Skip to main content
NCBI home page
BMC Genomics logoLink to BMC Genomics
. 2006 Nov 6;7:284. doi: 10.1186/1471-2164-7-284

The complete mitochondrial genome of the sea spider Nymphon gracile (Arthropoda: Pycnogonida)

Lars Podsiadlowski 1,✉, Anke Braband 2
PMCID: PMC1636051  PMID: 17087824

Abstract

Background

Mitochondrial genomes form units of genetic information replicating indepentently from nuclear genomes. Sequence data (most often from protein-coding genes) and other features (gene order, RNA secondary structure) of mitochondrial genomes are often used in phylogenetic studies of metazoan animals from population to phylum level. Pycnogonids are primarily marine arthropods, often considered closely related to chelicerates (spiders, scorpions and allies). However, due to their aberrant morphology and to controversial results from molecular studies, their phylogenetic position is still under debate.

Results

This is the first report of a complete mitochondrial genome sequence from a sea spider (Nymphon gracile, class Pycnogonida). Gene order derives from that of other arthropods so that presumably 10 single tRNA gene translocations, a translocation of the mitochondrial control region, and one large inversion affecting protein-coding genes must have happened in the lineage leading to Nymphon gracile. Some of the changes in gene order seem not to be common to all pycnogonids, as those were not found in a partial mitochondrial genome of another species, Endeis spinosa. Four transfer RNAs of Nymphon gracile show derivations from the usual cloverleaf secondary structure (truncation or loss of an arm). Initial phylogenetic analyses using mitochondrial protein-coding gene sequences placed Pycnogonida as sister group to Acari. However, this is in contrast to the majority of all other studies using nuclear genes and/or morphology and was not recovered in a second analysis where two long-branching acarid species were omitted.

Conclusion

Extensive gene rearrangement characterizes the mitochondrial genome of Nymphon gracile. At least some of the events leading to this derived gene order happened after the split of pycnogonid subtaxa. Nucleotide and amino acid frequencies show strong differences between chelicerate taxa, presumably biasing phylogenetic analyses. Thus the affinities between Pycnogonida and Acari (mites and ticks), as found in phylogenetic analyses using mitochondrial genes, may rather be due to long-branch attraction and independently derived nucleotide composition and amino acid frequency, than to a real sister group relationship.

Background

Due to their evolutionary history as derived endosymbionts, mitochondria have retained genetic material – the mitochondrial genome. Much of their original gene content was eliminated or transferred to the nucleus [1], while only a small proportion of genes has persisted to the present. In triploblastic animals the circular mitochondrial genome is sized around 11–20 kilobases and contains typically 37 genes: 13 protein-coding genes, two ribosomal RNA genes and 22 transfer RNA genes [2]. Mitochondrial genomes serve as a simple model for modes and mechanisms of gene rearrangements and genome evolution and provide large datasets for phylogenetic analyses. The frequent use of mitochondrial genes for inferring phylogenetic relationships of animals is due to their universal distribution among taxa, strongly conserved regions in some genes (facilitating universal PCR primer sets) and the absence of paralog genes [3]. However, the incidental presence of nuclear copies of mitochondrial genes [4] and strong differences in nucleotide composition between taxa [5] may complicate phylogenetic analyses.

During the last ten years mitochondrial genome data have played an important role in redefining arthropod relationships. The position of mitochondrial trnL2 is changed in crustaceans and hexapods, but not in chelicerates and myriapods [6]. Also from sequence-based analyses of mitochondrial [7,8] and nuclear genes [9-11] the Pancrustacea hypothesis found strong support, while the traditional Tracheata hypothesis, mainly based on morphological data, is now widely rejected. Mitochondrial genome data also provided strong evidence towards the identification of the formerly enigmatic Pentastomida (tongue worms) as aberrant crustaceans [12]. While those hypotheses were collectively supported by nuclear and mitochondrial data, some other hypotheses obtained with mitochondrial genome data are highly disputed, as for example the polyphyly of hexapods [7,8] or the phylogenetic position of pycnogonids [13].

