Keywords

12.1 Introduction

Brassica juncea, commonly known as Indian mustard, is an allopolyploid (AABB) formed by the cross between B. rapa (AA) and B. nigra (BB). This B. juncea (2n = 36) allopolyploidy was discovered by Prakash and Chopra in 1991 by artificial resynthesizing (Prakash and Chopra 1991). Cytological investigations using flow cytometry revealed the genome size to be 922 Mb (Yang et al. 2016). Previously, the B. juncea genome size was believed to be 1068 Mb, of which, 529 Mb and 632 Mb has been acquired from B. rapa and B. nigra, respectively. The genome of Brassica spp. has been used to examine gene function divergence and genome evolution as a result of polyploidy, widespread duplication, and hybridization. These investigations and characterization are performed under the title of “Multinational Brassica Genome Project (MBGP)”, established in 2002 (https://www.brassica.info/home/about_mbgp.html); the project’s main goal was to distribute and organize Brassica genomic resources for the benefit of the global Brassica research community. This helped greatly to study the qualitative and quantitative information of the diversity within Brassica spp. It allowed to generate a connection between phenotypic character and genomic loci responsible and to characterize the genes, and detail the substantial numbers of functional proteins within the organism. The database powered by these services has also enabled the identification and characterization of molecular genetic markers and the construction of high-resolution genetic maps. From this, gathering information of plant evolution complexity and its translation to the model plant has be come easy.

Genomic sequencing only provides information of one allelic form of each gene, but to find all variations in the heritable phenotype, three approaches are essential; gene expression information, comparison with known gene sequences, and genetic association studies. Further, functional genomics provides insights of the potential role of a gene by analyzing the pattern of gene expression. Since few genes are highly conserved, their core function can be predicted by comparing with annotated genes or available databases. Identifying the genetic markers can make it easy to pick certain alleles in the genome which are associated with important function. Genome sequence becomes applicable for crop improvement through marker-assisted breeding. Marker-assisted selection is likely to connect with a novel agronomic characteristic when an alternative allele of a gene is sequenced across different genotypes. These findings permit the identification of closely related genes with improved quality and quantities which target the improvement of biological pathways. The sequence of the B. juncea genome will identify orthologous genes, aim for germplasm enhancement. Table 12.1 provides with the basic understanding of the difference in the diploid and amphiploid genetic makeup of the Brassica species. The model plant Arabidopsis has high structural similarity with Brassica genes especially in the coding regions (~85% nucleotide conservation) (Cavell et al. 1998). Therefore, genome sequencing will serve as the foundation for translational genomics of this crop based on the functional insights of genes obtained from Arabidopsis. Here, we will contextualize the structural and functional details of B. juncea genome sequencing initiatives using various next-generation sequencing platforms.

Table 12.1 Difference in the diploid and amphiploid genetic makeup of the Brassica species
Full size table

12.2 Technological Advancements in Plant Genome Sequencing

There are several sequencing technologies that can be applied for genome sequencing in plants. These are primarily divided into two approaches, one of which is whole genome shotgun sequencing. In the shotgun sequencing approach, the complete genome is broken down into fragments ranging between 5 and 20 kb. Sequencing each fragment can be done by any suitable sequencing method and the sequence assembly can be done with computational tools to construct the whole genome sequence. Another approach is the application of bacterial artificial chromosomes (BACs) alongside genetic mapping, fingerprinting, and end sequencing. Till date, no technology exists that can read DNA from one end to the other of even moderately sized chromosome.

