A first nuclear and mitochondrial genomic portrait of Robinson Crusoe’s (Juan Fernández Island) spiny lobster Jasus frontalis (Crustacea: Decapoda: Achelata)

Abstract

Background

The Juan Fernández rock lobster Jasus frontalis (H. Milne-Edwards, 1837) is found along the Juan Fernández archipelago and Desventuradas Islands in the southeastern Pacific Ocean where it supports a profitable but likely unsustainable artisanal fishery. Only a few genomic resources are available for this species to support efficient managing practices. In this study, we took a first look at the genome of J. frontalis profiting from recently developed bioinformatics tools that mine for biological insight from low sequencing-depth datasets. Specifically, we have (i) estimated the nuclear genome size, (ii) assembled and described in detail the mitochondrial genome, (iii) discovered, quantified, and annotated nuclear transposable elements, and (iv) examined the phylogenetic position of the genus Jasus among spiny lobsters (family Palinuridae) using a maximum likelihood phylogenetic analysis based on translated mitochondrial protein coding genes (PCG’s).

Results

Using an in-silico k-mer approach, the haploid nuclear genome size (GS) calculated for J. frontalis ranged between 1.42 Gbp (estimated using k-mer = 54 bp) and 1.65 Gbp (with k-mer = 18). Our GS estimates are well below the GS (= 4.56 Gbp) determined for the same species using flow cytometry. In J. frontalis, the circular mitochondrial genome is 15,514 bp in length and contains 22 transfer (tRNA) genes, 13 PCGs, and two ribosomal RNA genes (12 S and 16 S ribosomal RNA). In the nuclear genome of J. frontalis, repetitive elements content varied between 41% (with k-mer = 54) and 64% (using k-mer = 18 bp). Nearly a third (64.94%) of the transposable elements were not annotated. Among annotated transposable elements, Long Interspersed Nuclear Elements (LINEs, 12.64%), DNA transposons (7.35%), and Long Terminal Repeats (LTRs, 5.44%) were the most common. Less abundant transposable elements included Rolling Circles (1.3%), Short Interspersed Nuclear Elements (SINEs, 0.64%), Satellite DNA (0.43%), and simple repeats (0.42%), among others. A phylomitogenomic analysis based on PCGs indicated supported the monophyletic status of the infraorder Achelata and the superfamilies Paniluridae and Syllaridae. The genus Jasus together with Sagmariasus verreauxi were recovered as a monophyletic clade that also provided support for the monophyletic status of lobsters that do not produce sound (“Silentes”).

Conclusion

The genomic resources developed in this study will support conservation strategies and efficient fisheries management in this spiny lobster that is likely overexploited.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12864-025-11794-5.

Background

In the species-rich crustacean order Decapoda, a clade with remarkable anatomical disparity [1], representatives of the infraorder Achelata (spiny and slipper lobsters) share three synapomorphies; (i) greatly enlarged second antennae, (ii) lack of claws (chelae) on the first, second, and third pair of pereopods, and (iii) a long-lived teleplanic larvae; the phyllosomata [2]. In this infraorder, several spiny and slipper lobsters reach relatively high population abundances and attain large body sizes, subsequently being targeted by subsistence, artisanal, or industrial fisheries, in subtropical, tropical, and/or warm temperate coasts in all continents except Antarctica [2].

Among spiny lobsters (Achelata: Palinuridae), the Juan Fernández rock lobster Jasus frontalis (H. Milne-Edwards, 1837) is found in the southeastern Pacific Ocean, particularly around the Juan Fernández archipelago and Desventuradas (= San Félix) Islands [2]). Jasus frontalis inhabits the rocky subtidal between 2 and 200 m depth (Retamal and Arana, 2000), attains a maximum body length of 48 cm [2] and is captured by a profitable artisanal fishery. Decreasing catches with increases in fishing effort observed during the last decades suggests that the species is experiencing overexploitation [3] but see [4]. The life history of J. frontalis is relatively well known [5–8], but only a few genomic resources are available for this species [9, 10].

Using a fragment of the mitochondrial protein coding gene cytochrome oxidase subunit I (cox1) [9] concluded that J. frontalis comprises a single population with high gene flow between the Juan Fernández and Desventuradas island systems. Most recently, [10] examined connectivity among populations located in the Juan Fernández and Desventuradas Islands using a genome complexity reduction-based sequencing approach (DaRTseq) finding evidence for two genetic populations, one restricted to each island system that are separated by 850 km. Further Lagrangian modelling suggested asymmetric larval connectivity between island systems, greater from the Juan Fernández archipelago to the Desventuradas Islands [10]. Additional genomic resources will be most valuable to continue improving our understanding about the biology of this heavily fished spiny lobster.