Pycnogonids or sea spiders are among the most bizarre arthropods, some of them with very large legs attached on a tiny body. Food uptake is performed by a pharyngeal suction tube, some species have even lost all head appendages (chelifores and pedipalps). Due to their derived morphology their phylogenetic position remains uncertain, although most workers consider them as primarily aquatic chelicerates [14]. Recent phylogenetic analyses using a combination of molecular and morphological data [11], or nuclear genes [9,10] support a basal position among chelicerates. In contrast, sequence data from partial mitochondrial genomes suggest an affinity to Acari (mites and ticks) [13], thus implying a terrestrial origin of pycnogonids. Recently, neuroanatomical data suggest that pycnognid chelifores are not positionally homologous to cheliceres [15], thus questioning pycnogonid affinities to Euchelicerata. However, hox gene expression data do not support the this view [16]. We report here the first complete mitochondrial genome sequence for a member of the Pycnogonida,Nymphon gracile. We use these data to analyse chelicerate relationships and to evaluate hypotheses of the phylogenetic position of Pycnogonida. We also discuss ancestral and derived features of the mitochondrial genome of Nymphon gracile and the influence of AT-content and differences of amino acid frequencies on phylogenetic analyses.

Results and discussion

Mitochondrial genome organization

The mitochondrial genome of Nymphon gracile is a circular DNA molecule of 14,681 bp length [GenBank:DQ666063]. All 37 genes expected for animal mitochondrial genomes have been identified. Gene overlaps (7 bp) exist between nad4 and nad4L, as well as between atp8 and atp6, as is reported for many other mitochondrial genomes. Six out of thirteen protein-coding genes show incomplete stop codons (T or TA), which is probably compensated by posttranscriptional polyadenylation [17].

Gene order (Fig. 1, Tab. 1) of the mitochondrial genome differs in many positions from that of the horseshoe crab Limulus polyphemus, which is considered to represent the euarthropod ground pattern [18]. One large segment (about 3,500 bp) containing the protein-coding genes cox1, cox2, nad2 and the transfer RNA genes trnC, trnW, trnI were probably subject to an inversion. Maintaining their original gene order, these genes were found on the opposite strand in Nymphon gracile compared to other arthropods. Probably trnK and trnD were also involved in the same inversion, meanwhile separated from cox2 by six tRNA genes, which have different positions in other arthropods. Further translocation events of single tRNA genes may have led to their actual position in Nymphon gracile. Three tRNAs were lost from the inverted segment (trnQ, trnM, trnY) and translocated to other positions in the mitochondrial genome. Two tRNA genes (trnE and trnR) have interchanged their positions, making it impossible to decide which of them was translocated and which stayed in its position. Supposing that trnK and trnD have changed their position due to the large inversion mentioned above, altogether ten tRNA genes have undergone individual translocation events. Except from those involved in the inversion, no other protein-coding or rRNA gene has changed its position compared to the ground pattern of Euarthropoda, represented by Limulus polyphemus [18]. The two trnL genes lie adjacent to another between nad1 and rrnL, as expected for the euarthropod ground pattern. This differs from the derived condition in Hexapoda and Crustacea (trnL2 between cox1 and cox2, trnL1 between nad1 and rrnL).

Figure 1.

Figure 1

Open in a new tab

Gene order in pycnogonids (Nymphon gracile, this study; Endeis spinosa, according to [13]). Comparison to the ground pattern of Euarthropoda (here represented by Limulus polyphemus and Heptathela hangzhouensis). Asterisks indicate that a gene is located on the opposite strand. Transfer RNA genes are depicted by their corresponding one-letter amino acid code. Colored genes have derived relative positions in the derived gene order of Nymphon gracile. Lines refer to putative independent translocation events.

Table 1.

