Introduction

Pigs (Sus scrofa) are one of the most ancient domesticated, socioeconomically valued and widely distributed livestock species across the world1. The process of pig domestication occurred independently in various regions from its wild ancestors, with evidence suggesting occurrences in western Asia around 8500 BC2,3, in China around 6500 BC4, and in Southeast Asia and Europe approximately 9000 years ago5,6. India is recognised as one of the centres for the domestication of pigs and domestic pigs potentially originating from Indian wild boars separate from European and other Asiatic lineages7. The genomic analyses revealed distinct mitochondrial haplotypes in Indian pig populations that were present in wild boar populations of India, but not in pigs from Europe and the East, indicating a localized event for domestication6.

Pig farming holds significant importance in the livelihood of rural tribal communities1. India possesses a rich diversity of pig genetic resources, with significant variations among populations. Fourteen indigenous pig breeds, documented in the country’s breed database, contribute significantly to the socioeconomic upliftment of rural poor pig farmers. Among these breeds, the Ghoongroo (GH), earlier known as Ghungroo, stands out as one of the most prolific pig breeds, primarily found in West Bengal and Assam which has the potential to be used in various breeding programmes8. This is the first registered pig breed of India exhibits distinctive features such as a black coat, a characteristic bulldog-like face, a cylindrical body shape, and large, drooping ears (Fig. 1a,b)9,10. Out of total 9.06 million pig population of the India, GH pig are 208,751 in number with 2.9% of total indigenous pigs of India. The highest population of GH is found in West Bengal (117.4 thousand) followed by Assam (82.9 thousand), and Uttar Pradesh11. The northeastern region of India, particularly Assam, emerges as a key hub for pig production, with GH and Doom being the predominant indigenous breeds traditionally raised in low-input backyard farming systems. These indigenous breeds possess inherent traits such as early sexual maturity, adaptability to harsh climate and management conditions and requirement of low input, disease resistance, strong maternal instincts, and desirable meat quality makes them the best enterprise for the weaker sections of society and the progressive farmers as well12. Being most prolific pig breed of India, GH pigs are very popular among the farmers traditionally rear pigs in low input backyard farming system13. However, to enhance growth and reproductive performance, breeds like Hampshire and Duroc have been introduced and crossbred with GH, resulting in crossbred pig varieties like Rani and Asha. Notably, these crossbred varieties retain the maternal genetic heritage of GH14.

Fig. 1
Fig. 1
Full size image

Images of adult (a) Ghoongroo Boar (b) Ghoongroo Sow (c) Asha Crossbred (d) Rani crossbred pigs.

Mammalian mitochondrial DNA genome (mtDNA) is a double-stranded molecule, composed of a H (heavy) strand and a L (light) strand and is approximately 16.5 kb in size that varies with the species viz. cattle 16.34 Kb, goat 16.64 Kb; sheep 16.61 Kb; buffalo 16.36 Kb and in pig 16.69 Kb15,16,17,18,19. MtDNA encodes crucial proteins of the electron transport chain. The location of genes varies in different species but most genes are located on the H-strand and only one or two genes are located on the L-strand. The mtDNA also has 22 tRNAs and 2 rRNAs that are involved in mtDNA transcript production and processing. Its maternal inheritance pattern and faster base substitution evolutionary rate allow for the investigation of evolutionary relationships within and between species20. Of the mtDNA genome, the control region i.e. D-loop was used for investigating the genetic population structure of closely related animals in restricted areas21,22.

The complete mtDNA of GH and its crossbreds, along with the assessment of matrilineal components and genetic diversity, represents a vital step towards conservation and genetic improvement. Several studies have been conducted to characterize the mitochondrial diversity in Indian domestic and wild pigs7,16,23. The maternal diversity in five Indian pig breeds (Mali, Niang Megha, Tenyi Vo, Ghoongroo, and Doom) alongside three exotic breeds (Hampshire, Duroc, and Large White Yorkshire) were identified using variations in 487 bp fragment of mitochondrial D-loop23. The study identified unique haplotypes within indigenous pig populations. Sharma et al.24 explored the maternal genetic diversity in Indian pig breeds by analysing the 464 bp fragments of mitochondrial D-loop. The study revealed 32 maternal haplotypes in which GH pigs indicated only two haplotypes with 10 variable sites. The earlier reports mainly focused on small fragments of mtDNA, however, the present study aimed to characterize the complete mtDNA of Indian GH pigs and its crossbred varieties, tracing domestication patterns based on maternal lineages. Phylogenetic analyses will shed light on relationships among GH and its crossbreds and to what extent they were affected by the modern commercial breeds (Duroc, Yorkshire and Landrace) in maternal lineage, thus providing valuable insights into the multiple matrilinear components and evolutionary history of GH pigs.

Methods and materials

Ethics statement

All blood sampling procedures were conducted with minimal invasiveness, adhering to humane endpoint guidelines to ensure the welfare of the pigs. Blood was collected only once from each pig, minimizing the stress and physiological impact on the animals. All experiments conducted in this study adhered to the guidelines set forth by the animal ethics committee of the institute, with approval no. NRCP/CPCSEA/1658/IAEC-20/2018, ensuring compliance with ethical standards for animal care and use.