This work is part of a broad research program that aims to develop genomic resources for marine vertebrates and invertebrates targeted by fisheries in the southeastern Pacific Ocean, including J. frontalis [11, 12]. In this study, we provide a first look at the genome of J. frontalis (Fig. 1) profiting from bioinformatics tools developed for mining biological insight from low sequencing-depth next generation sequencing (NGS) short-read datasets. In particular, (i) using a k-mer strategy we have estimated the nuclear genome size in the studied lobster, (ii) assembled the mitochondrial genome and analyzed it in detail, and (iii) quantified and annotated transposable elements in the nuclear genome of J. frontalis. Lastly, (iv) we have explored the phylogenetic position of the genus Jasus and J. frontalis within the family Palinuridae using the information gathered from translated mitochondrial protein coding genes. The biological insight generated in this study will direct the assembly of a chromosome-level nuclear genome in J. frontalis, contribute to the development of biomonitoring strategies targeting this species using environmental DNA, and aid in surveying the public marketplace to identify mislabeling of spiny lobsters commodities.

Fig. 1.

The Juan Fernández rock lobster Jasus frontalis (top) and sankey diagram (bottom) of the Kraken2 report results obtained for Jasus frontalis filtered reads before genome size estimation using a k-mer approach. Photograph published with permission from Erasmo Macaya

Methods

DNA extraction, library preparation, and sequencing

The J. frontalis tissue sample utilized in this study is deposited at the Clemson University Crustacean Collection. A muscle tissue sample (approx. 3–4 mm³) preserved in ethyl alcohol (95%) was dissected from a pereiopod using sterilized forceps and immediately transferred to a 1.5 mm sterile centrifuge tube (Eppendorf, Germany) for genomic DNA (gDNA) extraction with the DNeasy^® Blood & Tissue kit (Qiagen, Germany) using the manufacturer’s protocol. Next, gDNA was shipped to Novogene Inc. (Palo Alto, CA) for next generation sequencing (NGS). Library preparation was performed using the Illumina TruSeq kit following the manufacturer’s instructions. NGS was performed using an Illumina HiSeq X Plus system (Illumina, San Diego, CA, USA) with a 2 × 150 cycle. A total of 31,202,775 paired (PE) reads were generated by the sequencing facility and can be retrieved from short read archive (SRA) repository at NCBI’s GenBank (Bioproject: PRJNA780953, BioSample: SAMN46384295; SRA accession number: SRR32094225).

Cleaning and decontamination of raw reads

We used the program fastp v.0.20.1 with default parameters [13] to remove Illumina adapters and low quality sequences (with Phred scores < 20) from the dataset. Then, the program Kraken2 v2.1.2 [14] and the database kraken2-microbial-fatfree (https://lomanlab.github.io/mockcommunity/mc_databases.html) was used to ‘decontaminate’ the set of high-quality reads from human, protozoan, fungal, bacterial, archaeal, and viral reads.

Genome size of Jasus frontalis

With the high quality and decontaminated set of reads, we used the program KMC 3 v. 3.2.1 [15] to calculate the k-mer frequency distribution of 13 different k-mer sizes; 18, 21, 24, 27, 30, 33, 36, 39, 42, 45, 48, 51, and 54 bp long. Next, the web server GenomeScope2 [16] and the program REPeat SPECTra Estimation (RESPECT) v.1.0.0 [17] were used to analyze each of the 13 different k-mer frequency distributions and estimate the nuclear genome size of J. frontalis.

Repetitive elements in the nuclear genome of Jasus frontalis

Following recommendations in [18, 19], first, we used the program HISAT2 v2.2.1 [20] to map the high quality and decontaminated set of PE reads to the mitochondrial genome of Jasus frontalis (assembled in this study, see below). Then, we used the reads that did not map to the mitochondrial genome (n = 29,268,589 PE reads [99.78%]) for exploring the nuclear ‘repeatome’ of J. frontalis with the pipeline DnaPipeTE v1.4c [18, 19]. DnaPipeTE first assembled repetitive elements with the software Trinity [21] to then annotated them with the program RepeatMasker [22]. Finally, DnaPipeTE quantified the abundance of each detected repetitive element by mapping a random sample of the reads onto the newly assembled repetitive elements [11, 18, 19]. We ran DnaPipeTE using two iterations of the Trinity assembler, each time using an independent set of read sampled at 0.15X [19]. Also, the Protostomia-specific database of transposable elements (from Dfam - [23]) was used during the analysis. Lastly, dnaPipeTE calculated the divergence (based on blastn) between transposable elements copies in the J. frontalis nuclear genome and their respective consensus sequences to determine the repetitive elements landscape of J. frontalis [19].