The initial microbial genome sequencing was primarily conducted by whole genome shotgun technology. In this approach genomic DNA is sheared to smaller insert libraries of 0.4–1.2 kb and 5–20 kb library. The 0.4–1.2 kb library is used for sequencing and the 5–20 kb library acted as template in assembly. This technology could not meet the demands of eukaryotic genome sequencing, which contains long range repeats. Several technologies such as BAC, yeast artificial chromosome (YAC), plasmic-based conventional clones with large-insert (PBC) and bacteriophage P1-derived artificial chromosome (PAC) were developed by several groups. Though the linear YAC vectors, with yeast centromere and arms, can carry inserts of 1000 kb, the limitations such as instability of insert, chimerism, less amenable in extraction and purification alongside similarly sized yeast chromosomes, led researchers to look for alternative technologies too (Anderson 1993). Meanwhile, in the early 1990s, the BAC library technology emerged as backbone technology to sequence the large complex eukaryotic genomes, while the microbial genome sequencing was dominated with whole genome shotgun technology (Zhang and Wu 2001). The BACs are vector derivatives of E. coli F factor with low copy number and autonomous replicative capability (Shizuya et al. 1992), and can host inserts of 120–200 kb efficiently, and even of 300 kb (Luo and Wing 2003). With larger insert sizes, the BAC-based sequencing gave an advantage during assembly over shotgun sequencing. The first BAC library for plants was developed for Sorghum bicolor (Woo et al. 1994), followed with rice (Tao et al. 1994), Arabidopsis (Choi et al. 1995), and other plants. Physical map of the candidate genomes was constructed using the corresponding species’ library, which in turn were used to assist whole genome sequence assembly.

The chromosome mapping, often termed as physical mapping, is the sequential arrangement of several overlapping contigs into putative chromosome, on which DNA markers such as genes of agronomic traits, repeats, transposons, serve as landmarks. This gives the landscape of DNA markers with estimated distance, across the assembled chromosome. Genome assembly is the process of arranging nucleotide sequences into the correct manner and represent model for the actual genome. Physical mapping of chromosomes/genome is required to precisely place the assembled scaffolds at the right place of the genome, thus increasing the contiguity precision of genome construction. For instance, the 135 Mb Arabidopsis genome is comprised of five chromosomes. Hauge et al. (1991) utilized cosmid library of 17,000 cosmid clones to construct the physical map of Arabidopsis thaliana genome. These 17,000 cosmid clones were assembled into 750 contigs, spanning 90–95% of the entire Arabidopsis genome, however, lacking the landscape and DNA marker information. Later Schmidt et al. (1995) utilized YAC clones to construct the physical map of Arabidopsis chromosome 4 with much more resolution. This study resulted in the construction of four contigs with information about the distribution of 112 DNA markers, 20 unmapped genes, transposable elements and other random genomic DNA fragments, and covered 90–95% of 21 Mb chromosome 4 of Arabidopsis (Schmidt et al. 1995). Subsequently the physical maps of chromosome 5 (Schmidt et al. 1997; Kotani et al. 1997), chromosome 2 (Zachgo et al. 1996), and chromosome 3 (Camilleri et al. 1998) were also constructed using available YAC libraries. Following this, a complete A. thaliana physical map with BAC was constructed (Mozo et al. 1999). This model genome was further enriched with mutant information and gene expression data for functional annotation which subsequently became the cornerstone for Brassica genome annotation.

12.3 Sequencing Platforms and Data Generation for Brassica Genome

Multiple platforms were employed in B. juncea genome projects for sequencing data generation. Yang et al. (2016) used the Illumina HiSeq 2000, Illumina HiSeq 2500 platforms for 13 paired-end (PE) and mate-paired (MP) Illumina libraries (175.8 ×) and PacBio RSII sequencing platform for 1 single-molecule reads library (12.03 ×), and 222 × of BioNano genome mapping data for B. juncea var. Tumida  T. genome sequencing (Yang et al. 2016). Following this, Paritosh et al.(2021) sequenced Brassica juncea var. Varuna with dual library system, employing Illumina HiSeq 1000 sequencer platform for llumina PE libraries (200–350 bp) and PacBio RSII platform for PacBio long range (30–50 kb) DNA (PB) libraries. The combination of PE and PB gave higher resolution and better contiguity (Paritosh et al. 2021). The progenitor species with A genome, the B. rapa was sequenced with Illumina GA II platform with shotgun sequencing libraries ranging three different insert sizes as long (∼2 kb, 5 kb and 10 kb), medium (∼500 bp), and short (∼200 bp), further supported with Sanger sequencing to fill the gaps in assembly (Wang et al. 2011). This reference genome was reannonated with sequencing data from Illumina 55 Gb (114 ×) PE, 8.7 Gb (18 ×) MP libraries with Illumina Hiseq2500 platform and 6.5 Gb (13.4 ×) reads of single-molecule sequencing PacBio PB libraries data with an average length of 12 Kb with PacBio Sequel II platform (Cai et al. 2017). The genome information is further improved with 19.40 Gb data generated obtained from PacBio library-based PacBio Sequel II platform (Zhang et al. 2018). Zhang et al. also utilized Illumina HiSeq 4000 platform, for Hi-C mapping with 2 × 125 bp reads, for improved chromosome. For the sequencing of the B genome progenitor B. nigra, Yang et al. (2016) generated 10 PE and MP Illumina libraries (95.99 ×) for B. nigra doubled haploid line (YZ12151). Perumal et al. (2020) used C2 and Ni100 cultivars. A total of 82 Gb (137 ×) for C2 cultivar and 115 Gb (192 ×) for Ni100 cultivar were generated with Illumina HiSeq 2500 and Oxford NanoporeMinION and GridION platforms were used for further assembly. Paritosh et al. (2020, 2021) constructed genome sequence for B. nigra var. Sangam with three different sized, PE libraries (200–350, 300–450, and 400–550 bp) and MP libraries (2–3, 4–6, and 10 kb) and Illumina HiSeq 1000 and Illumina MiSeq platforms.

