Assembly and characterization of the mitochondrial genome of Camellia pitardii Coh.-St

Technical Summary: Assembly and Characterization of the Camellia pitardii Mitochondrial Genome

Problem Statement
Camellia pitardii Coh.-St. is an economically significant wild tree species in the genus Camellia, valued for its seed oil and ornamental qualities. Despite its importance, the species lacks a systematically characterized mitochondrial genome (mitogenome). While nuclear genome assemblies and phenotypic studies exist, the structural architecture, gene content, and evolutionary dynamics of the C. pitardii mitogenome remain unexplored. This gap is critical because plant mitochondrial genomes are known for their structural complexity, including size variations, abundant repetitive sequences, and frequent recombination events that create dynamic isoforms, which short-read sequencing alone often fails to resolve accurately.

Methodology
To address the lack of genomic data, the authors employed a hybrid assembly strategy combining Illumina NovaSeq 6000 short reads and Oxford Nanopore long reads.

• Sequencing and Assembly: High-molecular-weight DNA was extracted from C. pitardii leaves collected in Guizhou, China. Nanopore long reads were mapped to a reference database to enrich mitochondrial reads, which were then error-corrected and assembled de novo alongside Illumina clean reads using Unicycler. Assembly graphs were visualized in Bandage to resolve complex branching topologies.
• Annotation and Analysis: The assembled genome was annotated using homology-based alignment (BLASTN) against published Camellia and angiosperm mitogenomes, supplemented by manual curation. tRNAs were identified using tRNAscan-SE.
• Repetitive Sequence Analysis: Simple sequence repeats (SSRs), tandem repeats, and dispersed repeats were detected using MISA, Tandem Repeats Finder, and BLASTN, respectively. Long repeats were analyzed to predict potential genomic substructures via homologous recombination.
• Codon and RNA Editing Analysis: Relative synonymous codon usage (RSCU) was calculated using MEGA. Putative C-to-U RNA editing sites were predicted using the PREPACT 3.0 web server with a stringent cutoff.
• Phylogenetic Reconstruction: Phylogenetic trees were constructed using conserved protein-coding genes (PCGs) from both mitochondrial (37 genes) and chloroplast (64 genes) genomes. Analyses utilized Maximum Likelihood (IQ-TREE) and Bayesian Inference (MrBayes) methods, with outgroups selected from related taxa.

Key Results

• Genome Assembly and Annotation: The C. pitardii mitogenome was assembled into a master conformation of 895,290 bp with a GC content of 45.68%. Annotation identified 38 protein-coding genes (PCGs), 3 rRNA genes, and 30 tRNA genes. Notably, pseudogenized copies of cox2, rps19, and sdh3 were detected, and several tRNA genes (e.g., trnM-CAT) were found in multiple copies.
• Repeat Landscape and Structural Dynamics: The genome contains 260 SSRs, 33 tandem repeats, and 620 pairs of dispersed repeats. Three pairs of long repeats (LR10, LR11, LR12)—two forward and one inverted—were identified. The authors hypothesize that homologous recombination across these repeats mediates the formation of eight potential alternative genomic substructures, explaining the complex, multibranched topology observed in the assembly graph.
• Codon Usage Bias: Analysis of the 38 PCGs revealed a distinct codon usage bias. Of the 28 highly preferred codons (RSCU > 1), 92.86% end in A or U, reflecting the overall AT-rich nature of the genome.
• RNA Editing: 667 putative RNA editing sites were predicted across the PCGs, involving 14 types of amino acid substitutions. The most frequent substitution was Serine to Leucine (S→L). Approximately 46.48% of editing events were predicted to convert hydrophilic residues to hydrophobic ones.
• Phylogenetic Position: Both mitochondrial and chloroplast phylogenies confirmed the monophyly of the genus Camellia. However, topological incongruence was observed between the two organellar trees regarding the placement of specific clades. For instance, C. pitardii grouped with C. drupifera and C. oleifera in the mitochondrial tree, whereas the chloroplast tree placed it in a different subclade with C. chekiangoleosa and C. gigantocarpa. The Bayesian posterior probability for certain mitochondrial clades was low, indicating poor resolution.

Significance and Claims
The study claims to provide the first systematically characterized mitochondrial genome for C. pitardii, filling a critical gap in the organellar genomic resources for the genus Camellia.

• Baseline Data: The assembled sequence and annotation serve as essential baseline data for comparative organellar genomics, phylogenetic reconstruction, and germplasm utilization within the genus.
• Structural Insight: By integrating long-read sequencing, the study resolves the complex structural architecture of the C. pitardii mitogenome, moving beyond a single linear representation to propose a dynamic model of eight potential substructures mediated by long repeats.
• Evolutionary Complexity: The observed discordance between mitochondrial and chloroplast phylogenies highlights a complex evolutionary history for Camellia organelles, potentially driven by introgressive hybridization, organellar capture, or incomplete lineage sorting. The authors caution that definitive conclusions regarding the monophyly of Camellia sect. Camellia require broader taxon sampling and nuclear genomic data.
• Functional Predictions: The identification of extensive RNA editing sites and codon usage patterns provides a bioinformatics foundation for future studies on post-transcriptional modifications and gene regulation in this species.

The authors maintain a modest stance, emphasizing that the predicted substructures and RNA editing sites require in vivo validation through high-depth sequencing, transcriptomics, or experimental assays to confirm their stoichiometry and biological reality.