Mitochondrial genome of Jasus frontalis

The mitochondrial genome of J. frontalis was ‘de novo’ assembled using the pipeline GetOrganelle v1.6.4 [24]. The mitochondrial genome of the congeneric rock lobster J. edwardsii (NCBI’s GenBank accession number PP178561) was employed as a ‘seed’ for the assembly that used k-mer sizes of 21, 55, 85, and 115 bp. Then, the pipeline MITOS2 [25] available in the web platform Galaxy [26] was used for in silico annotation of the newly assembled mitochondrial molecule. A circular map of the assembled and annotated mitochondrial genome was created using the web server Genome Vx (http://wolfe.ucd.ie/GenomeVx/— [27]). The software MEGA X [28] was employed to calculate nucleotide composition of the entire mitochondrial genome and particular genes. Also, codon usage and Relative Synonymous Codon Usage (RSCU) was estimated using the ‘Codon Usage Tool’ in the web server Sequence Manipulation Suite (https://www.bioinformatics.org/sms2/codon_usage.html— [29].

Selective pressures in mitochondrial PCGs were examined using the software KaKs_calculator 2.0 [30]. For each PCG, the number of nonsynonymous substitutions per nonsynonymous site (dN), synonymous substitutions per synonymous site (dS), and the ratio dN/dS (= ω) were estimated using the congeneric rock lobster J. edwardsii as the outgroup (GenBank accession number PP178561). Observed ω ratios, are expected to be equal to 1, < 1, or > 1, if a PCG are exposed to neutral, purifying (negative), or diversifying (positive) selection, respectively [30].

We used the program MITFI [31] implemented in MITOS2 to predict the secondary structure of each mitochondrial tRNA gene and visualization of each predicted secondary structure was conducted with the web server Forna (http://rna.tbi.univie.ac.at/forna/; [32].

The mitochondrial control region (CR) was examined in detail. First, we determined the presence of microsatellites (= Simple Sequence Repeats, SSRs) and short tandem repeats in the CR using, respectively, the web servers Microsatellites Repeats Finder (http://insilico.ehu.es/mini_tools/microsatellites/— [33]) and Tandem Repeats Finder (https://tandem.bu.edu/trf/trf.html— [34]). Furthermore, we detected the existence (or not) of molecular ‘hairpins’ or ‘stem and loops’ in the studied CR using the web platform RNAfold (http://rna.tbi.univie.ac.at//cgi-bin/RNAWebSuite/RNAfold.cgi— [35]).

Phylogenetic placement of Jasus frontalis

We examined the phylogenetic position of the genus Jasus and J. frontalis using a maximum likelihood (ML) phylogenetic analysis based on mitochondrial PCGs (amino acids). The ML phylogenetic analysis was performed using the herein assembled mitochondrial genome of J. frontalis, the congeneric J. edwardsii mitochondrial genome available in NCBI’s GenBank but with no companion paper, and other 17 cofamilial lobsters with annotated mitochondrial genomes available in GenBank. Other 7 species belonging to family Scyllaridae, one species belonging to the infraorder Caridea, and one species belonging to the infraorder Astacidea, were included as outgroups, as in [11]. The ML phylogenetic analysis proceed by first aligning each of the 13 mitochondrial PCG sequences using the software MUSCLE [36] as implemented in the program MEGA X. Then, we used the program GBlocks [37] to trim poorly aligned regions in each PCG alignment. Finally, all alignments were concatenated and submitted to the web server IQ-TREE version 1.6.10 for ML analysis [38] that used the software ProtTest [39] to partition the concatenated dataset and select the best fitting models of sequence evolution for each partition, as in [11]. Support for nodes in the retrieved ML tree topology was determined using 1,000 (ultra-fast) bootstrap iterations and the Shimodaira–Hasegawa approximate likelihood ratio test ([SH]-aLRT).

Results and discussion

A total of 60,362,500 (96.73%) high quality clean reads remained for downstream analyses after low quality sequences and Illumina adapters were filtered out (by the program fastp) from the dataset (raw reads) made available by the sequencing facility (n = 62,405,550 total reads). Also, a total of 2.81% (n = 847,991) of the clean reads were classified as contaminants by the program Kraken2 (Fig. 1). Most of the contaminant reads were of bacterial (0.937%) and fungal (0.742%) origin and only a few from protozoan (0.38%) and viral (0.0491%) origin (Fig. 1). No read was classified as human (or chordate) and no taxon-specific overrepresentation was detected among contaminant reads.