12.4 k-mer Analysis and Error Correction

k-mers are short sequence substrings obtained from the insert reads. These are used to estimate the genome size, by dividing the total number of the k-mers with the depth of the major peaks. k-mers are also used to remove the errors during sequence assembly. A multiple appearance of specific k-mer in the read data reflects errorless sequencing, and error sequences were omitted with lower frequencies. K-mers of 16nt (Wang et al. 2011), 17nt (Yang et al. 2016; Perumal et al. 2020), 21nt (Paritosh et al. 2020), 24nt (Perumal et al. 2020) for short reads and even k-mers of 91 bp (Cai et al. 2017), and 100 bp (Paritosh et al. 2021) were used to cover the complex region with PacBio reads.

12.5 Tools and Technologies Used in B. Juncea and Its Progenitor Genomes Assembly

The de novo assembly with short reads of B. juncea var. Tumida reference genome was done by ALLPATHS-LG and the gaps were filled with Pacbio RS II reads (Yang et al. 2016). The Irys View RefAligner utility was utilized to draft assemble the genomic scaffolds, and the scaffolds were anchored to the chromosomal map obtained from BioNano data (Yang et al. 2016). The availability of B. juncea var. Tumida as reference made the B. juncea var. Varuna genome assembly easier. The assembly was carried out using Canu assembler (V1.4) with PacBio sequence reads and the resolution of the sequences was cross examined by mapping with Illumina short read sequences using BWA-MEM. This allowed the B. juncea var. Varuna genome to be more precise (Paritosh et al. 2021). The B. rapa reference genome (Wang et al. 2011) was de novo assembled using SOAPdenovo and was validated using NUCmer. The constructed chromosome assemblies were porous with gaps, in particular with the repetitive regions (Wang et al. 2011). The reference genome was reassembled to contigs by Cai et al. (2017) with new reads, as well as with the reads of Wang et al. (2011) using SOAPDenovo2. These assembled contigs were then linked into scaffolds using SSPACE (Cai et al. 2017). These scaffolds were then examined with BAC library sequences by BLAST as well as with Illumina short reads and anchored into super-scaffolds using GapCloser. The PacBio sequences were also used for gap filling using PBjelly_V15.2.20. The reassembled genome was then further validated with CEGMA analysis (Cai et al. 2017). Zhang et al. (2018) again constructed B. rapa genome by de novo assembly aided with Hi-C mapping data for better assembly. Using Canu (v1.5), the PacBio reads were assembled and the resolution was enriched by aligning with Illumina short reads using BWA (v0.7.15) and the assembly was then validated with Pilon (v1.22). The aligned and assembled scaffolds were further anchored onto chromosomes using BioNano optical mapping data with Lachesis. The Lachesis was also used to construct the spatial genome to visualize proximal interacting sequences with Hi-C data and plotting with ggplot2 package (Zhang et al. 2018). For B. nigra reference genome de novo assembly, the Illumina reads were assembled using ALLPATHS-LG and the gaps were then filled using GapCloser. The generated scaffolds were then aligned by BLAST with publicly available BAC library sequences of B. nigra. These scaffolds were then assembled into pseudo-chromosomes using available genetic map. The assembled genome sequence was further validated with CEGMA v.2.3 (Yang et al. 2016). Perumal et al. (2020) independently sequenced genomes of two other varieties of B. nigra. The generated Nanopore reads were assembled using CANU 1.6 and crosschecked with three other assemblers; SMARTDenovo, wtdbg, Miniasm (Perumal et al. 2020). The assembled contigs were iterated with high-quality Illumina reads using PILON. These iterated contigs were then scaffolded using Dovetail’s HiRise pipeline and were anchored into pseudo-molecules using available genetic maps as references (Perumal et al. 2020). Paritosh et al. (2020, 2021) sequenced genome of B. nigra var. Sangam. The genome was assembled with Nanopore reads using CANU 1.6 and the contigs were mapped with high-quality Illumina reads using BWA-MEM (v0.7.12) for higher resolution and iteration with PILON (v1.23) (Paritosh et al. 2020). In parallel, the group also assembled the genome of B. nigra var. Sangam with Illumina reads using MaSuRcA, and the BWA-MEM was used to align and position these contigs with PacBio reads. The scaffolding was done with SSPACE-LongRead.pl script and SSPACE-STANDARD-3.0.pl script (Paritosh et al. 2021). The assembled contigs were anchored onto chromosomes using BioNano optical map data and the generated genetic data was polished with PILON (Paritosh et al. 2021).