Animals and sampling

This study focused on the indigenous pig breed GH and its crossbreds found in the Bengal and Assam regions of India. A total of five GH pigs were used for the characterisation of the complete mtDNA genomic sequence and 11 GH pigs were used for the characterization of the complete D-loop sequence of the mitochondrial genome. Additionally, crossbred varieties, namely Asha and Rani, with maternal components of GH, were included in the study. Seven animals from the Asha variety and six from the Rani variety were used for complete D-loop sequence analysis, while one animal each from Rani and Asha crossbreds was employed for complete mitogenome sequencing. The samples were taken from different locations in the breeding track of GH from West Bengal and Assam and care were taken in including only one sample from one location such that the sample used in the study represent the true population. However, samples of crossbreds were collected from Assam region only where the crossbreds were developed and distributed. Rani (Fig. 1d) is a crossbred pig variety developed by crossing ♀ GH and ♂ Hampshire pigs, with 50% blood inheritance of each breed14. Asha (Fig. 1c), on the other hand, is a crossbred pig variety obtained by crossing ♀ Rani with the terminal sire ♂ Duroc, thereby maintaining mitochondrial inheritance from GH pigs exclusively. Blood samples of 5 ml each were collected from the anterior vena cava using a sterile needle and BD vacutainer, and then stored at − 20 °C until DNA extraction.

DNA extraction

Genomic DNA was extracted from the blood samples using the standard phenol–chloroform method25,26. The quality of extracted genomic DNA was checked on agarose gel electrophoresis and the DNA samples without any smearing and having intact bands were used for further study. The purity and concentration of DNA were determined using a NanoDrop spectrophotometer, with samples having an A260/A280 ratio falling between 1.7 and 1.9 considered suitable for downstream applications. The DNA samples having good quality and purity were stored at − 20 °C until further use.

PCR amplification of mitochondrial genome

The primers sequences, amplification temperatures and product size along with the amplification conditions were similar to our earlier study7. The sequences of all 30 pairs of primers along with annealing temperature and product size were presented in Supplementary Table 1. Briefly, 30 pairs of overlapping primers were used for the amplification of the complete mtDNA of GH and its crossbreds. The PCR was carried out in a 25 μl reaction mixture having 2.5 μl of 10× PCR buffer (with Mg2+), 0.5 μl of forward and reverse primer each (10 pm/μl), 0.5 μl of dNTPs (10 mM), 0.2 μl Taq polymerase (1 unit), 1 μl DNA samples (50 ng/μl), and 19.8 μl of NFW. The PCR condition was conducted in a thermocycler (Applied Biosystems) involving an initial denaturation at 95 °C for 7 min, followed by 30 cycles of denaturation at 95 °C for 30 s, annealing at Ta for 30 s (Ta for different set of primers were either 58 or 58.5 or 59 °C as mentioned in supplementary table 1), and extension at 72 °C for 30 s followed by a 5 min final extension at 72 °C. The PCR products were then analyzed by 2% agarose gel electrophoresis. The PCR products showing intact specific bands and without any smearing was used for downstream works and processed for sequencing.

Sequencing of amplicon and structure analysis mitogenome

The purified PCR products were subjected to Sanger sequencing using 3500 Series Genetic Analyzers (Applied Biosystems). Sanger Sequencing not only sequences individual DNA fragments sequentially, but also guarantees full coverage of the reference genome. This is achieved through the utilization of overlapping fragments, effectively eliminating any gaps, and ensuring a comprehensive 100% coverage of the targeted sequence27. The obtained sequences from Sanger sequencing were trimmed and edited using DNAstar and Megalign software version 15.0.0 (https://www.dnastar.com/software/lasergene/megalign-pro/). The annotation of complete mtDNA sequences was finalized using MITOS2 tools in the Galaxy (https://usegalaxy.org/?tool_id=toolshed.g2.bx.psu.edu%2Frepos%2Fiuc%2Fmitos2%2Fmitos2%2F2.1.9%2Bgalaxy0&version=latest) webserver Platform28,29,30. The circular structure of the complete mitogenome was constructed using Proksee (https://proksee.ca/) Server31. The structure of the tRNA sequence identified in the complete mitogenome was predicted using the tRNAscan-SE 2.0 (https://trna.ucsc.edu/tRNAscan-SE/) web server32. The Nucleotide frequencies, G + C content, and A + T content of mitogenome were determined using EditSeq of Laser gene (DNA STAR Inc.). The skewness of protein-coding genes in the mitogenomes was calculated using the formula: GC skew = (G − C)/(G + C) and AT skew = (A − T)/(A + T)33.