Nuclear genome size of Jasus frontalis

Using an in-silico k-mer approach, the haploid nuclear genome size (GS) calculated for J. frontalis with the software RESPECT ranged between 1,420,361,528 bp (1.42 Gbp) estimated with a k-mer = 54 bp and 1,645,914,013 (1.65 Gbp) using a k-mer size equal to 18. In turn, using the software GenomeScope2, GS in J. frontalis ranged between 43,017,128 bp (k-mer = 18) and 16,545,164 bp (k-mer = 51) (Table 1). Using flow cytometry, genome size for J. frontalis has been previously estimated to be 4.56 Gbp [40]. In the congenerics Red rock lobster J. edwardsii and Southern rock lobster J. novaehollandiae, genome size, also estimated using flow cytometry, is 4.89 and 5.21 Gbp, respectively [40]. In the family Paniluridae, genome size varies between a minimum of 3.08 Gbp in the Pink spiny lobster Palinurus mauritanicus [40] and a maximum of 5.43 Gbp in the Caribbean spiny lobster Panulirus argus [41]. Lastly, a recent study reports a genome size of 2.65 Gbp for Panulirus ornatus based on the assembly length of a chromosome-level genome [42]. Our calculations of genome size based on k-mers for J. frontalis are well below the range reported for congeneric and cofamilial species and the difference might be explained by the small number of reads used for the analysis; the program GenomeScope2 [16] is known to bias downwards genome size estimations with decreasing sequencing depths used for calculation [43]. In turn, the program RESPECT is supposed to provide accurate genome size estimations when working with low sequencing depth datasets [17] and benchmarking of this software has been shown to accurately estimate relatively large genome sizes (up to 3 Gbp) using low sequencing depth datasets (1x) [17]. For instance, in the Atlantic Salmon Salmo salar and the Cape sengi Elephantulus edwardii, with (assembly-based) genome size estimates of 2,966,890,203 and 3,843,982,861 bp, respectively, RESPECT estimated the correct genome size with a relatively low error (−5%, and 18%, respectively, see Fig. 3 and Supplementary Table A in [17] using a sequencing depth of 1x. Importantly, the sequencing depth used for our calculations corresponds to 2.05x as estimated by the software RASUSA [44] assuming a genome size equal to 4.56 Gbp [40]. We argue that our results testify to the limitations of RESPECT to accurately estimate genome size with very low sequencing depth (2x) in species with relatively large genomes (> 4.5 Gb). We note that repetitive element content is high in the studied species (see below) and such high repetitive element content is known to bias genome size estimates down, as well [16]. Additional benchmarking is needed to explore the effect of sequencing depth, repetitive element content, and their interaction on genome size prediction in tools based on k-mer profiling. As for now, our information suggests that the nuclear genome size in J. frontalis is at least 1.65 Gbp.

Table 1.

Genome size and repetitive element content in Jasus frontalis estimated based on Kmer analyses by two bioinformatics tools, GenomeSCOPE2 and RESPECT

	GenomeSCOPE2		RESPECT
k-mer size	Genome Size	Uniqueness ratio	Genome Size	Uniqueness ratio
18	43,017,128	13	1,645,914,013	36
21	39,195,731	12.8	1,640,100,813	48
24	36,343,008	13	1,628,403,821	50
27	33,749,311	13. 2	1,614,469,121	51
30	31,229,679	13. 5	1,598,548,919	53
33	28,902,065	13.9	1,580,813,776	54
36	26,633,202	14.3	1,561,811,388	55
39	24,609,641	14.7	1,541,546,640	55
42	22,847,492	15.1	1,519,672,316	57
45	21,098,615	15.5	1,496,473,627	57
48	19,456,857	16	1,472,210,024	58
51	17,927,442	16.5	1,446,773,757	58
54	16,545,164	16.9	1,420,361,528	59

Fig. 3.

