Identification of TaPPH-7A haplotypes and development of a molecular marker associated with important agronomic traits in common wheat
- 191 Downloads
Premature senescence of flag leaf severely affects wheat yield and quality. Chlorophyll (Chl) degradation is the most obvious symptom during leaf senescence and catalyzed by a series of enzymes. Pheophytin pheophorbide hydrolase (Pheophytinase, PPH) gene encodes a Chl degradation hydrolase.
In this study, the coding, genomic and promoter sequences of wheat TaPPH-A gene were cloned. The corresponding lengths were 1467 bp, 4479 bp and 3666 bp, respectively. Sequence structure analysis showed that TaPPH-A contained five exons and four introns. After the multiple sequences alignment of TaPPH-A genome from 36 accessions in a wheat diversity panel, four SNPs and one 2-bp InDel were observed, which formed two haplotypes, TaPPH-7A-1 and TaPPH-7A-2. Based on the SNP at 1299 bp (A/G), a molecular marker TaPPH-7A-dCAPS was developed to distinguish allelic variation (A/G). Using the molecular markers, 13 SSR, and 116 SNP markers, a linkage map of chromosome 7A were integrated. TaPPH-A was mapped on the chromosome region flanked by Xwmc9 (0.94 cM) and AX-95634545 (1.04 cM) on 7A in a DH population. Association analysis between TaPPH-7A allelic variation and agronomic traits found that TaPPH-7A was associated with TGW in 11 of 12 environments and Chl content at grain-filling stage under drought stress using Population 1 consisted of 323 accessions. The accessions possessed TaPPH-7A-1 (A) had higher TGW and Chl content than those possessed TaPPH-7A-2 (G), thus TaPPH-7A-1 (A) was a favorable allelic variation. By analyzing the frequency of favorable allelic variation TaPPH-7A-1 (A) in Population 2 with 157 landraces and Population 3 with 348 modern cultivars, we found it increased from pre-1950 (0) to 1960s (54.5%), then maintained a relatively stable level about 56% from 1960s to 1990s.
These results suggested the favorable allelic variation TaPPH-7A-1 (A) should be valuable in enhancing grain yield by improving the source (chlorophyll content) and sink (the developing grain) simultaneously. Furthermore, the newly developed molecular marker TaPPH-7A-dCAPS could be integrated into a breeding kit of screening high TGW wheat for marker-assisted selection.
KeywordsGene cloning Allelic variation Molecular marker TGW Chl content Wheat
Derived cleaved amplified polymorphic sequence
Grain number per spike
Pheophytin pheophorbide hydrolase
Quantitative trait loci
Spike number per plant
Single nucleotide polymorphism
Wheat (Triticum aestivum L.) provides the staple food source for 30% human population in the world . It is estimated that the global demand for wheat will increase by a further 40% before 2020, because of an increasing world population . Therefore, breeding for high-yield varieties has still been a major objective in wheat breeding programs . Thousand-grain weight (TGW) is an important yield-contributing trait. The significantly genetic improvement in wheat grain yield is partially attributed to increased TGW .
Delayed senescence, or stay-green, enables leaf to maintain a longer greenness after anthesis, and contributes to a longer grain-filling period [5, 6]. Gregersen, et al.  reported a positive correlation between delayed crop senescence and grain yield. A two-days delay in onset of senescence increased 11% carbon fixed in Lolium temulentum L . Furthermore, the yield of wheat stay-green mutant tasg1 was 9.5% higher than the wild type . Thus it has been regarded as a desirable characteristic for the production of a number of crops including wheat. Chlorophyll (Chl) degradation is the main indicator of leaf senescence and catalyzed by a series of enzymes [10, 11, 12, 13, 14]. Pheophytin pheophorbide hydrolase (Pheophytinase, PPH) is a key enzyme in Chl degradation, specifically hydrolyzes pheophytin a to pheophorbide a [15, 16]. Mutagenesis or overexpression of PPH can lead to a stay-green or premature senescence phenotype in Arabidopsis and rice [15, 16]. In addition, overexpressing LpPPH also accelerated Chl degradation, and the expression level was positively related to leaf senescence . Therefore, the expression of PPH gene affected Chl degradation, further affected yield and quality of crops.