12.6 Sequencing of Progenitor Diploid Genomes

Population analysis-based genetic linkage maps were earlier constructed for B. oleracea (Slocum et al. 1990), B. rapa (Song et al. 1991; Chyi et al. 1992), B. napus (Ferreira et al. 1994), B. nigra (Lagercrantz and Lydiate 1995), and other Brassica species as well (Thormann et al. 1994). B. rapa (AA) and B. nigra (BB) are the genomic progenitors of B. juncea (AABB). RNA-seq based single nucleotide polymorphism (SNP) marker mapping revealed collinearity between B. nigra genome and B. juncea B genome, and the B. juncea A genome with genome of B. rapa (Paritosh et al. 2014). The estimations by several groups put B. rapa (AA) genome to be 450–550 Mb, and confirmed to be of 442.9 MB with BioNano data (Zhang et al. 2018). The B. rapa genome underwent genome triplication, and localized rearrangements such as deletions, insertions, inversions, and substitutions (Hong et al. 2008). Wu et al. (2000) constructed binary vector-based library for B. rapa for cytoplasmic male sterility locus studies. Subsequently, a genome-wide BAC library was constructed for B. rapa jointly by multiple groups (Mun et al. 2008; Hong et al. 2006), and 12,017 BAC-end sequences were analyzed for TEs, SSRs, centromeric satellite repeats and genes, that could also act as markers. The coverage of genome with TEs (14%), 43,000 genes (16.8%) and 1392 different SSRs, with estimated 110,000 SSRs (~one SSR/~4.8 kb), were mapped with this and BAC-end sequence analyses (Hong et al. 2006, 2008). Meanwhile, multiple groups developed different B. rapa genetic linkage maps marked with different genetic markers (Kim et al. 2006; Choi et al. 2007). Wang et al. (2011) genome sequenced B. rapa by whole genome shotgun sequencing technology with short (∼200 bp), medium (∼500 bp), and long (∼2, 5 and 10 kb) Illumina GA II libraries, and constructed an annotated draft genome. The paired short read sequences upon assembly covered 222 Mb deriving unique scaffolds. The long-read MP sequences (≥2 kb) were not initially used to avoid chimeric reads and incorrect sequences that are common to such libraries. The obtained unique scaffolds were then matched with paired-end sequences to assemble the unique contigs and was matched and corrected with the available BAC libraries (Mun et al. 2008; Hong et al. 2006), and previously assembled chromosome sequences by BAC technology (Mun et al. 2010). Thus, the sequence assembly covered 283.8 Mb, with an estimated cover of >98% of the gene space (Wang et al. 2011). The gaps in draft genome were predominantly with repeat and centromeric regions. Search analyses with 214,425 expressed sequence tags (ESTs) and 52,712 unigenes from NCBI and other sources identified 41,174 putative coding genes. Analysis of B. rapa gene family revealed 1003 (5.9%) families out of 16,917 to be lineage specific (Wang et al. 2011), while the rest shared features closely with related plants such as Arabidopsis, Carica papaya, and Vitis vinifera. The genome data is further upgraded with more short reads and long PacBio reads with reannotation (Cai et al. 2017). The gaps in the previous versions are predominantly in the repeat regions and centromeric regions are due to usage of short read sequences. Zhang et al. (2018) overcome this drawback and further enriched the B. rapa genome sequence and developed a reference genome employing third generation sequencing technologies such as optical mapping, Hi-C, and single-molecule sequencing. This latest study with integrated approach put the B. rapa genome size to 442.9 Mb, shrinking the previous estimations of 450–550 Mb (Zhang et al. 2018). Optical mapping of B. rapa genome was performed alongside PacBio long-read sequencing with a focus on centromeric and repeat regions to fill the gaps from previous genome versions. Hi-C libraries were also generated to visualize the proximal and physically interacting DNA elements. The resequencing and reannotation provide the B. rapa genome with 45,985 protein-coding genes with 45,411 genes (98.75%) annotated on chromosomes, leaving behind 574 genes (1.25%) on scaffolds. The B. nigra (BB) genome has an estimated size of 570–607.8 Mb, while studies by Paritosh et al. (2020) suggested the B. nigra genome to 522 Mb. The B. nigra underwent whole genome duplication, and subsequent rearrangements, making it a complex genome for sequencing and annotation. Yang et al. (2016) genome sequenced the B. nigra genome, covering 396.9 Mb of the genome, and annotated 49,826 genes. Perumal et al. (2020) assembled B. nigra genome with whole genome shotgun Illumina PE (300–700 bp insert) sequencing, Roche/454and Illumina MP libraries (3–45 kb insert) and long reads from Nanopore sequencing. The short reads were assembled with the help of long reads and the genome was annotated with RNA seq data (Perumal et al. 2020). Perumal et al. (2020) predicted 59,877 and 67,030 genes in two different genotypes, of which 55,022 (92.0%) and 59,780 (89.2%) genes were annotated with Uniprot database. Following this, Paritosh et al. (2020) utilized Optical Mapping and Nanopore long-read sequencing to sequence and assemble the B. nigra genome with higher contiguity. The study identified 57,249 putative genes of which 42,444 genes were matched with transcriptome data. Table 12.2 enlists the sequencing and annotations efforts of B. juncea progenitors.

