Introduction

Chloroplast is a cell organelle that provides energy for plants and algae via the process of photosynthesis. Other biological processes occur in chloroplast including the production of starch, lipids, amino acids, vitamins, and key pathways of sulfur and nitrogen metabolism.¹ During evolution, chloroplasts were thought to arise from endosymbiosis between photosynthetic bacterium and non-photosynthetic host.² Plant plastid (cp) contains highly conserved genomes in terms of structure and gene content compared to those of mitochondrial and nuclear genomes.^3,4 Individual chloroplast contains up to 1,600 copies of cp genome or plastome.⁵ In angiosperm, ex., monocots, cp DNA is circular and genome size ranges between 120-160 kb and featured with a quadripartite organization of two copies of inverted repeats (IRs) (20-28 kb), and a large (80-90 kb) and a small (16-27 kb) single-copy region, namely LSC and SSC, respectively. The cp genome mostly harbors ~4 rRNAs, ~30 tRNAs and ~80 protein-coding genes in addition to introns and intergenic spacers (IGS).⁶ Chloroplast genome is maternally inherited and studies of its structure, sequence variation, and diversity are useful in cytoplasmic breeding and non-inherited transgene insertions.⁵ Differences in gene content have been detected among angiosperm cp genomes,^7-9 however, no records were made at the plant species level.

In the past, the advent of Sanger sequencing method has enabled the elucidation of genetic information, however, it was hampered by technical details, costs, time and data resolution. The next-generation sequencing (NGS) technology has overcome these problems and revolutionized the science of genomics more appropriately. NGS revealed unlimited insights into genomes and transcriptomes of many species during the last few years.

Wheat is among the most widely cultivated field crops worldwide. Cultivated wheats can be either hexaploid (T. aestivum, AABBDD, 2n=6x) or tetraploid (Triticum durum, AABB, 2n=4x). Complexity of wheat nuclear genome in terms of genome types and size makes it difficult to be sequenced and assembled. The draft genome of the A-genome progenitor (e.g., T. urartu, AA) has been assembled and assigned as a reference genome for further comparison with polyploid genomes.¹⁰

A number of studies used the whole genome approach in order to detect SNPs and INDELs in the mitochondrial (mt) and cp genomes.^5,11-13 Nonetheless, utilization of SNP/INDELs of plastome in detecting genetic distances is a challenging task. With the possibility that half of the cp genome has analogue sequences in mitochondrial genome and due to the incidence of intra-varietal heteroplasmy, drawing dendrograms to describe the relationships among cultivars based on organellar SNPs/INDELs is a challenging task. Although heteroplasmy has been reported as a rare event in cp genomes,¹⁴ earlier studies indicated higher probabilities.^15,16 We speculate that polymorphism due to partial genome transfer and heteroplasmy should be removed before we approach to detect SNPs/INDELs among genotypes.

The available reference cp genome of the hexaploid “Chinese Spring” cultivar was previously sequenced based on the constructed genomic library and the assembled clone-contigs.³ In the present study, we have detected the structure and gene content of wheat plastome based on the new era of NGS with nine wheat cultivars. Eight of these cultivars are hexaploids and one is a tetraploid. We also attempted to detect genetic distance within hexaploid species or between the two wheat species based on SNPs/INDELs of cp genomes. 

Methods

Sampling and DNA Isolation

Nucleic acids were isolated from leaf tissues (~ 1 g) of 14-day-old etiolated seedlings of nine wheat cultivars (Table 1) using the modified procedure of Gawel and Jarret¹⁷. DNAs were treated with RNase A (10 mg/ml) and incubated at 37^oC for 30 min to remove RNA contaminants. Then, DNAs were shipped in liquid nitrogen to BGI, China for deep sequencing using the Illumina HiSeq 2000 platform.

Table 1: Wheat cultivars examined along with their geographic locations, ploidy levels and pedigrees.

Mapping of Reads to Reference CP Genome

Between 101.34 to 195.28 million 100-bp paired-end reads were generated for each cultivar from 500-bp insert library. Adapter sequences in the raw data were deleted, and reads with 50% low quality bases (quality value ≤ 5) or more were discarded. The remaining sequences of different cultivars were first mapped to the published wheat mt genome (acc. no. AP008982) before mapping to cp genome (acc. no. AB042240) using CLC Genomics Workbench (version 3.0, http://www.clcbio.com/user  manuals). All cp reads that aligned to mt genome were removed before cp genome assembly.