Marker-assisted selection is considered a potential approach to accelerate the process of wheat breeding [18, 19]. Single nucleotide polymorphism (SNP) mainly refers to DNA sequence polymorphisms caused by single nucleotide changes and small insertions/deletions [20, 21]. Comparing with other marker types, such as RFLP, RAPD, AFLP, SSR and ISSR, SNP is a kind of more valuable marker due to its high abundance and relative stability, cost efficiency, and high-throughput scoring [22, 23]. Additionally, SNP has been also proved to be a kind of very effective molecular markers in high-density genetic map construction, fine mapping, genetic diversity, association analysis, functional marker development and marker-assisted selection breeding [24, 25]. Along with the development of high density genetic map of SNPs, association analysis has been proved to be an effective tool for identifying the relationship between polymorphic sites of target genes and important quantitative traits, and has been widely performed in many plant species, such as Arabidopsis [26, 27], rice [28, 29], maize [30, 31, 32], and wheat [33, 34].
Cloning and characterization of TaPPH-A gene
Sequence polymorphism and molecular marker development for TaPPH-A
To detect polymorphisms in the genomic and promoter sequences of TaPPH-A, all accessions in wheat diversity panel were used to amplify the TaPPH-A gene, and the amplified fragments were sequenced. Multiple sequence alignment found that four SNPs (G/A, A/G, C/T, A/C) and a 2-bp InDel were observed in the genomic sequence, while no sequence variation was detected in the promoter sequence of TaPPH-A (Fig. 2a and b). From the start codon (ATG), four SNPs were located at 1038 bp, 1299 bp, 1468 bp, and 3790 bp, and the 2-bp InDel was identified from 1902 to 1903 bp. The four SNPs and the 2-bp InDel together formed two haplotypes, TaPPH-7A-1 and TaPPH-7A-2 (Fig. 2b).
Based on the SNP at 1299 bp (A/G), a molecular marker was developed and named TaPPH-7A-dCAPS. It contained a base G mismatched to C in the downstream primer EcoRI-R and a restriction enzyme EcoRI site (Fig. 2c). To characterize the observed TaPPH-A allelic variation in large wheat populations, the following three experiments were performed: firstly, the genome-specific primer pair, G-F/R, was used to amplify genomic sequence of TaPPH-A, and a 4125 bp fragment was obtained. Secondly, 0.2 μL PCR product of 4125 bp was used as the template to amplify a 202 bp fragment by the molecular marker primer pair, EcoRI-F/R. Thirdly, the PCR product of 202 bp possessed allelic variation TaPPH-7A-1 (A) formed a EcoRI site GAATTC (Fig. 2c), and was digested into 174 bp and 28 bp by restriction enzyme EcoRI (Fig. 2d). Whereas the corresponding PCR product possessed allelic variation TaPPH-7A-2 (G) was GGATTC (Fig. 2c), and corresponding product remained 202 bp (Fig. 2d).
Genetic mapping of TaPPH
Association of TaPPH-7A with yield-related traits
Association analysis of TaPPH-7A allelic variation and yield-related traits in 12 environments
Association of TaPPH-7A with Chl content
Association analysis of TaPPH-7A allelic variation and Chl contents in two environments
Quantitative real-time PCR analysis of TaPPH-7A expression
Geographic distribution of TaPPH-7A allelic variation in Chinese wheat production zones
The frequency of TaPPH-7A allelic variation in wheat breeding history
Sequence polymorphism of TaPPH-7A
During the evolution, domestication and breeding in wheat, the genetic information had undergone two doublings and mutations under natural conditions, which eventually resulted in a rich genetic diversity. Somers, et al.  reported the level of sequence polymorphism was 1 SNP every 540 bp of EST sequence using a bioinformatics strategy based on a large wheat EST database from 12 cultivars. Ravel, et al.  identified that the SNP frequency was 1 SNP/334 bp in the genomic sequence including coding and non-coding region, and 1 SNP/267 bp in coding region in wheat. In this study, polymorphisms of TaPPH-7A gene sequences from the wheat diversity panel (36 accessions) was detected by direct sequencing. The SNPs occurred in both exons and introns region of TaPPH-7A (Fig. 2). This study allowed us to estimate an average of 1 SNP for every 1120 bp in the genomic sequence including coding and non-coding region, and 1 SNP every 1467 bp in coding region, 1 SNP every 1004 bp in non-coding region, but no polymorphism detected in the promoter region. As expected, the coding region had lower SNP frequency than the non-coding region, and this result is consistent with earlier studies . However, the SNP frequency of the whole genomic sequence was much lower than those in the previous studies [36, 37]. This result suggested that TaPPH-7A is a relatively conservative gene in evolution.
