were presented in the matR gene of all individuals. No SSR was found in the genes of all mitogenomes.
Comparative Plastome Analysis
The colinear analysis showed that there was no large-scale structural rearrangements among Dipelta plastomes. The overall sequence identities of 14 plastomes were plotted by using the mVISTA program. As expected, high sequence similarity was observed across the plastomes, and the non-coding regions exhibited higher divergence than the coding counterparts (Figure 4). Three genes (ycf2, rpl23, and accD) showed the highest variability with a percentage of variation larger than 0.3%. For the non-coding regions, three intergenic spacers (trnI-CAU-ycf2, rbcL-accD, and ndhH-rps15) and rps18 intron were more divergent (>45%) than others (Figure 5). These divergence hotspots would provide useful markers for studies at the population level. Interestingly, although some genes (such as ycf3 and nadF) were less variable than the genes mentioned above, their polymorphic sites could still offer valuable information for Dipelta species discrimination (Supplementary Table A3).
FIGURE 4
FIGURE 5
We compared the SC/IR boundaries and their adjoining genes of 14 plastomes. The junctions of SSC/IR were conserved with a few bases shift. The contraction and expansion of IRs were mainly located in the LSC/IR boundary. The rpl23 gene straddled the LSC/IRb region and extended to the IR region with 6∼183 bp among different individuals, leading to a distinct shift of the LSC/IRb boundary. The flexibility of LSC/IRa was ascribed to the changeable position of the trnI-CAU gene in IRa with 171–348 bp away from the boundaries (Figure 6).
FIGURE 6
Codon Usage Pattern and Adaptive Selection Analysis
Here, we analyzed codon usages of the protein-coding genes in the plastomes and mitogenomes from three Dipelta species. The number of encoded codons ranged from 25,065 to 25,152 in plastomes and 3,256–3,284 in mitogenomes (Supplementary Table A4). Leucines and cysteines were the most and least abundant amino acids, respectively. UGC for cysteine was the least frequently used codon in both genomes. The codon with the highest frequency was AUU for isoleucine in plastomes, but UUU for phenylalanine in mitogenomes. Codons with A/U at the third nucleotide position were used more frequently than G/C according to the estimation of RSCU values (RSCU > 1) in both genomes (Supplementary Table A4). For instance, 29 of the 30 preferred codons in plastomes (RSCU > 1) were ended with A/U, while only three did so among the 32 less frequently used codons (RSCU < 1). The A/U bias at the third position of codons could also be identified by the AT contents of codons. The mean AT values were 53.5, 61.3, and 68.9% for the first, second, and third positions in plastomes, while the corresponding values were 50.1, 57.4, and 59.7% in mitogenomes, respectively.
The 77 common protein-coding genes in plastomes of Dipelta species were used for positive selection analysis (Supplementary Table A5). The genes of ycf1, rpl23, and ycf2 potentially had experienced positive selection with 1, 2, and 16 selective sites, respectively (Supplementary Table A5). By contrast, no positive selection site was detected in all 12 mitochondrial genes.
Phylogenetic Analysis and Molecular Dating
Phylogenetic position and interspecific relationships of Dipelta were analyzed with the complete plastomes, ITS sequences, and 12 mitochondrial gene sequences, respectively (Supplementary Table A1). Two deep lineages of Caprifoliaceae s.l. and Adoxaceae in Dipsacales were well-supported in the maximum likelihood (ML) tree based on plastomes and ITS sequences as consistently clarified in the previous studies (; Xiang et al., 2020) (Figure 7). Organellar genomic data demonstrated that Dipelta was a monophyletic group closely related to Kolkwitzia. D. elegans was a sister to D. yunnanensis and D. floribunda. Contrasted to the monophyly of each species in the phylogenetic trees of organellar genomes (Figures 7A,B), two individuals of D. yunnanensis were clustered with D. floribunda in the ITS tree (Figure 7C), indicating the potential introgression between these two species.