Table 12.2 Sequencing and annotation of B. juncea progenitors
Full size table

12.7 Genome Sequencing of Cultivated B. Juncea

Sequencing of B. juncea has served to resolve the complicated allopolyploid genome using the tools and technologies described earlier. B. juncea carries an allopolyploid genome (AABB). The draft genome sequence for B. juncea was constructed (Yang et al. 2016) with multiple sequencing technologies primarily with PE and MP Illumina libraries, and certain portions with long-read Bionano and PacBio sequencing. Flow cytometric analysis suggested the genome size as 922 Mb. The first draft genome of B. juncea var. Tumida was done using shotgun reads by de novo assembly, single-molecule long reads (PacBio sequencing), genetic mapping and genomic (optical) mapping (BioNano sequencing). The assembled B. Juncea var. Tumida genome resolved the complication of allopolyploid genome, and it’s constructed by de novo assembly. Subsequently, it was also demonstrated that the homeolog expression dominance in B. juncea to explore the A and B subgenomes for their transcriptional behavior. The draft genome covered 85% (784 Mb) of the genome and annotated with 80,050 genes, 21lncRNAs, 2,638 tRNAs, 511 rRNAs, 3,725 small RNAs, 1,402 microRNAs, and 15,418 small nuclear RNAs. The B. juncea genome appears to have lost identified 562 and 545 genes in A and B subgenomes (Yang et al. 2016). Genome-wide RNA-seq analysis of the homoeologous genes from different developmental stage, tissues and two newly resynthesized B. juncea revealed, in all the samples, on average 16.2% of genes displayed homeolog expression dominance, whereas only 8.2% of genes showed expression dominance toward to BjuB over BjuA excluding B. juncea resynthesized lines. On the basis of this observation, its inferred that no significant global dominance for the subgenomes further strengthening recent polyplodization of B. juncea (Yang et al. 2016). The sequencing and assembly of the B. juncea genome facilitate the agricultural trait improvements in this important crop.