In this study, only one synonymous mutation SNP (C/T) was identified in the exon region at 1468 bp of TaPPH-7A. However, the TaPPH-7A associated with TGW and Chl content, and the relative expression levels of TaPPH-7A were lower in wheat genotypes with TaPPH-7A-1 (A) than those with TaPPH-7A-2 (G) (Fig. 5), so the reason for the phenotypic and gene expression variation may be alternative splicing. Many previous studies had showed that variations in the introns can significantly affect gene expression or phenotypic variation by alternative splicing of pre-mRNA [39, 40, 41, 42]. Thus, variations in introns are important and worthy of future investigation.
TaPPH-7A is a novel locus related to TGW and Chl content
Using QTL mapping, it is possible to dissect loci controlling genetic variation and characterize these loci based on map position, phenotypic effects, gene actions, pleiotropic effects, and epistatic interactions with other QTL in a segregating population . Several QTL for Chl content and TGW were detected on wheat chromosome 7A in various mapping populations. Using the DH population derived from the cross of Hanxuan 10 × Lumai 14, Shi, et al.  found two QTL for Chl content at flowering and grain-filling stage on chromosome 7A with 7.32 and 8.36% of phenotypic variation, respectively, and Yang, et al.  detected a QTL for Chl content in the interval Xwmc488-P2071–180 on chromosome 7A under water deficient. Bhusal, et al.  also reported a QTL for Chl a content on chromosome 7A close to the marker Xwmc388 under heat stress. Ilyas, et al.  mapped a QTL for total Chl content on 7A flanked by Xbarc49 and Xgdm14. As for TGW, a few QTL have been reported on chromosome 7AL in wheat, too. For example, Wang, et al.  reported a QTL for TGW close to the marker Xcfa2257 in a natural population, which explained up to 21.99% of the total phenotypic variation. Groos, et al.  also found a QTL for TGW in the vicinity of Xgwm282, explaining phenotypic variation ranging from 5.2 to 10.3% in six environments. In the present study, TaPPH was located on chromosome 7A between markers Xwmc9 and AX-95634545 within an interval of 1.98 cM by QTL mapping (Fig. 3b). Based on the common wheat consensus maps [50, 51], TaPPH-7A is considered to be near to the centromere, but is far from markers Xwmc488 and P2071–180 , Xwmc388 , Xbarc49 and Xgdm14 , Xcfa2257 , and Xgwm282 . Thus, TaPPH-7A was likely a novel locus related to Chl content and TGW on chromosome 7A.
TaPPH-7A-dCAPS is a stably and effectively molecular marker for assisted selection breeding
Functional molecular marker was developed from polymorphic site within gene causally affecting phenotypic trait variation . Converting SNPs to CAPS or dCAPS markers enable SNPs to be more conveniently applied in selecting preferred alleles in marker-assisted breeding . In addition, the assay procedure of CAPS or dCAPS marker was simple. However, it is difficult to develop CAPS or dCAPS markers from identified wheat genes because of the large allohexaploid genome [53, 54]. In order to solve this problem, we firstly designed genome-specific primer to distinguish three ortholog genome sequences, then identified SNPs by comparing the genomic sequences of target gene from different wheat genotypes, finally developed molecular markers based on SNP. Using the method described above, a molecular marker TaPPH-7A-dCAPS was developed based on the SNP at 1299 bp in this study (Fig. 2c). Two allelic variation, TaPPH-7A-1 (A) and TaPPH-7A-2 (G), were detected by scanning the Population 1 using the developed molecular marker (Additional file 5: Table S3). We made further efforts to found that favorable allelic variation TaPPH-7A-1 (A) was associated with high TGW in 11 of 12 environments and Chl content at grain-filling stage under drought stress using Population 1 (Fig. 4). And we also found that genotypes with favorable allelic variation TaPPH-7A-1 (A) had lower relative expression level of TaPPH-7A than those with TaPPH-7A-2 (G) (Fig. 5). Therefore, it is deduced that the favorable allelic variation, TaPPH-7A-1 (A), could simultaneously improve the source (chlorophyll content) and sink (the developing grain) after flowering, and eventually contribute to wheat grain weight. Thus, the newly developed molecular marker, TaPPH-7A-dCAPS, is a stable and effectively molecular marker for grain yield, and can be used to marker-assisted selection breeding in wheat.