FIGURE 7
We calculated the molecular divergence time of nine Dipsacales species based on the complete plastomes (Figure 8). The divergence time of the Adoxaceae and Caprifoliaceae s.l. was about 81.2 Ma (95% highest posterior density, 95% HPD = 77.30–82.73 Ma), suggesting the early Campanian divergence of these two families. Dipelta split from Kolkwitzia in the Eocene, with an estimated age of 38.69 Ma (95% HPD = 36.46–43.56 Ma). The divergence of extant species in Dipelta was estimated at about 33–37 Ma (95% HPD = 21.21–39.58 Ma) in the Eocene/Oligocene boundary. The diversification of Dipelta was dated in the Miocene, as so did the genera in Adoxaceae (Figure 8).
FIGURE 8
Discussion
Genome skimming is an effective approach to produce a large-scale of genomic data for studies in DNA barcoding, biodiversity, and phylogenomics (; ; ). By using this technology, we determined and compared the complete plastomes, partial mitogenomes, and nuclear ITS fragments of 14 individuals from three Dipelta species. The 2G raw data generated abundant DNA sequences of three genomes with high sequencing depth. These high-copy fractions of genomic DNA are beneficial to deal with the complex phylogenetic issues unresolved by the traditional DNA sequence fragments (; ).
Comparison of Genomic Variation
The plastomes in Dipelta were highly conserved in gene numbers, gene orders, and GC contents, with no structural rearrangements. However, the sizes of plastomes were variable within and among species. The contraction and expansion of IRs, as a common evolutionary phenomenon, was regarded as an index of length variation in the angiosperm plastomes (). In the 14 Dipelta plastomes, the SC/IR boundaries showed similar characteristics with slight shifts in gene positions (Figure 2). The variation of SC/IR borders was considered as a product of the dynamic and random expansion of IRs (). To evaluate the potential impacts of the junction changes, we analyzed the relationships between the SC/IR boundaries and divergence hotspot regions in 14 plastomes. Contrast to other genes flanking the junctions, the rpl23 gene spanned the LSC/IR boundary and also had high nucleotide variation, indicating the influence of the shift of the border regions (Figures 5, 6) (). The SC regions had a higher sequence divergence than IRs, which might be contributed to copy correction between IR sequences by gene conversion ().
Repetitive sequences also played an important role in sequence divergence (Yi et al., 2013). We characterized both complex and simple repeat sequences in the 14 individuals. High diversity in length and nucleotide components was presented in the genus (Figures 2, 3). These abundant repeat sequences provided valuable markers for systematic and population studies (Yi et al., 2013). Interestingly, some repeat sequences tended to exhibit species-specific characters. Component and distribution pattern of SSR were quite similar in D. floribunda and D. yunnanensis. The complement repeats only appeared in D. elegans, and neither the complement repeats nor the reverse repeats were found in D. yunnanensis. The reliability of these repeat sequences for species discrimination and phylogenetic analysis is worthy of clarifying in future works.
Among the highly variable genes in Dipelta, ycf2 generally exhibited a high sequence divergence in most of the angiosperms, which was treated as an alternative marker for investigating angiosperm relationships (). The divergence of rpl23 gene was specific in Dipelta. The accD gene was considered as a pseudogene in all plastomes of Dipelta when the annotation was followed the previous study (). This gene encodes a subunit of acetyl-CoA-carboxylase, which participates in fatty acid synthesis, playing an important role in leaf development (). Its pseudogenization had been reported in several species of Caprifoliaceae (; ). Loss or transfer of accD was also known in some taxa (; ). When examining the complete CDS sequences of Caprifoliaceae and Adoxaceae, we found that the accD genes in Caprifoliaceae had slight shorter open-reading-frames (ORF) with highly variable 5′-ends and conserved 3′-ends. If being annotated according to plastomes in Adoxaceae, the accD gene with complete start and stop codon would be aligned with numerous indels at the 5′-end and tended to be treated as a pseudogene. Therefore, we proposed that the accD gene in Caprifoliaceae was functional and annotated in error due to the divergent 5′-ends. However, further studies with more plastome samples are necessarily needed to test its function and divergence in Dipsacales.
Contrasted sharply with the variability of plastomes, mitochondrial genes were considerably conserved. Only 17 polymorphic sites were presented in the 9.9-kb-long sequences. Mitogenome generally contains many repetitive elements and demonstrates frequent gene gain/loss and genomic rearrangements (). However, its nucleotide substitution rate in the coding regions is rather low. All 12 mitochondrial genes involved in the study are protein-coding genes, which results in the conserved sequences of the mitogenome. The aligned atp9 gene was the most variable fragment with several long indels. High gene diversity of atp9 was also detected in the natural population of wild carrot (). However, when blasted in NCBI, besides annotated atp9 per se, the sequence could also match the intergenic spacer in the reference mitogenome with high expected values. Therefore, further works are needed to test whether the atp9 gene is functional.