Sequence Annotation

Annotation was carried out by mapping cp genome sequences with BLAST hits (identity 90% and overlap 90%)¹⁸ to known plastid genes. Then, sequences were tested for consistency of the ORFs using NCBI online tool of the ORF finder (http://www.ncbi.nlm.nih.gov/projects/gorf/, the standard genetic code was applied). Gene and exon boundaries were determined by alignment of homologous genes from wheat and several other common plastid angiosperm genomes. The tRNA genes were identified by using BLAST search tools,¹⁸ and the tRNAscan-SE program (version 1.4 with default parameters).¹⁹ Repetitive sequences were identified using the REPuter (version 2.74; length ≥ 50 bp; mismatch ≤ 3 mismatches).²⁰ Then, information on tandem repeats were identified using a tandem repeat finder (http://tandem.bu. edu/trf/trf.html, Benson²¹).

Identification of SNP and INDELs and Phylogenetic Analyses

As extra step of filtering was made by the removal of sequences in the reference cp genome corresponding to the gaps of ≥ 5 bp in all the nine wheat cultivars to avoid bias in the resulted INDELs analysis. Gaps in the cp genome of the nine cultivars that generated by the reference cp genome with less than 5 bp were considered insertions. However, gaps generated during alignment only in the reference cp genome were all considered as deletions. The mapping results after the third filtering were, then, used for SNPs/INDELs identification based on a Bayesian algorithm according to the BioScope software (version 1.3) guide used as visual double-check. Only SNPs/INDELs with a read depth of ≥ 30, mapping quality of ≥ 30 and SNPs/INDELs quality of ≥ 20 were retained.

Data matrices of different cultivar pairs were entered into TFPGA (version 1.3) and analyzed using qualitative routine and dissimilarity coefficients were utilized in drawing dendrogram using unweighted pair group method with arithmetic average (UPGMA) and Neighbor Joining (NJ) routine using NTSYSpc (version 2.10, Exeter software). The bootstrap value was set to 100. All other parameters are set as default. 

Results and Discussion

Mapping of Reads to Reference Genome

The number of reads mapped to the cp genomes of the nine wheat cultivars ranged between 281,499-2,169,718 with CG representing 38.31% and mapped reads average representing 1.1% of the total reads (Table 2, Supplementary Files 1-9). Mapping of the reads to the reference wheat cp genome (acc. no. AB042240, Ogiharaet al.,³) resulted in 100% coverage of the genome. Removal of reads aligned to the wheat mt genome reduced the number of cp reads to 219,147-1,440,201, which represents an average of 0.73% of the total reads with mean filtered coverage of 644-1,450 (Table 2). As all reads that mapped to mitochondrial genome were eliminated, we confidently declare that intra-cultivar heteroplasmy for the different cultivars does not exist in alignment with the results in cp genomes of many other angiosperms, ex., B. hygrometrica, in which no intraSNPs were found.²² The intraSNPs have been demonstrated to be present in both cp and mt genomes in rice.²³ Additionally, in our earlier study on date palm cp genome following the same approach of removal of reads mapped to mt genome, we detected a number of intraSNPs that reflects plastid heteroplasmy.²⁴ This data confirmed that date palm cp genomes are heteroplasmic and scoped the light on the necessity to be cautious when analyzing SNP from data generated from next generation sequencing of total genomic DNA of other crop plants.

Table 2: Statistics of DNA numerical data analysis for the nine wheat cultivars aligned to the chloroplast reference genome (acc. no. AB042240).