Phylogenetic analysis and genetic distance analysis

Phylogenetic analysis, using the complete mtDNA sequences of pigs generated in this study, was conducted, along with sequences from Indian wild boar, Asian pig breeds (5), European Pig breeds (4) and African warthog (AWH) downloaded from NCBI GenBank for comparison. The nucleotide sequences were aligned using the MUSCLE algorithm34 of Molecular Evolutionary Genetics Analysis (MEGA) 11 (https://megasoftware.net/dload_win_beta)35. Detailed information regarding each substitution model, including their respective BIC, AICc values, Maximum Likelihood value (lnL), and the number of parameters, is provided in Supplementary Table 2. The substitution model utilized for the alignment was thoroughly assessed, and the Tamura-Nei model with gamma distribution rates among sites (TN93 + G), which yielded the lowest Bayesian information criterion (BIC) score, was selected as the most suitable for both alignment and phylogenetic analysis. This analysis involved 18 nucleotide sequences. There were a total of 18,320 positions in the final dataset. Subsequently, the aligned sequences were employed to construct a phylogenetic tree using the maximum likelihood (ML) method with 1000 bootstrap replications in MEGA 11, aimed at elucidating the matrilineal components in GH and its crossbred pig varieties.

Apart from the complete mitogenome sequence, complete D-loop sequences were also used for the phylogenetic and genetic distance analysis. For this purpose, the complete D-loop sequences of European pig breeds, which contributed to the development of crossbred varieties of GH pigs, were retrieved from the NCBI GenBank (https://www.ncbi.nlm.nih.gov/). The list of sequences used for the phylogenetic analysis is provided with their accession numbers in Supplementary Tables 3a,b. The nucleotide sequences were aligned using the MUSCLE package of MEGA 11 employing the HKY + G model as the best fit, which yielded the lowest Bayesian information criterion (BIC) score, and accounts for varying nucleotide frequencies and differing rates of transitions and transversions. Details of each substitution model including their respective BIC, AICc values, Maximum Likelihood value (lnL), R, f and the number of parameters, are provided in Supplementary Table 4. The phylogenetic analysis encompassed 28 nucleotide sequences and a total of 1330 positions in the final dataset. These aligned sequences were used to construct a phylogenetic tree via the ML method with 1000 bootstrap replications.

The phylogenetic tree constructed in MEGA was visualized using the FigTree v.1.4.4 (http://tree.bio.ed.ac.uk/software/figtree/) software36. The genetic distances among the sequences were calculated using the maximum composite likelihood (MCL) model in MEGA 1135,37. Since all pairwise distances in a distance matrix have correlations due to the phylogenetic relationships among the sequences, the sum of their log-likelihoods is a composite likelihood. The pairwise distances and the related substitution parameters could be accurately estimated by maximizing the composite likelihood38. The AWH was selected as an outgroup for phylogenetic tree analysis because it is well known to be different from Eurasian wild boars and this pig has been commonly employed in past phylogenetic investigations of pigs39,40,41. The haplotypes were identified from sequences of complete mitogenome and complete D-loop sequences using DNA Sequence Polymorphism software package (DnaSP v.6) (http://www.ub.edu/dnasp/)42. The network of haplotype was generated by the minimum spanning network method (epsilon = 0) using PopART v.1.7 (https://popart.maths.otago.ac.nz/download/)43. The nucleotide diversity, no. of segregating sites, Tajima’s D value and haplotype frequency were also analysed using DnaSP v.642.

Results and discussion

This study was done to characterise the mitochondrial genome of GH and its crossbreds. The complete mitogenome and D-loop sequences were used for phylogenetic analysis and genetic distance estimation and to access the matrilineal components in these pigs. The D-loop region of the mitochondrial genome has high mutation rate and variable than any other region of the nuclear or mitochondrial genome44 and thus important region for the phylogenetic analysis and evolution of animal breeds45.

Sequencing and submission of complete mtDNA genome

The entire mtDNA of 5 GH and one each of the Rani and Asha crossbred varieties were amplified using 30 pairs of overlapping primers and sequenced by Sanger sequencing. All the fragments, including the D-loop, 2 rRNA, 22 tRNA, 13 coding genes, and repeat regions, were aligned to obtain the complete mtDNA genome of each pig and the sequences were deposited into the NCBI GenBank database and assigned accession numbers MT501674, MZ703184, OM617468, OM634652, MZ647672 for GH breed; ON706057 for Rani; and ON715893 for Asha. Apart from the complete mtDNA genome, complete D-loop sequences were amplified and sequenced using Sanger sequencing from GH, Asha and Rani pigs. The sequences were trimmed and edited using DNAstar and Megalign software and these sequences were submitted to NCBI GenBank database and received the accession numbers for GH (OP185718, OP185719, OP185720, OP185721, OP185722, and OP185723); Asha (ON934748, ON934749, ON934750, ON934751, ON934752, and ON934753) and Rani (OP352470, OP352471, OP352472, OP352473, and OP352474).

The base composition of the mtDNA genome

The complete mtDNA of all the GH and its crossbred viz. Rani and Asha were found to be of size 16,690 bp and for all the mitogenome, viz. GH, Asha and Rani, the approximate base composition was 34.7% for Adenine (A), 25.8% for Thymine (T), 13.3% for Guanine (G) and 26.17% for Cytosine(C), while the G + C content was, 39.5% for all the breeds studied. The composition of different nucleotide base compositions of the mitochondrial genome is depicted in Table 1, which shows that majority of the nucleotides in the mitogenomes were AT-rich with 60.52% of total bases. The nucleotide composition and organization of the mitogenome of GH and its crossbreds was like other pigs viz. Min pig18. The size of the mitogenome in these pigs was comparable with the earlier reports of pigs in Asian as well as European pig breeds7,16,41,46,47. The nucleotide composition in mtDNA of I Pig was 34.66% for A, 26.24% for C, 13.35%, for G and 25.75% for T, the A + T content was 60.41%33 while in Indian wild boar was 26.25, 13.40, 25.79 and 34.56% for C, G, T and A, respectively and A + T content was 60.357.