Codon Bias and Adaptive Selection
The codon usage pattern is essential for cellular function and is considered as an important factor for species divergence and molecular evolution (). In plastomes and mitogenomes of Dipelta, each individual showed a similar codon usage pattern according to the RSCU values (Supplementary Table A4). However, preferred codons differentially existed between the two cytoplasmic genomes. As an example, UAA for the stop codon was preferred in plastomes, while a bias to UGA was favored in mitogenomes. Preferred codon usage bias was found in ACC for threonine and GCC for alanine in mitogenomes, both of which were less used in plastomes. The difference in favored codon usage pattern was closely related to the origin and evolutionary history of two organellar genomes (Xu et al., 2015). In Triticum aestivum, neutral mutation and translation coupled factors played a critical role in shaping the codon usage pattern of mitogenome. For plastomes, selected constrain was potentially a major factor leading to the codon usage bias (Zhang et al., 2007). Despite the same tendency of A/U at the third codon position, the AT contents of the third position in mitogenomic codons were illuminatingly lower than those in plastomes (59.7 vs. 68.9%). The difference of codon GC usage between two genomes could be ascribed to several factors, such as GC contents of genes, mutation bias, natural selection, transcription levels, etc. (Xu et al., 2015; ).
Positive selection analysis was performed for protein-coding genes of plastomes and mitogenomes. Only three chloroplast genes, rpl23, ycf1, and ycf2, were identified under positive selection pressure. The rpl23 gene, encoding the large subunit ribosomal protein 23, was also one of the divergent hotspot regions. The ycf1 and ycf2 genes, the two largest chloroplast genes encoding the envelope membrane protein translocon in the chloroplast (), are essential for cell survival in plants (). They generally evolved at rapid mutation rates (Zhou et al., 2019) and were reported as selected genes in several species of Caprifoliaceae (; ). Three allopatric Dipelta species possess heterogeneous habitats (). These positively selected genes may have played significant roles for Dipelta species to adapt to the diverse environments.
Phylogenetic Relationships and Divergence Age Estimation
Interspecific relationships of the small endemic genus of Dipelta were still unclear (; ). In this study, we successfully applied the plastome, mitogenome, and nuclear ITS sequences to elucidate the phylogenetic issues of Dipelta. Our result suggested that Dipelta was monophyletic and closely related to Kolkwitzia, as identified in systematic works of Caprifoliaceae (; ; ). Within the genus of Dipelta, all data supported that D. elegans was a sister to the other two species. This was also supported by SSR and AFLP data (), but inconsistent with chloroplast fragment sequences, in which D. yunnanensis was the earliest divergent species (). The incongruence was mainly ascribed to the limited sampling number (one individual per species) and chloroplast data (nine fragments) in the latter study. Dipelta elegans is the most restricted species distributed only in the north Qinghai-Tibet Plateau (QTP) of China. D. elegans has unique morphological traits, i.e., the largest epicalyx and the smallest nectary bulge with the entire calyx (). It also harbors many species-specific nucleotide components shared with Kolkwitzia amabilis, which might be ancestral rather than recent derived or convergent (Supplementary Table A3). Liu and colleagues proposed that D. elegans retreated to the current distribution area during the glacial period as several species in QTP (Xu et al., 2010). The widespread allopatric distribution of D. yunnanensis and D. floribunda was an outcome of population expansion in the opposite direction during the period of Pleistocene (). In the nuclear ITS tree, two individuals of D. yunnanensis were grouped with D. floribunda (Figure 7C). These individuals of D. yunnanensis are located in the marginal region of its natural range adjacent to D. floribunda. The incongruence between nuclear and cytoplasmic phylogenetic trees could be explained as interspecific introgression in the past. Introgressive hybridization is commonly found in their sympatric zone () and possibly occurred before the last interglacial period ().