Favorable allelic variation TaPPH-7A-1 (a) was selected in wheat breeding history
Wheat is one of the most important staple food crops in China. With the development of economy, the breeding objective is constantly changing. Before 1960s, the main breeding objective was to increase TGW and further improve yield. From 1970s to 1990s, the breeding objective was changed to improve agronomic traits such as plant height, quality, spike number and grain weight per spike . The change of Chinese wheat breeding objective from 1950 to 1960s resulted in the positive selection of favorable allelic variation TaPPH-7A-1 (A) and the rapid increase in TGW before 1970 (Fig. 7). From 1970s to 1990s, the frequency of favorable allelic variation TaPPH-7A-1 (A) maintained a relatively stable level about 56%, but TGW continuously increased (Fig. 7), the possible reason may be that other genes contributing to TGW were selected, such as TaSnRK2.3-1A (Hap-1A-1) and TaSnRK2.3-1B (Hap-1B-1) , TaSPL21-6D-HapII&III , and TaSAP7-B (C) . Moreover, the frequency of favorable allelic variation TaPPH-7A-1 (A) was increased from landraces (Population 2) to modern cultivars (Population 3) in Chinese ten wheat Zones. However, in Population 3, the maximum frequency of TaPPH-7A-1 (A) was only 63.6% in Zone VI (Additional file 4: Table S2, Fig. 6a and b). Thus, there is a large potential for increasing TGW by selecting favourable allelic variation TaPPH-7A-1 (A) in high-yield wheat breeding.
The TaPPH-7A gene was cloned. Four SNPs and one 2-bp InDel were observed. A molecular marker TaPPH-7A-dCAPS was developed based on a SNP at 1299 bp (A/G). Favourable allelic variation TaPPH-7A-1 (A) was found to be associated with high TGW and Chl content, but not with SN and GN in Population 1. The frequency of favourable allelic variation TaPPH-7A-1 (A) was maintained a relatively stable level of about 56% from 1960s to 1990s in Population 3, thus the favourable allelic variation should be valuable, and could be selected to increase grain yield by improving the source (chlorophyll content) and sink (the developing grain) simultaneously. The newly developed molecular marker TaPPH-7A-dCAPS could be integrated into a breeding kit of screening high TGW wheat for marker-assisted selection.
Chang 4738, a wheat variety with a high TGW and a slowly chlorophyll degradation rate after anthesis, was used to clone TaPPH-7A gene. A wheat diversity panel (36 accessions, Additional file 6: Table S4) was chosen to detect polymorphisms in TaPPH-7A genomic sequence. Fifteen wheat species of different ploidy and a set of nulli-tetrasomic lines of Chinese Spring were used for chromosome location of TaPPH. A DH population , derived from a cross of Hanxuan 10 × Lumai 14, consisting of 150 lines, was used for linkage mapping of TaPPH.
Three wheat germplasm populations were used as plant materials to analyze the allelic variation in TaPPH-7A. Population 1 consisted of 323 accessions (Additional file 3: Table S1), including 275 modern cultivars, 36 advanced lines, and 12 landraces, was used for association analysis of the target gene allelic variation and phenotypic traits. Population 2 (157 landraces) and Population 3 (348 modern cultivars)  (provided by Dr. Xueyong Zhang, at Chinese Academy of Agricultural Sciences) were used to analyze the frequency of favorable allelic variation in different wheat production zones and wheat breeding history, and to evaluate the utilization value of the newly developed molecular markers.
All wheat accessions were legally obtained from Chinese Crop Germplasm Resources Information System (http://www.cgris.net/zhongzhidinggou/index.php).