The constructed B. juncea genome was further enriched for improved contiguity using long read sequencing technologies (Paritosh et al. 2020). The chromosome-scale genome assembly of B juncea generates a complete architecture of the A and B genome. They subjected B. juncea var. Varuna to SMRT sequencing on the PacBio RSII platform. This study provided marked improvement in B genome assembly of Varuna variety, unlike earlier sequenced Tumida variety, exposing extensive gene block fragmentations and gene block associations than reported earlier. The quality of genome sequence was improved to the extent that, the longest contig covered 35.8 Mb and the largest scaffold covered 72.1 Mb. RNA-seq guided gene annotations revealed 101 959 putative genes, in A (46,381) and B (55,578) subgenomes, of which 93% genes were validated with uniprot entries (Paritosh et al. 2021). Transposable elements covered 385 Mb (45.8% of B. juncea genome) of the genome, with higher representation in B subgenome covering 259 Mb of 51% of the subgenome than the A subgenome, that covered 113 Mb of 33.9% of A subgenome (Paritosh et al. 2021). Further, we have also discussed organellar genome sequencing in details. The details of genome sequencing efforts in B. juncea are summarized in Table 12.3.

Table 12.3 Publicly available B. juncea genome sequences, assembly and associated initiatives useful for genome annotation
Full size table

12.8 Genome Annotation

Genome annotation is a method of genome study to identify functional elements along the sequence. It is the description of individual genes and their proteins or predicting the meaning of the genome sequence. Such prediction can be structural annotation (including location of coding regions, open reading frames (ORFs), and their corresponding regulatory motifs, functional annotation (integration of additional data such as spatio-temporal expression profile at transcriptomic and proteomic level, interaction network, physiological and biochemical functions), and essential components (CDS, mRNAs, pseudogenes, promoters, and poly-A signals)). Annotation of B. juncea var. Varuna genome was carried out for transposable elements, centromeres, and genes. For transposable element de-novo prediction approach, centromeric region is identified by correlation plot and RNA-seq analysis is performed for gene annotation. There are 1590 consensus repeats generated when merged with Arabidopsis thaliana transposable element database, repeats were classified into retrotransposons, DNA transposons, and other repeats. Comparing to the A chromosome [(113 Mb (~33.9%)], the B chromosome [~259 Mb (~51%)] has more repeats. A genome of B. juncea has a centromeric region already identified in B. rapa (A genome) and B. oleracea (C genome), i.e., CentBr1 and CentBr2 but absent in the B genome centromeric region. However, seven new B genome-specific repeats along with three new A genome-specific centromeric repeats were identified from B. juncea. In gene annotation, a total of 40,208 full-length high-quality sequences were obtained and a total of 105,354 genes were predicted of which 48,270 in the A genome and 57,084 in the B genome (validated by the non-redundant proteins in the UniProt plant database). Other studies have found a total of 82,008 genes–38,232 (~79%), 43,776 (~76%) within the A and B genome, respectively (Paritosh et al. 2014; Yang et al. 2016). B. juncea var. Varuna, B genome has shown high similarity with B. nigra genome. Orthologous study has also identified a total of 19,404 orthologous groups of which 39,383 genes of A had orthologs in B, and 42,727 genes of B had orthologs in the A genome. Similarly, with using non-redundant nucleotide and NCBI protein sequences, SWISS-PROT, Gene ontology (GO), Cluster of Orthologous Groups (COG), and Kyoto Encyclopedia of Genes and Genomes (KEGG) databases, the Tumida variety was also annotated for 80,050 protein-coding genes. Additionally, Yang et al.(2016) identified 511 rRNAs, 2,638 tRNAs,3,725 small RNAs, 21 long noncoding RNAs, 15,418 small nuclear RNAs and 1402 microRNAs, from the B. juncea genome and even extracted syntenic ortholog gene pairs (28,228 and 28,917) from the B. juncea subgenomes. Gene loss has been observed during speciation, a total of 562 (A subgenome) and 545 (B subgenome) genes were considered as lost, referring to their common ancestral genomes. Likewise, many structural and functional annotations have been performed to investigate the structure, function, and regulation of the genome which are enlisted in Table 12.4.