The divergence times about Caprifoliaceae species have been estimated by using the complete chloroplast genomes with fossil calibrations (; ). We uncovered the split of the Adoxaceae and Caprifoliaceae s.l. at about 81.2 Ma at the Cretaceous/Tertiary boundary, similar to the previous study () but younger than that announced by . The dating bias was partly due to the difference in samples/data and fossil constraints (). Dipelta originated in the Eocene (36.46–43.56 Ma) () and diversified in the Eocene/Oligocene boundary (33–37 Ma) (Figure 8), supporting the hypothesis of pre-Oligocene intergeneric divergence and post-Eocene infrageneric diversification in Caprifoliaceae (). The divergence age of Dipelta species was much older than those recently reported (; ). Interestingly, Dipteronia, another Tertiary relict woody genus with two allopatric species, has similar geographical distribution pattern to Dipelta: one species locates in Central China, like D. floribunda; the other was restricted to Southwest China, as D. yunnanensis. The interspecific divergence within Dipteronia was dated back to the early Miocene (31.19 Ma) () or the early Eocene (52.7 Ma) (). The unexpectedly identical divergent ages indicated that the diversification of these two Tertiary relict genera was potentially traced to a common history event, the global climate transit from the greenhouse to the icehouse conditions in the Paleogene (). The divergence between the south and north species in Dipelta and Dipteronia potentially resulted from parallel evolution, driven by geographical isolation and natural selection of the heterogeneous environments. Surviving the climate oscillations in the late Tertiary and Quaternary, they become the living fossils in the East Asian flora.
Statements
Data availability statement
The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/ Supplementary Material.
Author contributions
Z-LL conceptualized the study, wrote, reviewed, and edited the manuscript. FP and QY were responsible for the software. ZZ, BX, and JH were in charge of validation. FP and YL were responsible for the formal analysis. BT was in charge of the resources. FP prepared and wrote the original draft. All authors have read and agreed to the published version of the manuscript.
Funding
This work was financially supported by the National Natural Science Foundation of China (31670219 and 31870321).
Conflict of interest
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Supplementary material
The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fgene.2020.573226/full#supplementary-material
SUPPLEMENTARY TABLE A1List of species and their GenBank accession numbers in the study.
SUPPLEMENTARY TABLE A2Features of plastomes, mitogenomes and ITS fragments in Dipelta species.
SUPPLEMENTARY TABLE A3Selected genes in plastome and mitogenome with potential applications in species discrimination.
SUPPLEMENTARY TABLE A4Codon usage and relative synonymous codon usage (RSCU) value for protein-coding genes in Dipelta cytoplasmic genomes.
SUPPLEMENTARY TABLE A5Likelihood ratio test (LRT) of the variable ω ratio under different models.
References
1
BaiG. Q.ZhouT.ZhangX.ChenX. D.YangJ.LiZ. H.et al (2017). Genetic differentiation and population genetic structure of the Chinese endemic Dipteronia Oliv, revealed by cpDNA and AFLP data.Forests8:424. 10.3390/f8110424
2
BeaulieuJ. M.TankD. C.DonoghueM. J. (2013). A Southern Hemisphere origin for campanulid angiosperms, with traces of the break-up of Gondwana.BMC Evol. Biol.13:80. 10.1186/1471-2148-13-80
3
BellC. D.DonoghueM. J. (2005). Dating the dipsacales: comparing models, genes, and evolutionary implications.Am. J. Bot.92284–296. 10.3732/ajb.92.2.284
4
BensonG. (1999). Tandem repeats finder: a program to analyze DNA sequences.Nucl. Acids Res.27573–580. 10.1093/nar/27.2.573
5
BoweL. M.CoatG.dePamphilisC. W. (2000). Phylogeny of seed plants based on all three genomic compartments: extant gymnosperms are monophyletic and Gnetales’ closest relatives are conifers.Proc. Natl. Acad. Sci. U.S.A.974092–4097. 10.1073/pnas.97.8.4092
6
ChenY. (2014). Medicinal Flora of China.Beijing: Peking University Medical Press.
