The complete chloroplast genome sequence of Citrus sinensis (L.) Osbeck var 'Ridge Pineapple': organization and phylogenetic relationships to other angiosperms

Abstract

Background: The production of Citrus, the largest fruit crop of international economic value, has recently been imperiled due to the introduction of the bacterial disease Citrus canker. No significant improvements have been made to combat this disease by plant breeding and nuclear transgenic approaches. Chloroplast genetic engineering has a number of advantages over nuclear transformation; it not only increases transgene expression but also facilitates transgene containment, which is one of the major impediments for development of transgenic trees. We have sequenced the Citrus chloroplast genome to facilitate genetic improvement of this crop and to assess phylogenetic relationships among major lineages of angiosperms.

Results: The complete chloroplast genome sequence of Citrus sinensis is 160,129 bp in length, and contains 133 genes (89 protein-coding, 4 rRNAs and 30 distinct tRNAs). Genome organization is very similar to the inferred ancestral angiosperm chloroplast genome. However, in Citrus the infA gene is absent. The inverted repeat region has expanded to duplicate rps19 and the first 84 amino acids of rpl22. The rpl22 gene in the IRb region has a nonsense mutation resulting in 9 stop codons. This was confirmed by PCR amplification and sequencing using primers that flank the IR/LSC boundaries. Repeat analysis identified 29 direct and inverted repeats 30 bp or longer with a sequence identity > or = 90%. Comparison of protein-coding sequences with expressed sequence tags revealed six putative RNA edits, five of which resulted in non-synonymous modifications in petL, psbH, ycf2 and ndhA. Phylogenetic analyses using maximum parsimony (MP) and maximum likelihood (ML) methods of a dataset composed of 61 protein-coding genes for 30 taxa provide strong support for the monophyly of several major clades of angiosperms, including monocots, eudicots, rosids and asterids. The MP and ML trees are incongruent in three areas: the position of Amborella and Nymphaeales, relationship of the magnoliid genus Calycanthus, and the monophyly of the eurosid I clade. Both MP and ML trees provide strong support for the monophyly of eurosids II and for the placement of Citrus (Sapindales) sister to a clade including the Malvales/Brassicales.

Conclusion: This is the first complete chloroplast genome sequence for a member of the Rutaceae and Sapindales. Expansion of the inverted repeat region to include rps19 and part of rpl22 and presence of two truncated copies of rpl22 is unusual among sequenced chloroplast genomes. Availability of a complete Citrus chloroplast genome sequence provides valuable information on intergenic spacer regions and endogenous regulatory sequences for chloroplast genetic engineering. Phylogenetic analyses resolve relationships among several major clades of angiosperms and provide strong support for the monophyly of the eurosid II clade and the position of the Sapindales sister to the Brassicales/Malvales.

Figure 1

Circular gene map of Citrus sinensis chloroplast genome. The thick lines indicate the extent of the inverted repeats (IRa and IRb, 26,996 bp), which separate the genome into small (SSC, 18,393 bp) and large (LSC, 87,744) single copy regions. Genes on the outside of the map are transcribed in the clockwise direction and genes on the inside of the map are transcribed in the counterclockwise direction.

Figure 2

Maximum parsimony tree based on 61 chloroplast protein-coding genes [69]. The single most parsimonious phylogram has a length of 53,085, a consistency index of 0.45 (excluding uninformative characters), and a retention index of 0.60. Numbers at nodes indicate bootstrap support values and branch length scales are shown at base of the tree. Taxa in red are members of the eurosid II clade. Thicker lines in tree indicate members of eudicots. Black bars indicate lineages that have lost infA. Accession numbers for taxa are: Pinus, NC_001631; Ginkgo, DQ069337-DQ069702, Amborella, NC_005086, Nuphar, DQ069337-DQ069702, Nymphaea, NC_006050; Acorus, DQ069337-DQ069702; Oryza, NC_001320; Saccharum, NC_006084; Triticum, NC_002762; Typha, DQ069337-DQ069702; Yucca, DQ069337-DQ069702; Zea, NC_001666; Calycanthus, NC_004993; Arabidopsis, NC_000932; Atropa, NC_004561; Cucumis, NC_007144; Eucalyptus, AY780259; Glycine, NC_007942; Gossypium, NC_007944; Citrus, DQ864733; Lotus, NC_002694; Medicago, NC_003119; Nicotiana, NC_001879; Oenothera, NC_002693; Panax, NC_006290; Ranunculus, DQ069337-DQ069702; Solanum lycopersicum, DQ347959; Solanum bulbocastanum NC_007943; Spinacia, NC_002202; Vitis, NC_007957.

Figure 3

Maximum likelihood tree based on 61 chloroplast protein-coding genes. The single maximum likelihood phylogram has a ML value of – lnL = 305916.24523. Numbers at nodes indicate bootstrap support values and branch length scale is shown at base of the tree. Taxa in red are members of the eurosid II clade. Thicker lines in trees indicate members of eudicots. Black bars indicate lineages that have lost infA.