Circular map of the mitochondrial genome of Juan Fernández rock lobster Jasus frontalis. Photograph published with permission from Erasmo Macaya

Repetitive elements in the nuclear genome of Jasus frontalis

In the nuclear genome of J. frontalis, repetitive elements content, as estimated by the software RESPECT, varied between 41% (with k-mer = 54) and 64% (using k-mer = 18 bp). In turn, the program dnaPipeTE estimated repetitive elements to comprise 43.81% of the nuclear genome in J. frontalis (Fig. 2). The latter value is within the range estimated by the software RESPECT. In general, the ‘repeatome’ comprises nearly a half or up to a third of the nuclear genome in J. frontalis. Repetitive genome content is poorly known in spiny lobsters. In P. ornatus, the only spiny lobster and representative of the Achelata with a chromosome-level assembled nuclear genome, repetitive elements comprise 65.67% of the genome [42]. In turn, genomic surveys have revealed repetitive elements content equal to 56.28% in the long-legged spiny lobster P. longipes [45], 57.64% in the Eastern Island spiny lobster P. pascuensis [11] and 69.02% in the Caribbean spiny lobster P. argus [43].

Fig. 2.

Transposable elements genome composition (circle graph) and landscape in the genome of Juan Fernández rock lobster Jasus frontalis

The program DnaPipeTE did not assign two thirds (66.92%) of the repetitive elements detected in the nuclear genome of J. frontalis to any known family of transposable elements specific to the Protostomia Dfam database (Fig. 2). Among those repetitive elements that were annotated, Long Interspersed Nuclear Elements (LINEs, 12.64%), DNA transposons (7.35%), and Long Terminal Repeats (LTRs, 5.44%) were the most common. Less abundant transposable elements included Rolling Circles (1.3%), Short Interspersed Nuclear Elements (SINEs, 0.64%), Satellite DNA (0.43%), and simple repeats (0.42%), among others (Fig. 2). The proportion of annotated transposable elements and repetitive element content herein estimated for J. frontalis is similar to that reported for the few other species whose ‘repeatome’ has previously been examined using low sequencing depth datasets [11, 43, 45]. In P. argus, P. longipes, and P. pascuensis, the portion of unannotated transposable elements in the nuclear genome correspond to 49.6%, 52.38%, and 64.94%, respectively [11, 43, 45]. However, in the chromosome-level genome assembly of P. ornatus, nearly all transposable elements were annotated by [42]. Also, transposable element content in J. frontalis is similar to that reported for the few other species of spiny lobsters with published genomic resources [11, 42, 43, 45].

Lastly, in J. frontalis the repetitive elements ‘landscape’ revealed a right-skewed distribution indicating a burst of transposable elements in the recent past (Fig. 2), similar to that reported for the Eastern Island spiny lobster P. pascuensis, one out of two species of spiny lobster in which the repetitive elements landscape has been explored [11]. In turn, in P. ornatus, the repetitive elements landscape exhibited at least 3 peaks suggesting multiple transposable elements burst both in the recent and distant past [42].

Future studies examining in detail the ‘repeatome’ of spiny lobsters most likely will reveal numerous novel transposable elements, considering the over-abundance of unannotated transposable elements in this and other spiny lobster genomes ([11, 43, 45, but see [42]). Furthermore, studies examining the repetitive elements landscape in spiny lobsters, the Achelata, and other decapod crustaceans have the potential to reveal connections between these bursts and the environmental conditions in the recent and distant past. Repetitive elements have recently been suggested to affect the capacity of their hosts to respond to environmental insults [46]. The role that transposable elements have in driving the ability of spiny lobsters and other marine organisms to react to global climate change remains to be examined.

Mitochondrial genome of Jasus frontalis

A circularized (complete) mitochondrial genome was assembled by the pipeline GetOrganelle with a coverage (average) of 131.5x per k-mer and 547.9x per nucleotide. The assembled mitochondrial genome of J. frontalis (GeneBank accession number PQ869192) is 15,517 bp long and encoded 13 protein-coding genes (PCGs), 22 transfer RNA (tRNA) genes, and 2 ribosomal RNA genes (12 S ribosomal RNA [rrnS] and 16 S ribosomal RNA [rrnL] (Table 2; Fig. 3). The newly assembled mitochondrial genome also contains a relatively short (595 bp) non-coding putative Control Region (CR). The length of the studied mitochondrial genome is similar to that reported for other closely related species; in the infraorder Achelata, mitochondrial genomes range from 15,470 bp in Sagmariasus verreauxi [47] to 16,826 bp in Thenus orientalis [48]. In the eastern rock lobster Jasus edwardsii (NC_088542), the only other representative of the genus Jasus with an assembled mitogenome, the length is 15,504 bp. Also, mitochondrial synteny in J. frontalis is identical to that reported before for all other cofamilial species [11, 49–51].

Table 2.