Table 12.4 Structural and functional annotations performed in B. juncea
Full size table

12.9 Organellar Genome Sequencing

12.9.1 Mitochondrial Genome Insights

Mitochondrial genomes in angiosperms are more complex compared to other organisms (Chang et al. 2011). Analyses of different angiosperm mitochondrial genome sequences have revealed several common characteristics; however, a number of diverse mitotypes have evolved within each genus/species. Sequencing and comparative investigations of six Brassica mitotypes indicates a mechanism for mitochondrial genome development in Brassica, which includes certain events of evolution like mutation, genome compaction, duplication, and rearrangements (Chang et al. 2011). Further, the analysis of the evolutionary relatedness between Brassica mitotypes provides information of mitochondrial genome of B. juncea. The size of whole single circular mitochondrial genome of B. juncea is 219,766 bp [GenBank: JF920288], which is reported to be similar to that of B. rapa, i.e., 219,747 bp. Whereas, the mitochondrial genome sizes of B. carinata and B. oleracea are greater than both B. juncea and B. rapa, i.e., 232,241 bp and 360,271 bp, respectively. The report suggests that the B. juncea mitotype is an evolutionary outcome of an inherited mitotype of (B. rapa) cam with certain modifications (Chang et al. 2011). The percentage of the total functional genes in the mitochondrial genome of B. juncea are almost the same compared to B. rapa, B. oleracea and B. napus mitotypes, expect B. carinata, where the similarity is 27.98% (Chang et al. 2011).

One of the main aspects of mitochondria genomes is producing cytoplasmic male sterility (CMS), which is essential to explore the heterosis in crop plants and is routinely utilized as a model to investigate the interactions between nuclear and cytoplasmic. These interactions are either spontaneous or formed through interspecific crossing, which results into an alloplasmic condition/cytoplasmic substitution (Gaikwad et al. 2006). CMS characterized, particularly by non-Mendelian inheritance and the suppression of the production of viable pollen. The CMS could result due to mutations, recombination, or rearrangements in the genome of mitochondria. Knowledge of the B. juncea mitochondrial genome is crucial to develop superior phenotypes. In this quest, numerous attempts are being undertaken to sequence and assemble the mitochondrial genomes of B. juncea lines. For instance, a study performed on five alloplasmic B. juncea lines revealed that the mitochondrial genomes from 221 to 256 kb (Wu et al. 2019), which is somewhat greater compared to the typical B. juncea size reported earlier (Chang et al. 2011). Another comparative study between the mitochondrial genomes of the hau CMS line and its iso-nuclear maintainer line in B. juncea showed a difference in the genome sizes and GC content (Heng et al. 2014). Mitochondrial genome of B. junceahau CMS is reported to be of 247,903 bp in size with a GC content of 45.08%, whereas that of another normal line (J163-4) and a maintainer line are 219,863 bp in size with a GC content of 45.23%. Further, the mitochondrial genome has numerous genes, of which, 35 are protein encoding, 25 are tRNA, 3 are rRNAs, and 29 are ORFs of unknown function. Whereas, genes in maintainer line are 36 for protein encoding, 22 for tRNA, 3 for rRNAs, and 31 are unidentified ORFs. In addition, sub-stoichiometrical coexistence of distinct mitotypes is confirmed in hau CMS lines as well as its maintainer lines in B. juncea (Heng et al. 2014). Further, it was demonstrated that a cytotoxic protein ORF288 associated with male sterility in B. juncea causes aborted pollen development (Jing et al. 2012). The toxicity generating region of ORF288 is reported to be located near the N-terminus, which repressed the growth when heterologously expressed in E. coli (Heng et al. 2018). However, heterologous expression of ORF288 portions indicates that the region which induces CMS is present between amino acids 73 and 288, whose heterologous expression did not inhibit E. coli growth. It is reported that the transcript levels of orf288 are associated with altered nuclear gene expression and the hau CMS system. Apart from orf288, it also reported that orf220 gene causes male sterility in B. juncea (Yang et al. 2010). Both positive and negative correlations have been reported between the occurrence of CMS and CMS-associated orfs, when CMS-associated orfs were targeted and expressed in mitochondria. Certain orfs can cause male- or semi-sterility, while some do not. Understanding of nuclear-mitochondrial compatibility is crucial as it can result in differential expression of mitochondrial genes, e.g., mitochondrial aptα gene (Gaikwad et al. 2006). Furthermore, it is essential to identify the molecular basis of mitochondrial recombination in the male sterile cytoplasmic hybrids to understand the underlying mechanism as well as the environmental factors affecting fertility reversion. For instance, analysis of floral bud transcriptome under both pollination and non-pollination state showed a variability in the expression of Muts HOMOLOG1 (MSH1) (a nuclear gene which regulates illegitimate recombination in plant mitochondria) in response to different sugars, which shows that physiological changes are involved in the pollination signaling and fertility reversion in CMS plants (Zhao et al. 2021). This is indicative that the gynodioecy is a reproductive plan of action that might incorporate ecologically responsive genes such as MSH1 as a switch for fertility-sterility transition under reproductive isolation. It was found that the mitochondrial genome in revertant lines of B. juncea CMS cytoplasm undergoes substoichiometric shift to suppress orf220 copy number, whereas MSH1-RNAi with increased orf220 copy number are male sterile (Zhao et al. 2016). This provides a valuable insight into substoichiometric shift in CMS induction, fertility reversion as well as interplay of MSH1.