Comparative Analysis of Plastomes of Several Angiosperms

Although the nuclear wheat genome (~16-17 Gb) is about 3-35 fold larger than other cereals, like rice (0.43 Gb) and barley (5.3 Gb), the plastid genome (133,812 bp) is the smallest among angiosperms including cereals, after Marchantia polymorpha (121,024 bp), and the total number of gene types (97), either protein coding, tRNA or rRNA genes, is the least among angiosperms (Table 3). The detailed gene content of wheat plastome is shown in Table 4. The largest known cp genome among angiosperms is that of Chara vulgaris (184,933 bp).²² Plastid genome of the latter species also has the highest AT% (73.8%) and repeats % (3.162%) among angiosperms. The coding percentage in wheat cp genome is intermediate among angiosperms; date palm cp genome has the highest (99.39%). The number of tandem repeats of wheat cp genome is the highest (5) among published cp genomes of other angiosperm. However, cp genome of Chara vulgaris possesses the highest number of long repeats (120) among angiosperms (Figure 1).

Figure 1: Number of tandem and long repeats in plastomes of several angiosperms. Click here to View Figure

Table 3: Comparative analysis of genomic features among 12 chloroplast genomes of angiosperms

*       Total/protein coding/tRNA/rRNA

**     This size was corrected (Bahieldin et al. 2014), which is 728 bp shorter than the published wheat plastome (Ogihara et al. 2000)

***   A number of 74 protein-coding genes and two unidentified ORFs (ycf3 & ycf4)

**** A number of 30 tRNA genes plus three new sequences detected in the present study

Table 4: The gene content across the nine assembled Triticum aestivum chloroplast genomes.

Plastome Structure

Plastid nucleotide sequence of G168, as a model, was submitted to the NCBI and received the accession no. KJ592713. Plastome of the wheat cultivar along with gene content were generated earlier by our group.²⁵ Our results indicated a number of three new non-coding genes, e.g., trnI, trnT and trnfM (Figure 2) located in the LSC region; of which the first two are shown in a cluster. Additionally, function of one out of three conserved ORFs, namely ycf5 was assigned after annotation (Figure 2). The latter gene, also called ccsA, functions as a cytochrome c-type biogenesis protein required for heme attachment to chloroplast cytochromes.²⁶ Functions of the two other conserved ORFs, namely ycf6 and ycf9 were also deciphered.²² Respectively, they are named pbsZ and petN genes. The first functions in photosystem II, while the second functions as a cytochrome in the generation of ATP via electron transport.

Figure 2: Plastome of wheat cultivar G168 indicating the gene content. Click here to View Figure

A total of 19,770 codons representing the coding capacity of all protein-coding genes of wheat cp genome were scored (Table 5). Among them, as high as 2,118 (10.71%) codons encode for leucine, while as low as 214 (1.08%) codons encode for cysteine. Yang et al.⁵ indicated that isoleucine and cysteine are the most and least amino acids in plastid genome in terms of number of codons in date palm cp genome, respectively, (see Table 1, Yang et al.⁵). The most frequent codon (825) was scored for AUU encoding isoleucine. Similar conclusion was reached by Yang et al.⁵ in their study on date palm cp genome. Our results also indicated that nucleotide frequencies vary at different codon positions. At the first position, “A” nucleotide is found the most frequent nucleotide (29.59%), followed by “G” (28.55%). The nucleotide “C” is the least (18.66%) at the first position. This indicates that purine is favored at the first position. At the second position, “U” is found as the most frequent nucleotide (32.70%), followed by “A” (27.61%). The nucleotide “G” scores the least (18.55%). At the third position, “U” also is the most frequent nucleotide (37.64%), followed by “A” (32.57%). The nucleotide “C” is the least frequent nucleotide (14.26%). These results indicate that “U” is favored for change at the second and third positions of the codon. Similar tendency of results was found when studying codon usage in mitochondrial genome of wheat.¹¹ This indicates that AT-rich genes in cp genome might be less conserved that CG-rich genes. Date palm also showed the same trend of results, except that nucleotide “C”, not “G”, is the least frequent at the first position of the codon in its plastide genome (Calculated from data in Table 1 of Yang et al.,⁵). The results of the relative synonymous codon usage (RSCU) indicated that UUA codon coding for leucine is the most common (2.07) compared to the other codons of leucine or for any other amino acids (Table 5). This indicates that cp genes display a non-random usage of synonymous codons. The results also indicated that UAA is the most frequently-used stop codon (54.9%). A number of 28, out of the sense 61 codons, covering all the 20 amino acids have tRNAs existed in wheat plastome. Interestingly, most of the tRNAs are specific for less frequent codons. Therefore, the phenomenon of codon preference in wheat plastome is not only explained by the frequency by which a certain codon of a given amino acid exists, but also by the availability of the cognate tRNA of such a codon (Table 5).