7
ChenY. S.DengT.ZhouZ.SunH. (2018). Is the east asian flora ancient or not?Natl. Sci. Rev.5920–932. 10.1093/nsr/nwx156
8
ChristensenA. C. (2013). Plant mitochondrial genome evolution can be explained by DNA repair mechanisms.Genome Biol. Evol.51079–1086. 10.1093/gbe/evt069
9
DarlingA. C.MauB.BlattnerF. R.PernaN. T. (2004). Mauve: multiple alignment of conserved genomic sequence with rearrangements.Genome Res.141394–1403. 10.1101/gr.2289704
10
DodsworthS. (2015). Genome skimming for next-generation biodiversity analysis.Trends Plant Sci.20525–527. 10.1016/j.tplants.2015.06.012
11
DongW. L.WangR. N.ZhangN. Y.FanW. B.FangM. F.LiZ. H. (2018). Molecular evolution of chloroplast genomes of Orchid species: insights into phylogenetic relationship and adaptive evolution.Int. J. Mol. Sci.19:716. 10.3390/ijms19030716
12
DonoghueM. J.BellC. D.WinkworthR. C. (2003). The evolution of reproductive characters in Dipsacales.Int. J. Plant Sci.164S453–S464. 10.1086/376874
13
DrescherA.RufS.CalsaT.CarrerH.BockR. (2000). The two largest chloroplast genome-encoded open reading frames of higher plants are essential genes.Plant J.2297–104. 10.1046/j.1365-313x.2000.00722.x
14
DrummondA. J.SuchardM. A.XieD.RambautA. (2012). Bayesian phylogenetics with BEAUti and the BEAST 1.7.Mol. Biol. Evol.291969–1973. 10.1093/molbev/mss075
15
FanW. B.WuY.YangJ.ShahzadK.LiZ. H. (2018). Comparative chloroplast genomics of Dipsacales species: insights into sequence variation, adaptive evolution, and phylogenetic relationships.Front. Plant Sci.9:689. 10.3389/fpls.2018.00689
16
FengY.ComesH. P.ZhouX. P.QiuY. X. (2019). Phylogenomics recovers monophyly and early Tertiary diversification of Dipteronia (Sapindaceae).Mol. Phylogenet. Evol.1309–17. 10.1016/j.ympev.2018.09.012
17
FonsecaL. H. M.LohmannL. G. (2020). Exploring the potential of nuclear and mitochondrial sequencing data generated through genome-skimming for plant phylogenetics: a case study from a clade of neotropical lianas.J. Syst. Evol.5818–32. 10.1111/jse.12533
18
FrazerK. A.PachterL.PoliakovA.RubinE. M.DubchakI. (2004). VISTA: computational tools for comparative genomics.Nucl Acids Res.32W273–W279. 10.1093/nar/gkh458
19
FreitasA.AnunciaçãoR.D’Oliveira MatieloC.StefenonV. (2018). Chloroplast DNA: a promising source of information for plant phylogeny and traceability.J. Mol. Biol. Methods1:1. 10.1007/978-1-4615-3276-7_1
20
GouldingS. E.OlmsteadR. G.MordenC. W.WolfeK. H. (1996). Ebb and flow of the chloroplast inverted repeat.Mol. Gen. Genet.252195–206. 10.1007/bf02173220
21
GuC. H.DongB.XuL.TembrockL. R.ZhengS. Y.WuZ. Q. (2018). The complete chloroplast genome of Heimia myrtifolia and comparative analysis within Myrtales.Molecules23:846. 10.3390/molecules23040846
22
HahnC.BachmannL.ChevreuxB. (2013). Reconstructing mitochondrial genomes directly from genomic next-generation sequencing reads-a baiting and iterative mapping approach.Nucl. Acids Res.41:e129. 10.1093/nar/gkt371
23
HarrisM. E.MeyerG.VandergonT.VandergonV. O. (2013). Loss of the Acetyl-CoA Carboxylase (accD) gene in Poales.Plant Mol. Biol. Rep.3121–31. 10.1007/s11105-012-0461-3
24
HoweC. J.BarbrookA. C.KoumandouV. L.NisbetR. E. R.SymingtonH. A.WightmanT. F. (2003). Evolution of the chloroplast genome.Philos. Trans. R. Soc. Lond. Ser. B Biol. Sci.35899–106. 10.1098/rstb.2002.1176
25
HuangJ. L.SunG. L.ZhangD. M. (2010). Molecular evolution and phylogeny of the angiosperm ycf2 gene.J. Syst. Evol.48240–248. 10.1111/j.1759-6831.2010.00080.x
26
JacobsB.LensF.SmetsE. (2009). Evolution of fruit and seed characters in the Diervilla and Lonicera clades (Caprifoliaceae, Dipsacales).Ann. Bot.104253–276. 10.1093/aob/mcp131