Table 1 Nucleotide base composition of the mitochondrial genome of Ghoongroo and its crossbred pigs.
Full size table

Annotation of complete Mitogenome

The mitogenome has one non-coding control region viz. displacement loop (D-loop), 22 transfer RNAs (tRNAs), 2 ribosomal RNAs (rRNAs), and 13 protein-coding genes which were similar to other pig mitogenomes48. The typical circular structure of the mitogenome of GH, Asha and Rani is depicted in Fig. 2a–c which depicts the H and L strands, position of all the genes, tRNAs, rRNAs and D-loop along with GC content and Skewness of GC or AT. The annotation of the complete mitogenome of the GH breed is shown in Table 2. The D-loop was 1254 bp long and located between tRNA-Pro and tRNA-Phe having repeat regions. The two rRNAs viz. 12S and 16S rRNAs were 960 and 1570 bp, respectively. The size of tRNAs was varied from 59 (tRNA-Ser-1) to 75 bp (tRNA-Leu, tRNA-Asn) in size. The mitogenome had a total of 6 overlaps among all the genes ranging from 1 to 43 bp long. Additionally, there were 11 non-coding spaces ranging from 1 to 32 bp in length. The protein-coding genes had a total sequence length of 11,409 bp ranging from 204 (ATP8) to 1821 (ND5), which is 68.36% of the total mitogenome. The H-strand of the mitogenome consists of all the protein-coding genes, tRNAs, rRNAs and D-loop, except the ND6 gene and 8 tRNAs (tRNA-Ala, tRNA-Asn, tRNA-Cys, tRNA-Tyr, tRNA-Ser, tRNA-GLU, tRNA-Pro) which were encoded on L-strand (Fig. 2).

Fig. 2
Fig. 2Fig. 2
Full size image

Circular structure of complete mitochondrial genome of (a) Ghoongroo (b) Rani and (c) Asha pigs.

Table 2 Annotation of the complete mtDNA genome of Indian Ghoongroo and its crossbreds.
Full size table

Consistent with our findings, the length of 13 protein-coding genes ranged from 204 to 1821 bp, and a total of 11,413 bp in I pig33. Furthermore, studies in various pig populations33,48,49, birds50, and humans51 have identified one protein-coding gene and eight tRNA genes encoded on the L-strand. In contrast, prior research on the mitochondrial DNA (mtDNA) of wild and domestic pigs revealed only one protein-coding gene (ND6) and seven tRNA genes (tRNAGln, tRNAAla, tRNAAsn, tRNACys, tRNAPro, tRNATyr, tRNAGlu) encoded on the L-strand52. However, another study indicated two protein-coding genes (COX3, ND6) and two tRNA genes (tRNAPro, tRNAGlu) on the L-strand53. Notably, the complete mitochondrial genome of Daweizi and Ningxiang pigs revealed that all mitochondrial genes were encoded on the L-strand, except for three tRNA genes (tRNAIle, tRNAAsp, tRNALeu), which were encoded on the H-strand54,55. The arrangement and orientation of genes in the present study align with earlier reports on vertebrate and pig mitogenomes16,33,48,52,56.

Each protein-coding gene begins with the start codon ATG, except for ND4L, which starts with GTG, while ND3, ND2, and ND5 start with ATA. The termination codons in six protein-coding genes (ND3, COX2, COX3, ND1, ND2 and ND4) were incomplete and were subsequently completed by the addition of 3’ A residues to the mRNA during post-transcriptional polyadenylation. The annotation and codon sequences in our study are consistent with those found in the mitogenomes of other pig breeds from India and South-East Asia like I, Nicobari16,33,41,48,52. In contrast to our findings, Singh et al.53 reported that all protein-coding genes had ATG as the start codon except for ND3, ND4, ND5, ND1, and ND2, which started with ATA. In the mitochondrial genome of Daweizi pig, ND2, ND3, and ND5 began with ATA, ND4L with GTG, and ND6 with ATT, while the remaining proteins initiated with ATG54. In Swedish pig breeds, the start codon for ND2 and ND4L was ATT and GTG, respectively57.

The nucleotide composition bias in mitogenome was estimated by GC and AT skews, and it showed that all genes on H stand coding for protein and rRNA were negatively GC-skewed and positively AT-skewed, which denoted cytosine and adenine biasedness. The most cytosine bias was observed in ATPase 8 (− 0.52), while the least cytosine bias was observed in 16S rRNA (− 0.11). In contrast, the 12S rRNA and 16S rRNA genes had the most adenine bias (0.25), while the COX1 gene had no bias for AT content, which means that its adenine content is equal to the thymine content. The ND6 gene was the only gene coded in the L strand that was positively skewed for GC content (0.55), whereas it was negatively skewed for AT content (− 0.36) (Table 2). The AT content of the D-loop was 57.81%, whereas the GC skew and AT skew were found to be − 0.26 and 0.14, respectively. The AT content of the D-loop (57.81%) was lesser than that of the I pig (60.09%), Ningxiang pig (60.52%) and Wild pig (60.65%)33,52. The AT content, AT skew, and GC skew were used to determine the nucleotide composition behaviour of mitochondrial genomes as well as related to phylogenetics58,59.