Annotation of the mitochondrial genome of Nymphon gracile

Gene Strand Position number Size(bp) Size (aa) Start codon Stop codon Intergenic nucleotides
trnY - 1–67 67 - - - 3
trnF - 71–134 64 - - - 0
nad5 - 135–1809 1675 558 ATA T 4
trnH - 1814–1876 63 - - - 2
nad4 - 1879–3261 1383 460 ATG TAA -7
nad4L - 3255–3535 281 93 ATG TA 13
trnT + 3549–3609 61 - - - 0
nad6 + 3610–4045 436 145 ATA T 12
cob + 4058–5180 1123 374 ATT T 0
trnS2 + 5181–5244 64 - - - 8
nad1 - 5253–6167 915 304 ATG TAA 0
trnL2 - 6169–6230 63 - - - 7
trnL1 - 6238–6299 62 - - - *
rrnL - 6300–7495 1196 - - - *
trnN - 7496–7556 61 - - - 6
trnQ - 7563–7629 66 - - - *
rrnS - 7630–8396 768 - - - *
trnD - 8397–8460 64 - - - 13
trnK - 8474–8540 67 - - - 8
trnG - 8549–8611 63 - - - 0
trnP - 8612–8674 63 - - - 4
trnM - 8679–8743 65 - - - 9
trnV + 8753–8820 68 - - - 2
trnA - 8823–8882 60 - - - 11
trnS1 + 8894–8952 59 - - - 30
cox2 - 8983–9667 685 226 ATG T 31
cox1 - 9699–11237 1539 512 ATA TAA 15
trnC + 11253–11314 62 - - - 1
trnW - 11316–11385 70 - - - -2
nad2 - 11394–12367 984 327 ATG TAA 1
trnI - 12369–12432 64 - - - 15
atp8 + 12448–12598 159 52 ATT TAA -7
atp6 + 12600–13214 615 204 ATG TAA 1
cox3 + 13216–14004 789 262 ATA TAA 17
nad3 + 14022–14367 346 115 ATC T 0
trnE + 14368–14434 67 - - - 0
trnR + 14435–14490 56 - - - 192
CR 14491–14681 192 - - - -
Open in a new tab

*Gene boundaries of rRNA genes determined by sequence of adjacent genes.

A large non-coding region is present between nad3-trnE-trnR and trnY-trnF-nad5. This is very likely to be the mitochondrial control region, which therefore must have been translocated, too. As the strand bias of nucleotide frequencies is comparable to other arthropods (see below) the control region seems not to be inverted as it is assumed for scorpions and two web spider species [13].

A partial (5105 bp) mitochondrial genome of another species of Pycnogonida, Endeis spinosa [GenBank:AY731173], was published recently [5,13] and revealed no differences in gene order to Limulus polyphemus in the segment ranging from nad2 to cox3. Therefore we presume that the large inversion recorded in Nymphon gracile must have happend after the split between the two clades. Assuming a larger taxon sampling the derived gene order of Nymphon gracile may serve as an apomorphic character supporting a subtaxon of Pycnogonida. Six of the ten individually translocated tRNA genes of Nymphon gracile found their new position between trnK and cox2. In Endeis spinosa these two genes are adjacent, therefore these six tRNA translocations may also have happened after the split of the two clades.

Secondary structure of RNAs

Out of 22 transfer RNAs usually present in metazoan mitochondrial genomes we have identified 18 by tRNA-scan [19]. The remaining four show derivations from the typical cloverleaf structure: the DHU stem and loop is extremely short or missing in trnA, trnN and trnS1, while it is definitely missing in trnR (putative secondary structures are shown in Fig. 2). Such aberrant secondary structures are often found in metazoan mitochondrial tRNAs, some taxa even show derivations in the majority of their tRNA genes (e.g. some nematodes [20], or web spiders [21,22]). It is not clear if the function of such derived tRNAs is maintained in every case, as there are reports of recruitment of nuclear tRNAs into mitochondria [1]. In spiders however, it is very likely that tRNAs are functional, despite lacking the TΨC stem and loop [23]. trnS1 from the horseshoe crab Limulus polyphemus is also missing the complete DHU stem and loop, while all other tRNAs in this species could be folded into cloverleaf structures [24].

