The complete chloroplast genome of Cinnamomum camphora and its comparison with related Lauraceae species

Samples and genome sequencing

Fresh young leaves of Cinnamomum camphora were obtained from the campus of Nanjing Forestry University. The chloroplasts of Cinnamomum camphora were isolated using the Sigma Chloroplast DNA Isolation kit (Sigma, St. Louis, MO, USA), and chloroplast DNA was extracted using DNaesy Plant Mini Kit (QIAGEN, Hilden, Germany). The purified DNA was subjected to hydroshearing, end repair and then interrupted randomly to construct 350 bp libraries. The complete library, with an average read length of 150 bp, was sequenced using Illumina Hiseq2000 platform.

Chloroplast genome assemble and annotation

To ensure accurate and reliable analyses, raw reads were proofread and assembled with SOAP denovo (Li et al., 2008). The generated contigs were assembled using the chloroplast genome sequence of Cinnamomum micranthum (KT348516.1) as a reference. The chloroplast genome sequences of Cinnamomum camphora were annotated through the online program Dual Organellar Genome Annotator (Solovyev et al., 2006). The annotation results were manually checked, including the start and stop codons, and adjusted by comparison to homologous genes from other sequenced chloroplast genomes. Transfer RNA (tRNA) genes were verified using tRNA scan-SE in organellar search mode with default parameters (Schattner, Brooks & Lowe, 2005). The circular chloroplast genome map of Cinnamomum camphora was drawn using the OGDRAW program (Lohse, Drechsel & Bock, 2007). Nucleotide frequency and relative synonymous codon usage (RSCU) (Sharp, Tuohy & Mosurski, 1986) were analyzed using DAMBE (Xia, 2013) on the protein-coding genes and only genes in inverted repeat region A (IRA) were used to represent repeated genes.

Repeat structures and simple sequence repeats (SSRs) analysis

REPuter was used to identify forward and palindromic repeats with a minimal size of 30 bp, hamming distance of three and over 90% identity (Kurtz et al., 2001). Tandem repeat sequences were identified in Cinnamomum camphora using Tandem Repeats Finder with default parameters (Benson, 1999). Simple sequence repeats (SSRs) were detected using the microsatellite identification tool MISA (http://pgrc.ipk-gatersleben.de/misa/) with the following thresholds: minimum SSR motif length of 10 bp, and 10 repeat units for mononucleotide SSRs, five repeat units for dinucleotide SSRs, four repeat units for trinucleotide SSRs, and three repeat units for tetra-, penta- and hexanucleotide SSRs. The maximum size of interruption allowed between two different SSRs in a compound SSR was 100 bp (von Stackelberg, Rensing & Reski, 2006). All of the repeats identified with the above programs were manually verified to remove redundant results.

Comparative analysis of different Lauraceae plastomes

To encompass the complete nucleotide diversity among Lauraceae species, the complete chloroplast genome sequences were aligned using MAFFT 7.222 software (Katoh & Standley, 2013), and manually adjusted with BioEdit software (Hall, 1999). DnaSP 5.0 software was used to conduct a sliding window analysis to calculate the nucleotide variability (Pi) values (Librado & Rozas, 2009). Pi is defined as the average number of nucleotide differences per site between two DNA sequences in all possible pairs in the sample population (Nei & Li, 1979). The window length was set to 600 bp, and the step size was set to 200 bp.

Phylogenetic analysis

A molecular phylogenetic tree was constructed using 26 different plant species with TreeBeST (http://treesoft.sourceforge.net/treebest.shtml). Among these 26 taxa, Abies koreana, Picea sitchensis, Pinus taiwanensis, Podocarpus lambertii, Cycas revolute and Ginkgo biloba were set as the outgroup. The 26 completed chloroplast genome sequences representing the lineages of angiosperms were downloaded from the NCBI Organelle Genome Resource database. The protein-coding sequences of related Lauraceae species were determined by MUSCLE (Edgar, 2004). A phylogenetic tree was constructed based on a neighbor joining analysis. The bootstrap probability of each branch was calculated by 1,000 replications.