Mitochondrial genome of Jasus frontalis with annotations of protein-coding, tRNA, rRNA, and control region sequences

Name	Type	Start	Stop	Strand	Length (bp)	Start	Stop	Anticodon	Continuity
trnY (gta)	tRNA	1	67	-	68			TAC	−3
cox1	PCG	65	1604	+	1540	ATC	T**		0
trnL2(taa)	tRNA	1605	1669	+	65			TTA	4
cox2	PCG	1674	2361	+	688	ATG	T**		0
trnK(ttt)	tRNA	2362	2425	+	64			AAA	0
trnD(gac)	tRNA	2426	2488	+	63			GTC	0
atp8	PCG	2489	2647	+	159	ATG	TAA		−7
atp6	PCG	2641	3318	+	678	ATG	TAA		−1
cox3	PCG	3318	4109	+	792	ATG	TAA		−1
trnG(tcc)	tRNA	4109	4173	+	65			GGA	0
nad3	PCG	4174	4527	+	354	ATT	TAA		−1
trnA(tgc)	tRNA	4527	4588	+	62			GCA	7
trnR(tcg)	tRNA	4596	4660	+	65			CGA	4
trnN(gtt)	tRNA	4665	4730	+	66			AAC	0
trnS1(tct)	tRNA	4731	4796	+	66			AGA	−2
trnE(ttc)	tRNA	4795	4864	+	70			GAA	0
trnF(gaa)	tRNA	4865	4930	-	66			TTC	3
nad5	PCG	4934	6662	-	1729	ATG	T**		0
trnH(gtg)	tRNA	6663	6727	-	65			CAC	−1
nad4	PCG	6727	8067	-	1341	ATG	TAA		−7
nad4l	PCG	8061	8363	-	303	ATG	TAA		2
trnT(tgt)	tRNA	8366	8431	+	66			ACA	0
trnP(tgg)	tRNA	8432	8497	-	66			CCA	2
nad6	PCG	8500	9018	+	519	ATC	TAA		−1
cob	PCG	9018	10,155	+	1138	ATG	T**		0
trnS2(tga)	tRNA	10,156	10,223	+	68			TCA	34
nad1	PCG	10,258	11,202	-	945	GTG	TAA		22
trnL1(tag)	tRNA	11,225	11,295	-	71			CTA	−25
rrnL	rRNA	11,271	12,625	-	1355				22
trnV(tac)	tRNA	12,648	12,719	-	72			GTA	−5
rrnS	rRNA	12,715	13,585	-	871				0
CR*		13,586	14,180	+	595				0
trnI(gat)	tRNA	14,181	14,246	+	66			ATC	−3
trnQ(ttg)	tRNA	14,244	14,312	-	69			CAA	6
trnM(cat)	tRNA	14,319	14,388	+	70			ATG	0
nad2	PCG	14,389	15,387	+	999	ATG	TAA		−2
trnW(gca)	tRNA	15,386	15,454	+	69			TGC	−1
trnC(tgc)	tRNA	15,454	15,517	-	64			GCA	0

The nucleotide composition of the positive DNA strand in the mitochondrial genome of J. frontalis was T = 31.59%, A = 33.29%, G = 13.41%, and C = 21.71%. The mitochondrial A + T content estimated for J. frontalis (64.88%) was similar to that reported for other cofamilial species, which range from 60.8% in P. echinatus [51] to 67.1% in P. homarus [52].

In the mitochondrial genome of J. frontalis, all but one (nad1) of the 13 PCGs utilized typical crustacean stop and start codons (Table 2). The PCG nad1 began with an unusual start codon, GTG, which has been previously observed in other spiny lobsters (Sagmariasus verreauxi– [47]; Panulirus argus– [50]. Also, all but four genes ended with canonical stop codons. The PCGs cox1, cox2, nad5, and cob terminated with the incomplete stop codon T. Incomplete stop codons (T and TA) are often reported for PCGs in the family Palinuridae, Scyllaridae, and other decapod crustaceans [11, 47, 49–51, 53–64].

Codons were not used proportionally in the mitochondrial PCGs of J. frontalis. Most frequently used codons were AT-rich, and included TTA (Leu, n = 258), TTT (Phe, n = 237), and ATT (Ile, n = 214). Least frequently used codons (excluding stop codons) were GC-rich, and included ACG (Thr, n = 5), CGT (Arg, n = 7), and CGC (Arg, n = 9) (Supplementary Materials, Table S1). The RSCU analysis also revealed that the most frequently used codons among the synonymous codons were AT-rich (Fig. 4). Our results align with the codon usage reported for the few other spiny lobsters in which this analysis has been conducted, including P. argus [50], P. echinatus [48], P. interruptus [51], and P. pascuensis [11]. Codon usage biases are often observed in eumetazoans and influence gene regulation, translation, transcription, and expression [65].