27
JacobsB.PyckN.SmetsE. (2010). Phylogeny of the Linnaea clade: are Abelia and Zabelia closely related?Mol. Phylogen. Evol.57741–752. 10.1016/j.ympev.2010.08.007
28
KatohK.StandleyD. M. (2013). MAFFT multiple sequence alignment software version 7: improvements in performance and usability.Mol. Biol. Evol.30772–780. 10.1093/molbev/mst010
29
KatzM. E.MillerK. G.WrightJ. D.WadeB. S.BrowningJ. V.CramerB. S.et al (2008). Stepwise transition from the Eocene greenhouse to the Oligocene icehouse.Nat. Geosci.1329–334. 10.1038/ngeo179
30
KhakhlovaO.BockR. (2006). Elimination of deleterious mutations in plastid genomes by gene conversion.Plant J.4685–94. 10.1111/j.1365-313X.2006.02673.x
31
KhanA. L.AsafS.LeeI. J.Al-HarrasiA.Al-RawahiA. (2018). First chloroplast genomics study of Phoenix dactylifera (var. Naghal and Khanezi): a comparative analysis.PLoS One13:e0200104. 10.1371/journal.pone.0200104
32
KikuchiS.BedardJ.HiranoM.HirabayashiY.OishiM.ImaiM.et al (2013). Uncovering the protein translocon at the chloroplast inner envelope membrane.Science339571–574. 10.1126/science.1229262
33
KimK. J.LeeH. L. (2004). Complete chloroplast genome sequences from Korean ginseng (Panax schinseng Nees) and comparative analysis of sequence evolution among 17 vascular plants.DNA Res.11247–261. 10.1093/dnares/11.4.247
34
KodeV.MuddE. A.IamthamS.DayA. (2005). The tobacco plastid accD gene is essential and is required for leaf development.Plant J.44237–244. 10.1111/j.1365-313X.2005.02533.x
35
KurtzS.ChoudhuriJ. V.OhlebuschE.SchleiermacherC.StoyeJ.GiegerichR. (2001). REPuter: the manifold applications of repeat analysis on a genomic scale.Nucl. Acids Res.294633–4642. 10.1093/nar/29.22.4633
36
LandreinS.BuerkiS.WangH. F.ClarksonJ. J. (2017). Untangling the reticulate history of species complexes and horticultural breeds in Abelia (Caprifoliaceae).Ann. Bot.120257–269. 10.1093/aob/mcw279
37
LandreinS.FarjonA. (2020). A monograph of Caprifoliaceae: Linnaeeae.Kew Bull.74:70. 10.1007/s12225-018-9762-5
38
LiD. Z.GaoL. M.LiH. T.WangH.GeX. J.LiuJ. Q.et al (2011). Comparative analysis of a large dataset indicates that internal transcribed spacer (ITS) should be incorporated into the core barcode for seed plants.Proc. Natl. Acad. Sci. U.S.A.10819641–19646. 10.1073/pnas.1104551108
39
LiuM. L.FanW. B.WangN.DongP. B.ZhangT. T.YueM.et al (2018). Evolutionary analysis of plastid genomes of seven Lonicera L. species: implications for sequence divergence and phylogenetic relationships.Int. J. Mol. Sci.19:4039. 10.3390/ijms19124039
40
LiuZ. L.TianP.LiJ. F.GuoC.BiffinE. (2013). A comparison of DNA sequences, SSR and AFLP for systematic study of Dipelta (Caprifoliaceae).Ann. Bot. Fenn.50351–359. 10.5735/086.050.0510
41
ManchesterS. R.ChenZ. D.LuA. M.UemuraK. (2009). Eastern Asian endemic seed plant genera and their paleogeographic history throughout the Northern Hemisphere.J. Syst. Evol.471–42. 10.1111/j.1759-6831.2009.00001.x
42
MandelJ. R.McAsseyE. V.RolandK. M.McCauleyD. E. (2012). Mitochondrial gene diversity associated with the atp9 stop codon in natural populations of wild carrot (Daucus carota ssp. carota).J. Hered.103418–425. 10.1093/jhered/esr142
43
MengX. X.XianY. F.XiangL.ZhangD.ShiY. H.WuM. L.et al (2018). Complete chloroplast genomes from Sanguisorba: identity and variation among four species.Molecules23:2137. 10.3390/molecules23092137
44
NevillP. G.ZhongX.Tonti-FilippiniJ.ByrneM.HislopM.ThieleK.et al (2020). Large scale genome skimming from herbarium material for accurate plant identification and phylogenomics.Plant Methods16:1.