Genome organization and gene features

The complete chloroplast genome size of Cinnamomum camphora was 152,570 bp, the same as Cinnamomum micranthum (Wu, Ho & Chang, 2016). The chloroplast genome had a single circular chromosome with a quadripartite structure, which included a pair of IR regions (19,886 bp) that were separated by a large single copy (LSC, 93,705 bp) and a small single copy (SSC, 19,093 bp) regions (Fig. 1). Coding regions (83,429 bp), comprising protein-coding genes (72,222 bp), tRNA genes (2,271 bp) and rRNA genes (8,936 bp) accounted for 54.68% of the genome, whereas non-coding regions (69,141 bp) accounted for the remaining 45.32% of the genome. The overall GC content of the Cinnamomum camphora was 39.13%. The IR regions had a higher GC content of 44.42%, while the GC contents of the LSC and SSC were 37.96% and 33.92%, respectively. The high GC percentage in the IR regions was similar to most reported chloroplast genomes, which could be the result of ribosomal RNA in this region (Asaf et al., 2016; Qian et al., 2013).

A total of 123 coding regions were identified in the chloroplast genome of Cinnamomum camphora, of which 107 were unique (Fig. 1 and Table 1). In total, 79 (73 unique) protein-coding genes were involved in the processes related to photosynthesis, the genetic system and several currently unknown functions. Additionally, 36 (30 unique) genes encoded tRNAs and eight (four unique) rRNA genes. Like the genes, the introns of chloroplast genomes were basically conserved. In Cinnamomum camphora, seven protein-coding genes (rps16, petD, ndhB (2×), atpF, rpoC1 and rpl2) contain one intron, and four protein-coding genes (ycf3, clpP and rps12 (2×)) contain two introns (Table S1). The loss of introns within the protein-coding genes reported in other plant species has not been found in the chloroplast genome of Cinnamomum camphora (Daniell et al., 2008; Jansen et al., 2008; Wu et al., 2009). As in many other land plants, rps12 was located with a single 5′ end in the LSC region and a repeated 3′ end in both of the IR regions (Raman & Park, 2016; Redwan, Saidin & Kumar, 2015; Yang et al., 2013). There were 15 genes duplicated in the IR regions, including three protein-coding genes, eight tRNAs and four rRNAs. Overall, the gene content, number and structure in chloroplast genomes were generally similar to those of other reported Lauraceae species (Hinsinger & Strijk, 2016; Song et al., 2015, 2016; Wu, Ho & Chang, 2016).

The sequence analysis indicates 79 protein-coding genes in this genome represented 63,654 bp and 21,218 codons. On the basis of the sequences of protein-coding genes, the frequency of codon usage was calculated (Table 2). Among these codons 2,203 (10.87%) encode leucine, while 255 (1.25%) encode cysteine, which are the most and least used amino acids, respectively. The overall codon bias pattern in of the Cinnamomum camphora genome tended to use A/U—ending codons, and among the 29 preferred codons in the Cinnamomum camphora chloroplast genome (RSCU > 1.0), 27 ended with A/U. This phenomenon is similarly observed in many other chloroplast genomes (Nie et al., 2012; Zhang et al., 2016).

Comparative analysis of different Lauraceae plastomes

We compared four other reported chloroplast genomes of representative taxa in the Lauraceae family with that of Cinnamomum camphora (Table S2). These taxa included Cinnamomum micranthum (152,570 bp, KR014245.1), Persea americana (152,723 bp, KX437771.1), Machilus yunnanensis (152,721 bp, KT348516.1) and Litsea glutinosa (152,618 bp, KU382356.1). To examine the level of sequence divergence of the Lauraceae family, the Pi values within 600 bp were calculated with DnaSP 5.0 software among the five chloroplast genomes of Cinnamomum camphora, Cinnamomum micranthum, M. yunnanensis, L. glutinosa and P. americana. Between the two Cinnamomum species, Pi values varied from 0 to 0.025 (ycf2) with a mean of 0.00362, indicating that the differences between the two Cinnamomum genomes were small (Fig. 2A). However, five of the genes (ycf2, rrn23, ycf1, trnL-UAG and ndhF) showed high levels of variation, which were much higher than the values of other regions (Pi > 0.02). Among the five Lauraceae species, the Pi values ranged from 0 to 0.3178 (ndhH) with a mean of 0.03361, indicating that the differences among different species of the genera in the Lauraceae family were greater than those between congeneric species (Fig. 2B). Particularly, the entire SSC regions were highly variable among the different genera in the Lauraceae family. Coding regions, including those of ycf1, psaC, ccsA, rpl32 and a set of genes named ndh, have been identified as highly variable regions in the SSC region. The SSCs often have a higher nucleotide substitution rates relative to the IRs in land plants (Perry & Wolfe, 2002; Zhang, Ma & Li, 2011). These comparatively highly variable loci are good for exploiting molecular markers and evaluating interspecies phylogenetic relationships (Chen et al., 2015; Dong et al., 2012; Timme et al., 2007).