The nucleotide composition bias within the mitogenome was assessed using GC and AT skews, revealing that all genes on the H strand encoding for proteins and rRNA exhibited negative GC-skew and positive AT-skew, indicating a bias towards cytosine and adenine. Among these, ATPase8 displayed the highest cytosine bias (− 0.52), while 16S rRNA showed the least (− 0.11). Conversely, 12S rRNA and 16S rRNA exhibited the highest adenine bias (0.25), while COX1 showed no bias in AT content, implying equal adenine and thymine content. Notably, the ND6 gene, the sole gene encoded on the L strand, demonstrated positive GC skew (0.55) and negative AT skew (− 0.36) (Table 2). The AT content of the D-loop was determined to be 57.81%, with corresponding GC and AT skews of − 0.26 and 0.14, respectively. This AT content was lower than that observed in I pig (60.09%), Ningxiang pig (60.52%), and Wild pig (60.65%)33,52. Utilizing AT content, AT skew, and GC skew aids in understanding the nucleotide composition patterns within mitochondrial genomes and their relevance to phylogenetics58,59.

Structure of tRNA

The mitogenome of GH and its crossbred had 22 tRNA genes encoded for 22 tRNAs which consists of 2 tRNAs for Leucine and Serine amino acids while one tRNA for each of the other amino acids. All the tRNAs except eight were encoded on the H strand. The tRNA structure predicted by the tRNAscan-SE server revealed that all the tRNAs have typical cloverleaf structures except Ser-I which was devoid of D-arm. The structure and number of tRNAs in our study were similar to earlier reports in pigs33. The structure of all the tRNAs is depicted in Fig. 3.

Fig. 3
Fig. 3
Full size image

tRNA structure of all 22 tRNAs found in the complete mitogenome of GH and its crossbreds.

Phylogenetic analysis and double matrilineal within Ghoongroo pigs

The aligned complete mitochondrial genome sequence of GH and its crossbred pigs identified 27 polymorphic sites (Table 3), with one insertion and one deletion in the rRNA region of the Rani mitogenome. Out of these polymorphic sites, other than the insertion and deletion, 9 were found in the control region of the D-loop while 2 in the rRNA region, 3 each in ND4L and ND5 gene, 2 each in COX1 and COX2 gene, and one each in ND4, ND6, ATP6 and Cytb (Table 3). Out of 27 polymorphic sites, 16 sites were present in the coding region of the mitogenome with mostly synonymous mutation and do not cause any change in the amino acid. Apart from complete mitogenomes, a complete D-loop region was also sequenced and aligned and a total of 18 polymorphic sites were observed in 1254 bp long D-loop (Table 4 and Supplementary Table 6). The phylogenetic analysis from the complete mitogenome sequence of GH and its crossbred along with the European pig breeds (Duroc, Large white Yorkshire, Hampshire, Landrace) and Asian pig breeds of different regions apart from India (Meishan, Banna Mini, I Pig, Ningxiang, Tibetan) was done using the AWH sequence as an outgroup. The phylogenetic analysis showed that different sequences of GH breeds were clustered in two different clades indicating two independent maternal lineages existing within GH. In one clade GH sequences were clustered with Rani crossbred while in other clade GH sequences were clustered with the Asha crossbred variety (Fig. 4). The presence of GH in two different clades indicated that GH breeds may have different matrilineal components and this may be the reason for their grouping in different clades. This hypothesis stands good if we see the position of Asha and Rani crossbreds, both crossbreds have mitochondrial inheritance of GH breed, Rani was clustered in one clade while Asha is in another clade which indicated that Rani crossbred used in this study may have developed from GH which have similar matrilineal components while Asha was developed from GH which have different matrilineal components. The phylogenetic analysis indicated that GH and its crossbred were present in separate clades from European and Asian pig breeds which shows independent domestication events in Indian Ghoongroo pig breed from other Asian and European pig breeds. The Indian Ghoongroo pigs might have originated from Indian wild boar locally independent from other Asian and European pig breeds. The same finding was also observed in studies by6,7 where independent domestication centres was reported in Indian, Asian and European regions. The haplotype analysis using a complete mitogenome also shows the same result (Fig. 5b, Supplementary Table 5b). There was a total of 1838 polymorphic or segregating sites, including 1539 singleton sites and 299 parsimony informative sites. The number of haplotypes generated from complete mitogenome sequences was 16 with haplotype diversity 1.00 ± 0.022. This represents a very high level of haplotype diversity and suggests a highly diverse population with a wide variety of haplotypes and few common ones. It was found that GH were present in three different haplotypes while Rani and Asha were present in separate haplotype. The haplotype of Rani has only 1 mutation different from one of the GH haplotypes while Asha haplotype has 3 mutations different from another GH haplotype (Fig. 5b and Supplementary Table 5b). It was also evident that GH and crossbreds had separate haplotype which may be due to the different matrilineal components within them as evident from the phylogenetic analysis. The nucleotide diversity in the complete mitogenome sequences were measured as 0.01855 ± 0.00903 while Tajima’s D was − 1.96091 with statistical significance of P < 0.05, which suggest that population is experiencing non-neutral evolutionary process. The nucleotide diversity indicating substantial genetic variation within the mitogenome sequences suggesting that the population has a diverse set of mitochondrial DNA sequences, which could be due to a large effective population size, long history of divergence within population, or high mutation rates in the mitogenome60. A high negative Tajima’s D value with statistical significance generally indicates an excess of rare variants compared to what is expected under neutral evolution. This may happen due to recent population expansion or purifying (negative) selection where deleterious mutations are removed more quickly, leading to an excess of low-frequency alleles61. This suggests deviation from neutrality, potentially indicating population expansion, purifying selection, or a recent bottleneck62. The analysis of molecular variance of haplotypes indicated the fixation index Phi_ST value as 0.95746 (P < 0.001, a significant complete mtDNA variation was found within (0.4866% and among (99.51% the population sequences of the pigs. A high phiST value of 0.99513 suggests that nearly all the genetic variation is present between population rather than within them. This means the populations are almost entirely distinct from each other genetically.