Field management and phenotypic assessment
Population 1 was planted at Experiment Stations at Changping (116°13′E; 40°13′N) and Shunyi (116°56′E; 40°23′N) in Beijing over 3 growing seasons, i.e. in 2015 at Shunyi, 2016 at Changping and Shunyi, and 2017 at Shunyi. The experiment field was divided into two plots with different water regimes: rain-fed (drought stressed, DS) and well-watered (WW). The DS plots were not irrigated during the whole growing season but had rainfalls of 173 mm, 143 mm, and 116 mm, respectively. The WW plots were irrigated with 750 m3/ha (75 mm): before winter, at booting, flowering and grain filling when the amounts of rainfall were insufficient during each corresponding period. In addition, a heat stress experiment (HS) was conducted by adding polythene covers over the plots at Shunyi in 2015 and 2016. The other field managements, such as fertilization, disease and pest control were the same as local production conditions. Phenotypic assessment was performed under 12 environments (E1 to E12). E1 to E12 indicated the environments at Shunyi in 2015 under WW, DS, WW + HS and DS + HS, Changping in 2016 under WW and DS, Shunyi in 2016 under WW, DS, WW + HS and DS + HS, Shunyi in 2017 under WW and DS, respectively.
Yield-related traits, including spike number per plant (SN), grain number per spike (GN) and TGW, were measured under 12 environments. Chl content (SPAD value) was tested with a handheld portable chlorophyll meter (SPAD-502, Konica-Minolta, Tokyo) at flowering and grain filling stages in two environments (E11 and E12) (Additional file 7: Table S4). The SN, GN, TGW and Chl content were measured using five plants, separately.
Population 3 was grown at Luoyang (112°45′E; 36°61′N) in Henan Province during the 2001–2002 and 2004–2005, and Shunyi (116°56′E; 40°23′N) in Beijing during 2009–2010 cropping seasons . TGW was measured in all the three environments.
DNA and RNA extraction and TaPPH-A gene cloning
Seedlings of Chang 4738 were used as experimental materials. Genomic DNA was extracted by the CTAB method . Total RNA was extracted using an RNAprep Pure Plant Kit (Tiangen, Beijing) following the manufacturer’s instructions. First-strand cDNA was synthesized with a FastQuant RT Kit (with gDNase) (Tiangen, Beijing).
Primer sequences used in this study
Primer sequence (5′-3′)
Cloning coding sequence
Cloning TaPPH-A genomic specific sequence
Cloning TaPPH-A promoter specific sequence
Marker TaPPH-7A-dCAPS developed for SNP-1299 (A/G)
The cDNA and genomic DNA of Chang 4738 were used as templates. PCR amplification was performed in 20 μL volume including 1 μL 50 ng/μL cDNA or genomic DNA, 0.2 μM of each primer, 0.2 mM dNTPs, 4 μL 5 × TransStart® FastPfu Buffer, and 0.4 μL (2.5 U/μL) TransStart® FastPfu DNA Polymerase (TransGen Biotech, Beijing). The amplification program consisted of an initial denaturation at 95 °C for 5 min, followed by 35 cycles of denaturation at 95 °C for 30 s, annealing at 58 °C for 45 s, and extension at 72 °C for 1–2 min, with a final extension at 72 °C for 15 min. The PCR products were separated by electrophoresis in 1.2% agarose gel. The target bands were purified using Gel Extraction Kit (BIOMIGA, China). The purified PCR product was cloned into pEASY®-Blunt cloning vector (TransGen Biotech, Beijing), then was transformed to 33 μL Trans1-T1 Phage Resistant Chemically Competent Cells (TransGen Biotech, Beijing) by heat shock. Six positive clones of each wheat genotype were sequenced. The sequence alignment was performed using SeqMan (DNASTAR Lasergene 7.1.0). The gene structure of TaPPH-A was determined using MegAlign (DNASTAR Lasergene 7.1.0) through alignment the coding and genomic sequences.
Molecular marker development
The sequences of TaPPH-7A cloned from the wheat diversity panel were aligned by SeqMan for screening polymorphism. Molecular marker(s) were developed based on polymorphism sites. A dCAPS marker was designed using dCAPS Finder 2.0 (http://helix.wustl.edu/dcaps/dcaps.html). The primer pair of dCAPS marker was EcoRI-F/R. Genotyping was performed by two rounds of PCR and one enzyme digestion following the procedure described by Miao, et al. .