45
PatelR. K.JainM. (2012). NGS QC Toolkit: a toolkit for quality control of next generation sequencing data.PLoS One7:e30619. 10.1371/journal.pone.0030619
46
PedenJ. F. (1999). Analysis of Codon Usage.PhD. Nottingham: University of Nottingham.
47
PengF. F.ChenL.YangQ.TianB.LiuZ. L. (2019). The complete chloroplast genome of Dipelta yunnanensis (Caprifoliaceae), a vulnerable plant in China.Mitochondr. DNA B Resour.4515–516. 10.1080/23802359.2018.1551084
48
PosadaD.CrandallK. A. (1998). MODELTEST: testing the model of DNA substitution.Bioinformatics14817–818. 10.1093/bioinformatics/14.9.817
49
Rousseau-GueutinM.HuangX.HigginsonE.AyliffeM.DayA.TimmisJ. N. (2013). Potential functional replacement of the plastidic Acetyl-CoA Carboxylase subunit (accD) gene by recent transfers to the nucleus in some angiosperm lineages.Plant Physiol.1611918–1929. 10.1104/pp.113.214528
50
StamatakisA. (2006). RAxML-VI-HPC: maximum likelihood-based phylogenetic analyses with thousands of taxa and mixed models.Bioinformatics222688–2690. 10.1093/bioinformatics/btl446
51
StraubS. C. K.ParksM.WeitemierK.FishbeinM.CronnR. C.ListonA. (2012). Navigating the tip of the genomic iceberg: next-generation sequencing for plant systematics.Am. J. Bot.99349–364. 10.3732/ajb.1100335
52
ThielT.MichalekW.VarshneyR. K.GranerA. (2003). Exploiting EST databases for the development and characterization of gene-derived SSR-markers in barley (Hordeum vulgare L.).Theor. Appl. Genet.106411–422. 10.1007/s00122-002-1031-0
53
TianB.FuY.MilneR. I.MaoK. S.SunY. S.MaX. G.et al (2019). A complex pattern of post-divergence expansion, contraction, introgression, and asynchronous responses to Pleistocene climate changes in two Dipelta sister species from western China.J. Syst. Evol.58247–262. 10.1111/jse.12524
54
TrevisanB.AlcantaraD. M. C.MachadoD. J.MarquesF. P. L.LahrD. J. G. (2019). Genome skimming is a low-cost and robust strategy to assemble complete mitochondrial genomes from ethanol preserved specimens in biodiversity studies.PeerJ7:e7543. 10.7717/peerj.7543
55
WangH. F.LandreinS.DongW. P.NieZ. L.KondoK.FunamotoT.et al (2015). Molecular phylogeny and biogeographic diversification of Linnaeoideae (Caprifoliaceae s. l.) disjunctly distributed in Eurasia. North America and Mexico.PLoS One10:e0116485. 10.1371/journal.pone.0116485
56
WangH. X.LiuH.MooreM. J.LandreinS.LiuB.ZhuZ. X.et al (2020). Plastid phylogenomic insights into the evolution of the Caprifoliaceae s.l.(Dipsacales). Mol. Phylogenet. Evol.142:106641. 10.1016/j.ympev.2019.106641
57
WangS.XieY. (2004). China Species Red List.Beijing: Higher Education Press.