Table 3 Polymorphic sites of complete mitochondrial genome sequence of Ghoongroo and its crossbreds.
Full size table
Table 4 Polymorphic sites of complete D-loop sequence of Ghoongroo and its crossbred.
Full size table
Fig. 4
Fig. 4
Full size image

Phylogenetic tree generated using complete mitogenome sequences of GH and its crossbreds using Maximum Likelihood methods using 1000 bootstrap replications in MEGA11 and Visualise using FigTree v1.4.4. Two distinct clusters can be seen of GH in red and violet colour where Rani is clustered in violet clade while Asha is clustered in red clade. Asian and European pig breeds are clustered separately from GH and its crossbred in brown and blue colour, respectively. Indian Wild boar is present separately from both the haplogroup. African Warthog was used as an outgroup.

Fig. 5
Fig. 5
Full size image

Median Joining Haplotype Network created with PopArt v1.7 showing the genetic relationships between all individuals using (a) Complete D-loop sequences (b) Complete mitogenome sequences. Colours represent different pig populations, mutational steps between haplotypes are shown as hatchmark across connection lines. Each haplogroup is represented by a circle, with the area of the circle proportional to the haplogroup’s frequency.

To further validate the results of double matrilineal components in GH with complete mitogenome samples, another phylogenetic analysis was conducted on the complete D-loop sequences of GH, Rani and Asha. We have also downloaded the complete D-loop sequences of Hampshire, Duroc and Landrace for phylogenetic analysis to see whether any influence of paternal mitochondrial components was present in Rani and Asha. The transition-to-transversion (Ts/Tv) ratio in the complete D-loop sequences from this study was notably higher (5.09:1) than the critical value of 2:1 to 4:1 reported by Perna and Kocher63 and Knight and Mindell64, indicating a high preference for transitions in GH and its crossbreds. This may reflect evolutionary pressures, mutational biases. Our study corroborates to earlier reports where the high ratio (3.85–10.5) was found in different pigs23,65. In this study, Phi_ST were found to be 0.95746 (P < 0.001, a significant mtDNA D-loop variation was found within (4.25%) and among (95.75%) the breed sequences of the pigs. A high phiST value of 0.95746 suggests that nearly all the genetic variation is present between population rather than within them. This means the populations are almost entirely distinct from each other genetically. Similar to the results of complete mitogenome analysis, GH and its crossbreds Rani and Asha were clustered in two distinct clades (Fig. 6). It was clearly indicated that some of the GH, Rani and Asha were clustered together in one clade while the rest of GH, Rani and Asha were clustered in another clade. It may be due to the fact that the animals within the same clade have the same matrilineal components while animals between the two clades have different matrilineal components. The European breeds which were used to produce crossbreds have clustered separately in the third clade which shows that GH crossbreds have no effect on the matrilineal components of European breeds, signifying European breeds were used as sire components only and no mitochondrial inheritance was found in crossbreds from these European breeds.

Fig. 6
Fig. 6
Full size image

Phylogenetic tree generated using complete D-loop sequences of GH and its crossbreds using Maximum Likelihood methods using 1000 bootstrap replications in MEGA11 and Visualise using FigTree v1.4.4. The frequency and branch length are presented above and below the branch, respectively. Three distinct clusters can be seen in blue, red and green colour where blue and red haplogroup represent the GH and its crossbred while green haplogroup represent European pigs. (a) Radial representation (b) rectangular representation.