Population structure and association analysis
Population 1 was performed on Wheat 660 K SNP Array, which consisted of 630,517 SNPs . By removing nucleotide variations with missing rates ≥0.2 and minor allele frequency (< 0.05), 395,681 SNPs were eventually used to detect the structure of Population 1 by software STRUCTURE 2.3.4 . A general linear model was performed in TASSEL 2.1 for analyzing significant associations between the target gene allelic variation and and phenotypic traits. Associations were considered significant at P < 0.05. Statistical analysis was conducted by SPSS 19.0 software (SPSS Corp., Chicago, IL, USA).
Analysis of gene expression with quantitative real-time PCR
A primer pair RT-F/R (Table 3) was designed to use for analyzing the expression of TaPPH-7A in different wheat genotypes. Quantitative real-time PCR (qRT-PCR) was performed in triplicate with Roche LightCycler® 96 using the SYBR® Premix Ex Taq™ (Tli RNaseH Plus) (Takara, Japan). The qRT-PCR reaction system with specific primer contains 10 μL 2 × SYBR Premix Ex Taq™, 0.4 μL 50 × Rox Reference Dye II, 0.4 μL (5 μM) of each primer (Table 3), 1 μL cDNA template, and 7.8 μL ddH2O. The reaction procedure was as follows: denaturation at 95 °C for 2 min; followed by 45 cycles at 95 °C for 20 s, 60 °C for 20 s, and 72 °C for 20 s. TaActin was used as the endogenous control of normalizing expression levels of different samples. Gene relative expression levels were calculated using the 2-△△CT method . The statistical analysis of ΔΔCT according to the method described by Zhang et al. .
We thank our colleagues Dr. Xueyong Zhang (Institute of Crop Science, Chinese Academy of Agricultural Sciences) for providing DNA samples of Population 2 (157 landraces) and Population 3 (348 modern cultivars).
DS and RJ designed the research and revised the manuscript. HW performed the experiments, analyzed the data, and drafted the manuscript. SW, XC, and CH helped to measure agronomic traits. All the authors read and approved the final manuscript.
This work was supported by the National Natural Science Foundation of China (31671607), the National Key R&D Program of China (2017YFD0300202), and National Science and Technology Major Projects for Cultivation of New Transgenic Varieties (2018ZX0800917B). The funding bodies provided only the experimental cost, including data collection, and publication fee for this study. However, the experimental design, data analysis and interpretation, and writing the manuscript were managed by the contributing authors.
Ethics approval and consent to participate
Consent for publication
The authors declare that they have no competing interests.
- 30.Guillet-Claude C, Birolleau-Touchard C, Manicacci D, Rogowsky PM, Rigau J, Murigneux A, et al. Nucleotide diversity of the ZmPox3 maize peroxidase gene: relationships between a MITE insertion in exon 2 and variation in forage maize digestibility. BMC Genet. 2004;5:19–29.PubMedPubMedCentralCrossRefGoogle Scholar
- 47.Ilyas M, Ilyas N, Arshad M, Kazi AG, Kazi AM, Waheed A. QTL mapping of wheat doubled haploids for chlorophyll content and chlorophyll fluorescence kinetics under drought stress imposed at anthesis stage. Pak J Bot. 2014;46:1889–97.Google Scholar
- 55.He ZH, Zhuang QS, Cheng SH, Yu ZW, Zhao ZD, Liu X. Wheat production and technology improvenment in China. J Agric For. 2018;8:107–14.Google Scholar
- 56.Miao LL, Mao XG, Wang JY, Liu ZC, Zhang B, Li WY, et al. Elite haplotypes of a protein kinase gene TaSnRK2.3 associated with important agronomic traits in common wheat. Front Plant Sci. 2017;08:368–78.Google Scholar
- 59.Li L, Mao X, Wang J, Chang X, Reynolds M, Jing R. Genetic dissection of drought and heat-responsive agronomic traits in wheat. Plant Cell Environ. 2019. https://doi.org/10.1111/pce.13577.
- 62.Li L, Peng Z, Mao X, Wang J, Chang X, Reynolds M, et al. Genome-wide association study reveals genomic regions controlling root and shoot traits at late growth stages in wheat. Ann Bot. 2019. https://doi.org/10.1093/aob/mcz041.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.