Figure 2.

Figure 2

Open in a new tab

Putative secondary structures of mitochondrial tRNA molecules from Nymphon gracile.

Nucleotide composition

Nucleotide compositions of protein-coding and ribosomal RNA genes clearly demonstrate a strand specific bias (Tab. 2). CG-skew is positive for all genes on (+)-strand and negative for genes on (-)-strand, AT-skew is positive or only slightly negative in (+)-strand genes, while strongly negative in (-)-strand genes. This is also true for genes which have changed from one strand to the other due to an inversion during pycnogonid evolution (cox1, cox2, nad2). Strand bias in nucleotide composition is probably due to asymmetries during the replication of mitochondrial genomes, leading to different mutational pressures on both strands. In the literature the exact mechanisms are controversially discussed: one strand stays single-stranded during replication in the strand-displacement model [25], or is subject to extensive incorporation of ribonucleotides in the strand-coupled model [26]. Almost all arthropods show this strand bias [5,13]. In some species a reversal is seen, probably due to a strand swap of the mitochondrial control region (e.g. in scorpions and web spiders [5], as well as in an isopod [27]). As seen from the pycnogonid and spider examples, strand reversal of single genes leads to a quick reversal of strand bias, too. In performing phylogenetic analyses one has to take into account these findings and probably has to modify evolutionary models [5].

Table 2.

Nucleotide composition and skews of Nymphon gracile mitochondrial protein-coding and ribosomal RNA genes.

Gene (+/- strand) Proportion of nucleotides %AT AT skew CG skew
A C G T
atp6 (+) 0.382 0.145 0.073 0.400 78.2 -0.023 0.328
atp8 (+) 0.510 0.116 0.000 0.374 88.4 0.154 1.000
cox1 (-) 0.263 0.123 0.169 0.446 70.8 -0.259 -0.158
cox2 (-) 0.308 0.123 0.143 0.426 73.4 -0.161 -0.077
cox3 (+) 0.373 0.150 0.120 0.357 73.0 0.021 0.108
cob (+) 0.355 0.144 0.107 0.394 74.9 -0.051 0.149
nad1 (-) 0.256 0.079 0.163 0.503 75.8 -0.326 -0.348
nad2 (-) 0.318 0.070 0.116 0.496 81.4 -0.218 -0.246
nad3 (+) 0.387 0.139 0.067 0.408 79.5 -0.025 0.352
nad4 (-) 0.256 0.075 0.136 0.533 78.9 -0.351 -0.288
nad4L (-) 0.278 0.053 0.178 0.491 76.9 -0.278 -0.538
nad5 (-) 0.247 0.080 0.133 0.540 78.7 -0.372 -0.249
nad6 (+) 0.371 0.153 0.048 0.428 80.0 -0.071 0.523
rrnl(-) 0.347 0.074 0.134 0.446 79.3 -0.125 -0.288
rrns (-) 0.350 0.078 0.167 0.405 75.5 -0.073 -0.363
total (+) 0.454 0.141 0.085 0.319 77.3 0.175 0.248
Open in a new tab

AT skew ((A%-T%)/(A%+T%)) and CG skew ((C%-G%)/(C%+G%)) according to [44]. Values from (-)strand genes in bold letters.

Phylogenetic analysis

Phylogenetic analysis using nucleotide sequences from all mitochondrial protein-coding genes (Fig. 3) reveals a strong support for a taxon comprised of Nymphon gracile and Acari (mites and ticks) from BI but not from ML bootstrap analysis. Similar results were published based on an analysis of five mitochondrial protein-coding genes [13]. This contradicts almost every other phylogenetic study which included pycnogonids (the only recent exception not based on mitochondrial genes is a combined analysis of 18S/28S sequences and morphological data, including fossils [28], where Pycnogonida, Acari and Palpigradi together form one clade – but with extreme character conflict). Recent multigene analyses place pycnogonids outside Arachnida: either (1) as sister group to Euchelicerata (Xiphosura + Arachnida), e.g. with a combined alignment of EF-1α, EF-2, and RNA-Pol II [10], as well as with a combined dataset of nine genes and morphology [11]; or (2) in an unresolved trichotomy with Euchelicerata and Myriapoda (together forming the clade Paradoxopoda or Myriochelata), e.g. using a combined aligment of 18S and 28S [9].