To further explore the genetic differentiation of the population, the haplotype analysis was performed and a median-joining network profile was generated (Fig. 5a). The haplotype network has a correlation with the results of the phylogenetic tree. There was a total of 125 polymorphic or variable sites, including 104 singleton sites and 21 parsimony informative sites. The number of haplotypes generated from complete D-loop sequences was 10 with haplotype diversity 0.720 ± 0.079. This indicates a moderately high level of haplotype diversity and suggests that the population has a fair amount of genetic variation, but there are some common haplotypes shared among individuals. Among the 10 haplotypes identified in this study, one haplotype was shared by GH and its crossbred, while other haplotypes were unique to GH, Rani, Asha and other European pig breeds. None of the haplotypes were shared between GH and its crossbreds with European pig breeds. It was found that GH, Asha and Rani shared a single largest haplotype (Fig. 5a and Supplementary Table 5a) which is obvious due to the same matrilineal components within them. It was also evident that GH and crossbreds had another separate haplotype which may be due to the different matrilineal components within them as evident from the phylogenetic analysis. Indian domestic breeds have unique and differentiated haplotypes and no haplotype sharing was found between Indian domestic pig breeds with exotic pig breeds23 and between wild and domestic pig breeds24. Larson et al.39 had also reported the similar findings from other regions. Our findings corroborate with the earlier report where there were 2 haplotypes in GH pig population based on 464 bp D-loop fragments in 10 individuals24. The nucleotide diversity measured 0.01333, while Tajima’s D was − 2.10929. These values suggest non-neutral evolution, potentially indicating an abundance of rare alleles, suggestive of positive selection or a selective sweep66. The nucleotide diversity found in our study is in line with the earlier reports where the nucleotide diversity was found to be 0.0124

The evolutionary divergence in the form of genetic distance between different breeds sequences was estimated using the Tamura-Nei model67 based on complete mitogenome sequences and presented in Supplementary Table 7. Apart from the pairwise genetic distance between breeds, the genetic distance between the groups were also estimated (Supplementary Table 8). The analysis showed that the genetic distance between two different matrilineal components of GH was 0.11% which is lowest than other groups comparisons. The genetic distance between GH and Asian Pigs was less (0.17%) than the distance between GH and European Pigs (1.31%). The genetic distance between GH and Indian Wild Boar was lesser than the distance between European pigs and Indian wild boar. As African warthog was used as an outgroup, the highest genetic distance was observed between AWH and other groups of pigs. However, pairwise genetic distance between breeds revealed that GH and its crossbreds were more distant to European pig breeds than the other Asian pig breeds. Among European pig breeds, the genetic distance with GH was highest in Hampshire (1.45%) and subsequently decreases in Landrace (1.35%), LWY (1.24% and Duroc (1.19%). Among Asian Pig breeds, the genetic distance was highest in Banna mini (0.25%) and subsequently decreases in I-Pig (0.24%), Meishan (0.14%), Tibetan (0.13%) and Ningxiang (0.12%). The results of genetic distance corroborated with the phylogenetic tree.

The pairwise percentage genetic distance between GH, GH crossbreds and European breeds and AWH was calculated using the maximum composite likelihood model38 based on complete D-loop sequences and presented in Table 5. The animals used for distance calculation were one from each haplogroup as depicted in supplementary table 5a. The analysis showed that the genetic distance between GH having different matrilineal components was more (0.47%) than the distance between GH and Asha (0.05–0.16%) or Rani (0.10%) crossbreds. As crossbreds Rani and Asha were developed from the GH matrilineal components, the genetic distance based on the D-loop should be negligible but a significant genetic distance was found between two GH and some of the Rani and Asha while with others it was negligible. It may be because Rani and Asha may have matrilineal components different from the GH for which distance was calculated. The genetic distance between GH and its crossbreds against European breeds revealed that the highest distance was between GH and Hampshire (1.27–1.39%) followed by Landrace (1.14–1.37%) and Duroc (0.85–1.11%). The results of genetic distance corroborated with the phylogenetic tree. Furthermore, after placing all the GH and its crossbreds in one group, European pigs in 2nd group and AWH in 3rd group, the genetic distance between the GH and European group was found to be 1.21% while between GH and AWH group was 6.09% while the distance between European and AWH was 6.9%. Our study corroborates with the earlier reports where the genetic distance between Indian wild boar and domestic pigs was 3.29%7 and 3.5%23 and European wild boars and domestic breeds were 1.16%46.

Table 5 Estimates of pairwise evolutionary divergence based on complete D-loop sequences of different pigs by the maximum composite likelihood method in MEGA11.
Full size table