Table 5: Codon usage and codon-anticodon recognition pattern for tRNA in nine assembled wheat chloroplast genomes

*RSCU: Relative synonymous codon usage

Snps and Indels Analyses and Cultivars Relationships

Across the different Egyptian cultivars, 564 SNPs and 160 INDELs were identified in the study, of which 230 and 4 are in the protein-coding regions, respectively (Table 6). The number of monomorphic SNPs and INDELs are 553 and 154, respectively. A number of 212 SNPs were found in the long inverted repeat (IR) regions, of which 104 were found in the IRa and 108 were found in the IRb region. One SNP, while none for INDELs, was found in the genic regions unique to one of the two inverted repeats (IRa) in the coding sequence of ndhB gene. The similarity of SNPs patterns in both IR regions is due to the fact that cp genome is conserved. However, there is a possibility that one single read within these regions might be mapped to either region. This possibility reduces the chance to detect the different patterns, if existed, of the IR region. Therefore, SNPs analysis using next generation sequencing of total genomic DNA should be taken cautiously. It is likely that the duplication of the IR region took place way after the occurrence of point mutations during evolution. Numbers of inter-cultivar polymorphic and cultivar-specific SNPs were nine and five, respectively (Table 6). The latter number was scored only in the intergenic spacers (IGS) region for cultivar BSF4. Among the polymorphic SNPs, six were found in the IGS regions, while only one was found in the introns (IN) of atpF gene and two SNPs were found in the GN regions of proA and rpl16 genes. Numbers of 15 and nine polymorphic and cultivar-specific INDELs were also found of which 10 and eight INDELs, respectively, are located in the IGS regions, while five polymorphic and one cultivar-specific INDEL, respectively, are located in the IN regions of the rpl16 gene.

Table 6: SNPs and INDELs within plastid genomes of the nine Egyptian wheat cultivars as sorted by position and region of the genome. Plastid genome of Chinese Spring cultivar was used as the reference genome (acc. no. AB042240). GN refers to protein-coding genic regions, IN refers to intron regions and IGS refers to intergenic spacer regions, S refers to synonymous substitution, NS refers to non-synonymous. Letters in INDELs refer to insertions and (-) refers to deletions. Red blocks refer to SNPs in the protein-coding regions. Green blocks indicate SNPs unique to one of the two inverted repeats (IR) regions. Blue block indicates the unique SNP to one IR (IRa) region. Orange blocks indicate INDELs within the IR region that showed similar patterns in the two regions.

The highest number of SNPs in the protein coding regions was scored for gene ndhB (31 and 30 in the IRa and IRb regions, respectively), followed by rpl2 (18 in each IR region), then matK (34) and psbC (25). One long INDEL of 12 nucleotide exists in the rpl2 gene and starts at nucleotide 4 of the gene (Figure 3). Nucleotide sequence of this INDEL encodes for four amino acids (LNNT). Two other INDELs that are 19-nt apart starting from nucleotide 160 of the gene were detected in the rpl2 gene (Figure 3). The first is an inserted nucleotide in the nine wheat cultivars, while the second was a deleted nucleotide compared to Chinese Spring cultivar. The latter two INDELs resulted in a frameshift of six amino acids, with a glycine amino acid in the middle remains unchanged, then default frame was regained (Figure 3). We concluded that rpl2 gene in the reference genome is 12-nt shorter than that of the Egyptian cultivars. It is unlikely that the change in these amino acids has posed any functional constraints on proteins encoded by either versions of the gene as they were proven to be effectively functioning.