Figure 3.

Figure 3

Open in a new tab

Maximum likelihood tree of chelicerate relationships – complete dataset. According to a nucleotide alignment (first and second codon positions) from 13 protein-coding genes. Numbers above branches indicate Bayesian posterior probabilities (upper) and bootstrap percentages of maximum likelihood analysis (lower). Branch lengths reflect substitutions per site.

As a consequence of an Acari-Pycnogonid clade one would have to assume that pycnogonids may either have had terrestrial ancestors, or that an independent transition to terrestrial life was undertaken by Acari on the one hand, and spiders and scorpions on the other hand (leaving the question open where to place the remaining arachnid taxa). At least the first hypothesis is contradicted by the fossil record: the oldest pycnogonid fossil is from the upper Cambrian, a time from which no terrestrial animal is known [29] – the first terrestrial arachnids were not found before the Silurian – a gap of about 70 million years.

To determine effects of long-branch attraction [30], a second analysis was performed, on a dataset where sequences from the long-branching acarids Varroa destructor and Leptotrombidium pallidum were omitted (Fig. 4). In the resulting tree Pycnogonida does not cluster with Acari, rather appears as sister taxon to Euchelicerata in the ML tree, but without good support from BI or ML bootstrap analysis. Chelicerata, Euchelicerata and Arachnida as well find no good support from these inferences. So it is very likely that long-branch attraction is one major reason for the clustering of Pycnogonida and Acari in phylogenetic analysis of chelicerate relationships with mitochondrial genes, while the true position of Pycnogonida remains far from being resolved by our analyses.

Figure 4.

Figure 4

Open in a new tab

Maximum likelihood tree of chelicerate relationships – reduced dataset. Alignment without the long-branching acarid species Varroa destructor and Leptotrombidium pallidum. Details of analyses and legends as in Fig. 3.

AT content and amino acid usage in chelicerate mitochondrial genomes

As mentioned above, some of the arachnid taxa show a reversal in nucleotide frequency bias (the scorpion Centruroides limpidus, the spiders Ornithoctonus huwena and Habronattus oregonensis, and the mite Varroa jacobsohni), which may be one reason for misleading results in a phylogenetic analysis [5,13]. However, it was shown before that a reversal in strand bias seems not to be the cause for the affinity between Acari and Pycnogonida [13].

With the exception of Limulus polyphemus all chelicerates in our phylogenetic analysis show branch lengths two or three times longer than those of hexapods, crustaceans and myriapods (Fig. 3, Fig. 4). This implies that arachnid and pycnogonid species have undergone more change in nucleotide sequence than Limulus polyphemus and the remainder of arthropod taxa in our study. With the exception of the scorpion Centruroides limpidus all arachnid and pycnogonid species in our study show a higher AT content in protein-coding genes than Limulus polyphemus and the outgroup taxa (Tab. 3). This is more striking if only third codon positions are compared. Strong variation in AT content between species may also lead to perturbance of phylogenetic analyses [31]. Independent evolution of higher AT content my lead to homoplastic similarities.

Table 3.

Amino acid usage and AT content of mitochondrial protein-coding genes from various arthropods.