Repeat and SSR analyses

Repeat motifs are very useful in the analysis of genome rearrangements and play an important role in phylogenetic analyses (Cavalier-Smith, 2002). A total of 40 repeat structures with a length at least 30 bp were detected in the chloroplast genome of Cinnamomum camphora. Similarly, 61, 45, 46 and 40 repeat structures were detected in Cinnamomum micranthum, L. glutinosa, M. yunnanensis and P. americana (Fig. 3D). Cinnamomum micranthum had the greatest total number of repeats, and the other four Lauraceae species showed similar numbers and patterns of repeats. For repeat analysis of Cinnamomum camphora, 12 forward and 12 palindromic repeats were found with a size of 30–44 bp, whereas only one forward and one palindromic repeats were 15–29 bp in length. Similarly, 13 tandem repeats were 15–29 bp, and one tandem repeat was 30–44 bp in length (Figs. 3A–3C). The presence of these repeats indicated that the region is a potential hotspot for genome reconfiguration (Asano et al., 2004; Gao et al., 2009). Additionally, these repeats were informative resources for developing genetic markers for phylogenetic and population genetic studies (Nie et al., 2012).

Simple sequence repeats, also known as microsatellites, are 1–6 bp repeating sequences that are widely distributed throughout the chloroplast genome. SSRs are typically co-dominant and have a higher degree of polymorphism. Here, we detected perfect SSRs with a minimum size of 10 bp in Cinnamomum camphora and four other Lauraceae species. The SSRs were interrupted by a maximum distance of 100 bp. Based on the analysis, 83 perfect SSRs were detected in the chloroplast genome of Cinnamomum camphora. Similarly, 88, 81, 82 and 86 were identified in Cinnamomum micranthum, L. glutinosa, M. yunnanensis and P. americana (Fig. 4A). The majority of the SSRs in these chloroplast genomes were mononucleotides, ranging from 54 in L. glutinosa to 65 in Cinnamomum micranthum. Additionally, only one hexanucleotide was present in all of the Lauraceae species. In Cinnamomum camphora, the most abundant motif, at 74.50%, was a run of mononucleotide A/T (Fig. 4B). This result confirmed that the chloroplast SSRs found in most plants are generally composed of polythymine or polyadenine repeats, and infrequently contain cytosine and guanine repeats (Choi, Chung & Park, 2016; Kuang et al., 2011). The total output consisted of 83 SSRs: 76% (63 SSRs) in the LSC region, 19% (16 SSRs) in the SSC region, and 5% (four SSRs) in the IR regions (Table S3). In comparison with the IR region, the SSRs were more prevalent in the LSC and SSC regions. SSRs in coding regions are prone to mutation and cause frame-shifts to occur, which renders the gene non-functional (Wang, Barkley & Jenkins, 2009). Chloroplast SSRs have been used to evaluate genetic variations among plant genotypes (Vendramin et al., 1999) and to investigate the genetic diversity of Lauraceae species (Santos, Spironello & Sampaio, 2008; Zhai et al., 2010). The SSRs in this analysis are good resource for developing molecular markers and will be applied to molecular marker-assisted breeding, population genetics and genetic linkage map construction (Deguilloux, Pemonge & Petit, 2004; Huang et al., 2014).

Phylogenetic analysis

Chloroplast genome sequences are useful for deciphering phylogenetic relationships among closely related taxa and for clarifying the evolutionary patterns of plant species (Jansen et al., 2007; Kyunghee et al., 2015). To examine the phylogenetic position of Cinnamomum camphora in the Lauraceae family, the core protein-coding genes that are common to all 26 chloroplasts were used to infer their phylogenetic relationships (Fig. 5). Several species in the gymnosperm were set as outgroups. The alignment analysis was conducted by MUSCLE. A neighbor joining analysis was performed with TreeBeST using 1,000 bootstrap replicates. The long branches indicated faster rates of plastid sequence evolution compared with other members in the tree. The tree suggested the correct phylogeny inference, followed by the latest angiosperm phylogeny group III (Bremer et al., 2009). The phylogenetic tree indicated that Calycanthus is most closely related to Lauraceae, which are both members of Laurales and Laurales forms a sister group to Magnoliids. In corroboration with other studies, Endiandra discolor were first separated from the other Lauraceae species (Hinsinger & Strijk, 2016; Rohwer & Rudolph, 2005). Furthermore, the position of Cinnamomum camphora was clustered with Cinnamomum micranthum, both of which are members of Cinnamomum. The chloroplast genome of Cinnamomum camphora will provide valuable and essential genetic information to further the phylogenetic resolution among angiosperms (Leebens-Mack et al., 2005; Moore et al., 2007; Ruhfel et al., 2014).