Figure 3: Alignment of the rpl2-encoded amino acids sequence of the cultivar Click here to View Figure

Based on the SNPs of the different nine cultivars in addition to the reference plastid genome, dendrogram was constructed (Figure 4). The tree was well-resolved with high bootstrap support for resolved nodes. This might be due to the fact that the Egyptian cultivars are closely related on one hand, and genetically distant from the reference genome, on the other hand. The results indicated the correspondence between tree topology and linage of eight out of the nine cultivars. The cultivar pairs G168/SWL, SHK94/SKH95 and MSR2/SDS13 are closely related. In other words, the cultivars with shared ancestors showed genetically closer relationships. As no information is available on the lineage of SKH95, it is likely that it shares a common ancestor with SKH94. Interestingly, the tetraploid cp genome was closely related to the other Egyptian hexaploid wheat cultivars as compared to the reference hexaploid wheat cultivar Chinese Spring. The SNPs/INDELs tree was not resolved and bootstrap support values were low (data provided upon request). This is due to the fact that some INDELs might be artifacts rather than real. The INDELs inside the IR region are more reliable as they should show similar patterns in the two inverted regions.

Figure 4: Phylogenetic analysis using chloroplast sequences from nine wheat cultivars Click here to View Figure

There are no intra-cultivar polymorphic SNPs were detected. This might be due to the fact that sequences of the mt genome mapped to the cp genome were filtered out and artifacts were removed before cp genome assembly. Generally speaking, intra-varietal heteroplasmy in the wheat cp genome within the studied cultivars does not exist in contradiction with previous reports in other plants.^5,27 

Conclusion

We conclude that plastome SNPs and INDELs successfully separated wheat cultivars and results aligned with the known ancestral information of the different genotypes.

Conflict of Interest

Authors declare no conflict of interest including grants, membership, employment, ownership of stock or any other interest or non‐financial interest. 