Fig. 4.

Relative synonymous usage in the 13 protein coding genes encoded in the mitochondrial genome of Juan Fernández rock lobster Jasus frontalis

The analysis of selective pressures in the mitochondrial PCGs of J. frontalis indicated that all studied genes are evolving under purifying selection (Ka/Ks < 1) (Table 3). We note that the Ka/Ks value was not estimated for nad4l given the absence of nonsynonymous codon substitutions in this gene. The Ka/Ks ratio calculated for cox1 (Ka/Ks = 0.84498) was notably high compared to other PCGs. In contrast, the Ka/Ks ratio for cob was very low (Ka/Ks = 5.22E-15). So far, pervasive purifying selection across all PCGs has been revealed in four other species of Palinurid spiny lobster [11, 50, 51] and a few other crustaceans (i.e. caridean shrimps– [66]; penaeid shrimps– [67, 68]; and brachyuran crabs– [69].

Table 3.

Selective pressure analysis of the protein coding genes in the mitochondrial genome of Panulirus pascuensis

PCG	KA	KS	KA/KS	P-value
atp6	0.00205697	0.135455	0.0151856	3.61E-12
atp8	0.007967	0.080501	0.098966	0.07828
cob	1.11E-15	0.21263	5.22E-15	< 0.00001
cox1	2.31544	2.74023	0.84498	0.266494
cox2	0.002148	0.140396	0.015302	9.93E-13
cox3	0.001697	0.178378	0.009516	4.73E-18
nad1	0.001437	0.232354	0.006184	7.00E-27
nad2	0.005765	0.208827	0.027605	8.35E-24
nad3	0.003537	0.309134	0.011441	3.42E-11
nad4	0.001041	0.255748	0.00407	6.01E-41
nad4l	0	0.143045	-	-
nad5	0.003099	0.204132	0.015181	2.19E-40
nad6	0.007979	0.252892	0.031552	8.54E-14

In the studied mitochondrial genome, tRNA genes ranged between 62 bp (tRNA-Ala) and 72 bp (tRNA-Val) and typically featured a standard ‘cloverleaf’ secondary structure (Table 2; Fig. 5). Two substantial exceptions to the standard secondary structure were observed. Firstly, tRNA-Ser1 possessed a T-loop, but was missing the TΨC helical stem. Although the effect is unknown, the truncation or absence of the TΨC helical stem in tRNA-Ser1 commonly occurs in other eumetazoans [31], including the spiny lobsters P. argus [50], P. japonicus [53], P. pascuensis [11], P. polyphagus [58], P. ornatus [54], and P. stimpsoni [49]. In addition, tRNA-Thr was missing the T-loop. The absence or extreme reduction of the T-loop in tRNA-Thr has not yet been reported within the Achelata.

Fig. 5.

Secondary structure of the tRNA genes in the Juan Fernández rock lobster Jasus frontalis. All tRNA genes exhibited a standard cloverleaf structure, with the exception of tRNA-Ser1 and tRNA-Thr. Depicted in the web server Forna [32]

In J. frontalis, the non-coding putative CR (595 bp long) is located between the 12 S ribosomal RNA and tRNA-I (Fig. 1) and is AT-rich (AT content = 70.62%), in line to that observed in other spiny lobsters with assembled mitochondrial genomes [11] and references therein). A total of seven dinucleotide microsatellites (repeated between 3 and 5 times) were detected in the studied CR (Supplementary Materials, Table S2). Furthermore, a single short tandem repeat, 17 bp long (consensus motif: 5’– TTTATTAAATGTACACT– 3’), repeated 3.4 times, was detected in the studied CR, in agreement to that observed in Panulirus stimpsoni [49] but in disagreement with observations in P. argus [50] and P. pascuensis [11]. Finally, hair-pin/stem-loop structures were observed along most of the length of the studied CR after predicting its secondary structure with the web server RNAFold (Supplementary Materials, Fig. S1). In line with our observations, hair-pins/stem-loops are often observed in the CR of the few other spiny lobsters in which the secondary structure of this region has been examined [11, 50]. Future studies examining in detail the organization of the CR have the potential to advance our understanding of the role that this relatively long non-coding region plays in mitochondrial replication and transcription.