The study first time used the complete mtDNA genome of GH and its crossbred varieties namely Rani and Asha pigs. The mtDNA genome sequencing was performed for GH pigs from different locations in their breeding tract from West Bengal to Assam. The study discovered that within the GH breed, a wide genetic diversity is present and it may have two subspecies, with exact same morphological characters, which is not possible to identify phenotypically. Genetic measures like sequencing and phylogenetic analysis of mtDNA have given the tools to identify the matrilineal components of GH pig and identification of such distinct clusters within a breed made possible. The complete mtDNA genome as well as complete D-loop sequences revealed the double matrilineal components within GH breeds. Moreover, the crossbred commercial varieties namely Rani (♀GH x ♂Hampshire) and Asha (♀Rani x ♂Duroc) where GH has been used as the maternal lineages also showed similar differentiation. The Rani and Asha were also differentiated into two different clusters confirming the double matrilineal components within the GH population. The complete mitochondrial genome data of five GH pigs as well as complete D-loop sequences of 11 pigs of this breed clearly validate that GH pig has maintained two different mitochondrial inheritances. Analogous patterns of gene flow parallel to these two maternal inheritances can also be seen in Rani and Asha, crossbred pigs developed using GH as the maternal lineage. In this connection, the phylogenetic analysis of the complete mitochondrial genome as well as complete D-loop sequences of Rani and Asha, validated the vertical inheritance of two mitochondrial genomes of GH subspecies among these crossbreds (Figs. 4 and 6). The GH pigs having two different matrilineage footprints might have encountered unusual natural/artificial selection pressure in the past, which led to the evolution of two distinct subspecies within the breed without affecting their phenotypic characters. The geographical isolation between Assam and West Bengal due to the Sankosh River, a tributary of the Brahmaputra River may be one of the reasons for the introduction of mitochondrial inheritance within GH from a related different ancestor. Therefore, from the study, it can be inferred that GH pigs have experienced a silent mitochondrial inheritance within the breed, which has changed their genetic framework but did not alter the phenotypic attributes of the animal.

This is the first study which revealed the two different matrilineal components within the same breed. As GH is distributed in a wide range of West Bengal and Assam states of India and wide variation is present in this breed but phenotypically all are the same. The double matrilineal components within GH may be due to their origination from not a single ancestor but from closely related maternal ancestors. This may also be possible due to the derivation/dispersal from the matrilineal pool of one region to the other regions because of the migration. A similar pattern has been observed in previous studies23,68. The GH breed might have originated locally from Indian Wild boars independent to other Asian and European pig breeds. The eight pig breeds of Shandong province of China had lower divergence and shared the same haplotype because of the possibility that they stemmed from closely related maternal ancestors, if not from a common ancestor69. The Dapulian Black and Laiwu Black breeds had independent maternal lineage but other indigenous pigs have extensive gene flow with other breeds69. The neighbour-joining tree analysis identified two distinct clades among individual pigs, with a Chinese domestic breed showing a scattered distribution across multiple breeds70. The commercial European breeds like Landrace, Hampshire, Large White Yorkshire and Duroc were clustered together in a separate clade independent of Indian pigs. Our findings align with earlier studies indicating a distinct phylogenetic separation between Indian pigs and European-American and Asian pig clades52. The separate matrilineage observed in Indian wild boars may stem from the differentiation of wild boar populations that originated from Island South East Asia (ISEA) and subsequently migrated to the Indian subcontinent before further dispersal to East Asia and eventually across Eurasia39. Additionally,6 reported that modern Indian domestic pigs have ancestral ties to local wild boar populations distinct from those found in Asia and Europe. Multiple domestication centres have been identified, including four for Chinese Native Pigs71 and six for native pigs in East Asia39. The results of this study is based on the 7 complete mitochondrial genome sequences as well as 24 complete D-loop sequences of GH and its crossbreds. We have included some of the sequences from the NCBI GenBank but due to limited complete D-loop sequences of Indian pigs in the databank, the study is limited by the small sample size. However, to estimate the genetic diversity and phylogenetic analysis using complete mitochondrial or Ccomplete D-loop sequences, preliminary studies were conducted with smaller sample size. The genetic diversity in 17 indigenous Chinese pig breeds and 3 European breeds was determined based on one sample from each breed only72. Similarly, the phylogenetic relationships between Asian and European pig breeds were determined using one or two samples from 19 breeds of pig46.

The GH pig is the most prolific pig breed in India and shows a wide range of variability in terms of reproductive and productive performance73 which may be linked to the two different genetic subpopulations of GH having different maternal lineages. The absence of a structured breeding program for indigenous pig breeds like GH has contributed to a decline in their population74. Consequently, it becomes crucial to characterize the mitochondrial DNA (mtDNA) of GH and its crossbred counterparts to assess genetic diversity and maternal lineages within these pigs. This information is pivotal for designing appropriate breeding programs aimed at conserving this significant pig breed. In contrast to many species where domestication has been one of the major causes for drastic change in their morphological parameters, in this study it has been particularly uncovered that GH pig due to multiple expansion events though maintaining two different maternal lines in their genome has experienced no phenotypic changes in their morphology. The mitochondrial genomic data of GH pigs having two matrilineage, annotated in this study, could be fine-tuned as a powerful tool to trace the origin of pig domestication, conserve their indigenous germplasm and identify the purity as well as elucidate the lineage of other nondescript pigs.

Conclusion

This study aimed to identify the molecular breed signatures of GH pigs through an integrative analysis of mtDNA expression and D-loop sequence data of GH pigs from different locations and their crossbred species (Rani and Asha). As evidenced by the mitochondrial genomic data GH pigs of this region have undergone multiple domestication events and have been carrying two different maternal lineages in their genome from the past, which is also apparent in their crossbred species. The study also underscores that GH pigs have undergone silent mitochondrial inheritance in their breed, without affecting their morphological attributes. The study identified two unique maternal legacies, which might be useful for their genotypic identification and thus work as an input for designing and implementing genetic strategies to conserve this important indigenous pig breed of India.