58
WangS. B.SongQ. W.LiS. S.HuZ. G.DongG. Q.SongC.et al (2018). Assembly of a complete mitogenome of Chrysanthemum nankingenseusing Oxford Nanopore long reads and the diversity and evolution of Asteraceae mitogenomes.Genes9:547. 10.3390/genes9110547
59
WangW. Q.MessingJ. (2011). High-throughput sequencing of three Lemnoideae (Duckweeds) chloroplast genomes from total DNA.PLoS One6:e0024670. 10.1371/journal.pone.0024670
60
WickeS.SchneeweissG. M.dePamphilisC. W.MullerK. F.QuandtD. (2011). The evolution of the plastid chromosome in land plants: gene content, gene order, gene function.Plant Mol. Biol.76273–297. 10.1007/s11103-011-9762-4
61
WinkworthR. C.BellC. D.DonoghueM. J. (2008). Mitochondrial sequence data and Dipsacales phylogeny: mixed models, partitioned Bayesian analyses, and model selection.Mol. Phylogenet. Evol.46830–843. 10.1016/j.ympev.2007.11.021
62
WuZ.RavenP.HongD. (eds) (2011). Flora of China.St. Louis, MO: Science Press.
63
WymanS. K.JansenR. K.BooreJ. L. (2004). Automatic annotation of organellar genomes with DOGMA.Bioinformatics203252–3255. 10.1093/bioinformatics/bth352
64
XiangC. L.DongH. J.LandreinS.ZhaoF.YuW. B.SoltisD. E.et al (2020). Revisiting the phylogeny of Dipsacales: new insights from phylogenomic analyses of complete plastomic sequences.J. Syst. Evol.58103–117. 10.1111/jse.12526
65
XuT. T.AbbottR. J.MilneR. I.MaoK.DuF. K.WuG. L.et al (2010). Phylogeography and allopatric divergence of cypress species (Cupressus L.) in the Qinghai-Tibetan Plateau and adjacent regions.BMC Evol. Biol.10:194. 10.1186/1471-2148-10-194
66
XuW. J.XingT.ZhaoM. M.YinX. H.XiaG. M.WangM. C. (2015). Synonymous codon usage bias in plant mitochondrial genes is associated with intron number and mirrors species evolution.PLoS One10:e0131508. 10.1371/journal.pone.0131508
67
YangZ. H.WongW. S. W.NielsenR. (2005). Bayes empirical Bayes inference of amino acid sites under positive selection.Mol. Biol. Evol.221107–1118. 10.1093/molbev/msi097
68
YiX.GaoL.WangB.SuY. J.WangT. (2013). The complete chloroplast genome sequence of Cephalotaxus oliveri (Cephalotaxaceae): evolutionary comparison of Cephalotaxus chloroplast DNAs and insights into the loss of inverted repeat copies in gymnosperms.Genome Biol. Evol.5688–698. 10.1093/gbe/evt042
69
ZhangW. J.ZhouJ.LiZ. F.WangL.GuX.ZhongY. (2007). Comparative analysis of codon usage patterns among mitochondrion, chloroplast and nuclear genes in Triticum aestivum L.J. Integr. Plant Biol.49246–254. 10.1111/j.1744-7909.2007.00404.x
70
ZhaoN.WangY.HuaJ. (2018). The roles of mitochondrion in intergenomic gene transfer in plants: a source and a pool.Int. J. Mol. Sci.19:547. 10.3390/ijms19020547
71
ZhouT.RuhsamM.WangJ.ZhuH.LiW.ZhangX.et al (2019). The complete chloroplast genome of Euphrasia regelii, pseudogenization of ndh genes and the phylogenetic relationships within Orobanchaceae.Front. Genet.10:444. 10.3389/fgene.2019.00444
Summary
Keywords
Dipelta, plastomes, mitogenome, intergenic transcribed spacer, divergence, phylogeny
Citation
Peng F, Zhao Z, Xu B, Han J, Yang Q, Lei Y, Tian B and Liu Z-L (2020) Characteristics of Organellar Genomes and Nuclear Internal Transcribed Spacers in the Tertiary Relict Genus Dipelta and Their Phylogenomic Implications. Front. Genet. 11:573226. doi: 10.3389/fgene.2020.573226
Received
16 June 2020
Accepted
26 August 2020
Published
25 September 2020
Volume
11 - 2020
Edited by
Lifeng Zhu, Nanjing Normal University, China
Reviewed by
Tao Zhou, Xi’an Jiaotong University, China; Iker Irisarri, University of Göttingen, Germany
Updates
Copyright
© 2020 Peng, Zhao, Xu, Han, Yang, Lei, Tian and Liu.
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Zhan-Lin Liu, liuzl@nwu.edu.cn
This article was submitted to Evolutionary and Population Genetics, a section of the journal Frontiers in Genetics
Disclaimer
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.