References

1. Bausher, M.G, Singh, N.D,, Lee, S.B,, Jansen, R.K., Daniell, H. The complete chloroplast genome sequence of Citrus sinensis (L.) Osbeck var ‘Ridge Pineapple’: organization and phylogenetic relationships to other angiosperms. BMC Plant Biol. 2006;6:21.
   CrossRef
2. Howe, C.J., Barbrook, A.C., Koumandou, V.L., Nisbet, R.E., Symington, H.A., et al. Evolution of the chloroplast genome. Trans. R. Soc. Lond. B. Biol. Sci. 2003;358:99–106; discussion 106-7.
   CrossRef
3. Ogihara, Y., Isono, K., Kojima, T., Endo, A., Hanaoka, M., et al. Chinese Spring wheat (Triticum aestivum) chloroplast genome: Complete sequence and contig clones. Plant Mol. Biol. Rep. 2000;18:243-53.
   CrossRef
4. Ogihara, Y., Yamazaki, Y., Murai, K., Kanno, A., Terachi, T., et al. Structural dynamics of cereal mitochondrial genomes as revealed by complete nucleotide sequencing of the wheat mitochondrial genome. Nucleic Acids Res. 2005;33:6235-50.
   CrossRef
5. Yang, M., Zhang, X., Liu, G., Yin, Y., Chen, K., et al. The complete chloroplast genome sequence of date palm (Phoenix dactylifera). PLoS ONE 2010;5:e12762.
   CrossRef
6. Chumley, T.W., Palmer, J.D., Mower, J.P., Fourcade, H.M., Calie, P.J., et al. The complete chloroplast genome sequence of Pelargonium x hortorum: organization and evolution of the largest and most highly rearranged chloroplast genome of land plants. Biol. Evol. 2006;23:2175-90.
   CrossRef
7. Hansen, A.K., Escobar, L.E., Gilbert, L.E., Jansen, R.K. Paternal, maternal, and biparental inheritance of the chloroplast genome in Passiflora (Passifloraceae): Implications for phylogenetic studies. J. Bot. 2007a;94:42-6.
   CrossRef
8. Hansen, D.R., Dastidar, S.G., Cai, Z., Penaflor, C., Kuehl, J.V., et al. Phylogenetic and evolutionary implications of complete chloroplast genome sequences of four early-diverging angiosperms: Buxus (Buxaceae), Chloranthus (Chloranthaceae), Dioscorea (Dioscoreaceae), and Illicium (Schisandraceae). Phylogenet. Evol. 2007b;45:547-63.
   CrossRef
9. Mardanov, A.V., Ravin, N.V., Kuznetsov, B.B., Samigullin, T.H., Antonovm A.S., et al. Complete sequence of the duckweed (Lemna minor) chloroplast genome: structural organization and phylogenetic relationships to other angiosperms. Mol. Evol. 2008;66:555-64.
   CrossRef
10. Ling, H.-Q., Zhao, S., Liu, D., Wang, J., Sun, H., et al. Draft genome of the wheat A-genome progenitor Triticum urartu. Nature 2013;496:87-90.
    CrossRef
11. Cui, P., Liu, H., Lin, Q., Ding, F., Zhuo, G., et al. A complete mitochondrial genome of wheat (Triticum aestivum Chinese Yumai), and fast evolving mitochondrial genes in higher plants. J. Genet. 2009;88:299-307.
    CrossRef
12. Fang, Y., Wu, H., Zhang, T., Yang, M., Yin, Y., et al. A complete sequence and transcriptomic analyses of date palm (Phoenix dactylifera) mitochondrial genome. PLoS ONE 2012;7:e37164.
    CrossRef
13. Khan, A., Khan, I.A., Heinze, B., Azim, M.K. The chloroplast genome sequence of Date palm (Phoenix dactylifera cv. ‘Aseel’). Plant Mol. Biol. Rep. 2012;30:666–78.
    CrossRef
14. Birky, C.W. Relaxed cellular controls and organelle heredity. Science 1983;222:468-75.
    CrossRef
15. Chat, J., Decroocq, S., Decroocq, V., Petit, R.J. A case of chloroplast heteroplasmy in Kiwifruit (Actinidia deliciosa) that is not transmitted during sexual reproduction. The J. Hered. 2002;93:293-300.
    CrossRef
16. Frey, J.E., Frey, B., Forcioli, D. Quantitative assessment of heteroplasmy levels in Senecio vulgaris chloroplast DNA. Genetica 2005;123:255-61.
    CrossRef
17. Gawel, N.J., Jarret, R.L. A modified CTAB DNA extraction procedure for Musa and Ipomoea. Plant Mol. Biol. Rep. 1991;9:262-66.
    CrossRef
18. Altschul, S.F., Madden, T.L., Schaffer, A.A., Zhang, J., Zhang, Z., et al. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res. 1997;25:3389-3402.
    CrossRef
19. Lowe, T.M., Eddy, S.R. tRNAscan-SE: a program for improved detection of transfer RNA genes in genomic sequence. Nucleic Acids Res. 1997;25:955-64.
    CrossRef
20. Kurtz, S., Choudhuri, J.V., Ohlebusch, E., Schleiermacher, C., Stoye, J., et al. REPuter: the manifold applications of repeat analysis on a genomic scale. Nucleic Acids Res. 2001;29:4633-42.
    CrossRef
21. Benson, G. Tandem repeats finder: a program to analyze DNA sequences. Nucleic Acids Res. 1999;27:573-80.
    CrossRef
22. Zhang, T., Fang, Y., Wang, X., Deng, X., Zhang, X., et al. The complete chloroplast and mitochondrial genome sequences of Boea hygrometrica: Insights into the evolution of plant organellar genomes. PLoS ONE 2012;7:e30531.
    CrossRef
23. Tang, J., Xia, H., Cao, M., Zhang, X., Zeng, W., et al. A comparison of rice chloroplast genomes. Plant Physiol. 2004;135:412-20.
    CrossRef
24. Sabir, J.S.M., Arasappan, D., Bahieldin, A., Abo-Aba, S., Bafeel, S., et al. Whole mitochondrial and plastid genome SNP analysis of nine date palm cultivars reveals plastid heteroplasmy and relationships among cultivars. PloS ONE 2014;9:e94158.
    CrossRef
25. Bahieldin, A., Al-Kordy, M.A., Shokry, A.M., Gadalla, N.O., Al-Hejin, A.M.M., et al. Corrected sequence of the wheat plastid genome. R. Biol. 2014;337:499-502.
    CrossRef
26. Feissner, R.E., Beckett, C.S., Loughman, J.A., Kranz, R.G. Mutations in cytochrome assembly and periplasmic redox pathways in Bordetella pertussis. Bacteriol. 2005;187:3941-9.
    CrossRef
27. Straub, S.C.K., Parks, M., Weitemier, K., et al. Navigating the tip of the genomic iceberg: next-generation sequencing for plant systematics. J. Bot. 2012;99:349-64.
    CrossRef