Phylogenetic position of the genus Jasus and Jasus frontalis

A total of 28 terminal nodes, 11,126 characters, and 6,099 informative sites were used during the ML phylogenetic analysis (Fig. 6). In line with previous analyses based on translated mitochondrial PCGs (i.e [11], our analysis recovered the infraorder Achelata and the families Palinuridae and Scyllaridae as monophyletic with bootstrap support (bv) values equal to 100, 100, and 98, respectively. In the family Palinuridae, the genus Jasus (represented by J. frontalis and J. edwardsii in our analysis, recovered as monophyletic [bv = 100]) and the eastern rock lobster Sagmariasus verreauxi clustered into a fully supported monophyletic clade [bv = 100] that exhibited an early branching position; this clade comprising spiny lobsters that do not produce sound (= “Silentes” lobsters) was sister to all other spiny lobsters that produce sound (= “Stridentes”) included in our analysis. Nonetheless, support for the Stridentes clade was moderate to poor (bv = 78). In the latter clade, the phylogenetic relationship among the species included in our analysis were identical to those recently recovered by [11].

Fig. 6.

Maximum likelihood phylogenetic hypothesis for the family Palinuridae and phylogenetic placement of the genus Jasus and the Juan Fernández rock lobster Jasus frontalis. The phylogenetic tree is based on the phylogenetic signal provided by translated protein coding genes. Numbers above branches near nodes represent bootstrap pseudoreplicates (N = 1,000) of the tree search

Overall, our analysis is in line with conclusion from previous studies examining phylogenetic relationships within the Achelata that used either a smaller species set and entire mitochondrial genomes [11, 50, 51] or a larger species set with a reduced count of genetic markers [70]. This study also confirms the notion that the information contained in mitochondrial PCGs can be used to resolve, reliably, natural relationships at and below the family and superfamily level in the Achelata, as shown by previous phylomitogenomic studies focusing in other infraorders of decapod crustaceans, including the Caridea [71] and Anomura [72], among others.

Conclusion

In this study, we developed genomic resources for the rock lobster J. frontalis, an economically important species in oceanic islands from the southeastern Pacific Ocean that is likely overfished. We used a low-pass (= low-coverage) Illumina PE short-read dataset to estimate nuclear genome size in J. frontalis, and to identify, classify, and quantify nuclear transposable elements in the studied lobster. Notably, the abundance of transposable elements in the nuclear genome of J. frontalis implies that the assembly of a chromosome-level nuclear genome in this species will require long-reads (i.e., Pacific Biosciences or Oxford Nanopore Technology) as well as chromosome conformation capture techniques. We also assembled the mitochondrial genome of the studied species and examined its position in the infraorder Achelata based on the phylogenetic signal provided by mitochondrial PCGs. Conservation initiatives and fisheries management focusing on the Juan Fernández rock lobster will be supported by the genomic resources produced in this study. For instance, the newly assembled mitochondrial genome can be used as a reference to estimate the abundance of this species indirectly and non-intrusively; using environmental DNA.

Abbreviations

Ala

Alanine amino acid

Arg

Arginine amino acid

bv

Bootstrap value

cm

Centimeters

cob

Cytochrome b subunit of the cytochrome bc1 complex

cox

Subunits of the cytochrome c oxidase

CR

Control Region

d_N

Non-synonymous substituions

d_S

Synonymous substitutions

Gln

Glutamine aminoacid

H

Heavy strand

Ka

Non-synonymous substitutions per non-synonymous site

Ks

Synonymous substitutions per synonymous site

Leu

Leucine aminoacid

Lys

Lysine amino acid

m

Meters

nad

Subunits of the NADH dehydrogenase complex

NCBI

National Center for Biotechnology Information

PCG’s

Protein coding genes

PE

Paired-end

rRNA

Ribosomal genes

rrnL

16 S ribosomal RNA

rrnS

12 S ribosomal RNA

RSCU

Relative Synonymous Codon Usage

Ser

Serine amino acid

Thr

Threonine amino acid

tRNA

Transfer RNA genes

Trp

Tryptophan amino acid

Val

Valine aminoacid

ω = d_N/d_S = Ka/Ks

The ratio of non-synonymous to synonymous substitutions

Authors’ contributions

JAB conceived and designed the experiments. JAB, SAK, AB performed the experiments, analyzed the data, prepared figures and/or tables, authored or reviewed drafts of the paper, and approved the final draft. VPR provided resources, authored or reviewed drafts of the paper, and approved the final draft.

Funding

The authors received no funding for this work.

Data availability

Sequences described here are accessible via GenBank: PQ869192, BioProject ID: PRJNA780953.

Declarations

Ethics approval and consent to participate

All methods were performed in accordance with the relevant guidelines and regulations.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.