12.9.2 Chloroplast Genome Sequencing

Chloroplast is a vital organelle in plants, which harbors genetic and enzymatic resource required for photosynthesis, as well as various important biosynthesis pathways such as pigments, fatty acids, vitamins, and amino acids. Understanding of the chloroplast genome is therefore essential to enhance the qualitative and quantitative traits of B. juncea. As chloroplast genome engineering of B. jucnea through chloroplast can increase the oil content, improve oil quality, as well as make it resistant to most of the abiotic and biotic factors. Prabhudas et al. (2016) are the first to sequence the complete chloroplast genome of B. juncea (Indian mustard). The sequencing carried out on Illumina Hiseq 2500 platform generated 100 bp paired raw reads, which were qualitatively trimmed followed by read quality assessment.

The chloroplast genome size is reported to be 153,483 bp, with 36.36% of GC content (Prabhudas et al. 2016). The assembled chloroplast genome exhibited a quadripartite structure consisting of an 83,286 bp large single copy region and an 17,775 bp single copy region, which are separated by a pair of IRa and IRb (inverted repeats a and b) of 26,211 bp each. The GC content was 34.12, 29.20, and 42.34% for large single copy region, single copy region, and inverted repeats, respectively. In the entire chloroplast genome, a total number of 113 genes were annotated, of which, 79 genes are protein encoding, 30 are tRNA, and 4 are rRNA. Total 15 genes including 9 protein-encoding genes and 6 tRNA genes had either a single of couple of introns. Gene duplications in inverted repeat regions were extended to 4 rRNA, 7 tRNA, as well as 6 protein-encoding genes.

B. juncea share the chloroplast genome with one of the hybridization donor B. rapa, which fits the U model (Li et al. 2017). The phylogenetic analysis indicated that the branch length of B. juncea chloroplast genome is close to B. oleracea, B. napus, and B. rapa more distant from B. nigra and B. carinata, suggesting a divergence from the two Brassica’s.

12.10 Conclusion

Rapid development in next-generation sequencing technologies has undoubtedly accelerated the sequencing programs of B. juncea genomes along with its progenitors to generate metadata at less cost and time. The B. juncea bio-projects have sequenced, assembled, analyzed and have annotated the genome to highlight the impact of recombination between progenitor-genomes of Brassica spp. Apart from providing fundamental knowledge about the evolution of B. juncea, the bio-projects also helped in the identification of candidate genes involved in various physiological and biochemical processes. Importantly, the information can be utilized form trait mapping to targeted tinkering of the genetic makeup of B. juncea and to further improve resistance to biotic and abiotic stresses, enhance climate resilience,  increase yield as well as quality and nutritional traits. The availability of the genomic data will further advance the knowledge of cultivated B. juncea lines and help to produce better tool kits for molecular breeding and crop improvement. Considering the level of diversity and wealth of the species in the Brassica genus, there is a lot yet to explore about  B. juncea to clarify the relationship between genotypes/phenotypes. Similarly, progress in proteome and transcriptome research has the capacity to reveal the structural and functional role of genes in B. juncea, altered in response to physiological and environmental processes and pathogen attacks. Overall, the information of nuclear and organellar genomes, as well as the increasing amount of “omics” resources will certainly contribute toward the effective use of B. juncea.