Taxon Species Ala Arg Asn Asp Cys Gln Glu Gly His Ile Leu Lys Met Phe Pro Ser Thr Trp Tyr Val AT% AT%
GCN CGN AAY GAY TGY CAR GAR GGN CAY ATY TTR AAR ATR TTY CCN TCN ACN TGR TAY GTN PCG PCG
CTN AGN total 3rd pos.
Acari Ornithodoros mou. 123 51 163 57 33 66 87 207 71 386 511 115 288 368 131 318 186 99 130 209 71,30% 78,50%
" Rhipicephalus san. 117 44 206 58 34 56 79 163 65 462 487 136 318 394 120 355 142 80 129 139 77,90% 89,50%
" Varroa destructor 100 48 210 54 29 39 90 160 65 412 507 118 350 367 113 317 123 82 189 163 79,20% 91,70%
" Amblyomma trig. 103 41 200 52 31 51 83 161 63 494 469 141 311 436 106 352 154 80 130 130 78,30% 88,40%
" Haemaphysalis fl. 109 42 190 52 33 48 81 163 69 488 476 137 328 389 113 362 140 82 135 147 76,60% 85,50%
" Carios capensis 119 50 158 63 32 59 81 207 69 414 536 120 293 364 139 339 150 96 131 172 72,50% 79,90%
" Ixodes hexagonus 115 48 161 64 32 49 79 181 67 436 495 118 296 385 137 385 140 95 121 182 71,10% 75,70%
" Leptotrombidium p. 137 39 100 62 26 65 100 206 69 360 503 147 194 414 131 432 131 88 73 151 71,40% 83,10%
Araneae Habronattus oreg. 138 52 152 67 24 48 89 206 67 362 494 84 338 317 116 386 134 95 158 234 73,80% 86,10%
" Ornithoctonus huw. 147 53 109 74 23 48 96 213 73 358 469 97 273 355 134 402 139 99 127 260 69,70% 78,40%
" Heptathela han. 135 53 136 54 29 64 90 218 73 369 561 100 274 344 140 375 151 100 114 192 71,40% 82,10%
Scorpiones Centruroides lim. 190 63 93 65 43 46 86 253 77 250 604 79 184 355 145 376 176 103 120 298 62,90% 68,10%
Pycnogonida Nymphon gracile 125 53 160 69 35 46 80 188 71 362 494 123 286 420 136 370 158 82 139 218 76,60% 90,50%
Xiphosura Limulus polyph. 176 62 147 59 49 67 86 237 78 346 565 83 212 330 152 387 178 110 121 221 66,30% 74,70%
Myriapoda Narceus ann. 205 50 110 66 31 80 74 276 75 298 606 68 196 276 162 342 206 113 128 283 62,10% 67,90%
" Lithobius for. 193 60 125 70 34 68 76 256 74 331 557 85 246 306 137 357 199 102 140 239 65,70% 72,00%
Crustacea Squilla mantis 225 62 132 74 42 73 82 259 80 272 560 83 235 317 136 324 210 97 152 243 68,10% 79,30%
" Triops can. 207 61 164 69 32 79 78 232 81 316 627 73 199 319 138 351 188 107 137 220 68,20% 77,70%
Hexapoda Tricholepidion ger. 184 58 148 72 44 73 76 245 80 324 564 74 234 310 144 376 223 104 147 226 67,60% 76,90%
Open in a new tab

Bold numbers indicate strong differences (+/-25%) to Limulus polyphemus (underlined). This species was chosen for comparison, because in the phylogenetic analyses (Fig. 3, 4) it shows a branch length comparable to those of crustacean, myriapod and hexapod species, while the remainder of chelicerates was subject to a higher degree of nucleotide substitution. See methods for complete names of species.

Looking at the amino acid composition of the protein-coding genes from chelicerates and other arthropods (Tab. 3), again strong differences are observed between Limulus polyphemus and the remainder of Chelicerata. Compared to mitochondrial proteins from Limulus polyphemus some amino acids are significantly less used in the pycnogonid species and most of the arachnids (Ala, Arg, Cys, Gln, Thr, Trp), while others are significantly more frequently used (Lys, Met). With a few exceptions, less used amino acids are coded by GC rich codons (Ala, Arg, Gly, Pro), while those more often used are coded by AT rich codons (Ile, Lys, Met, Phe, Asn). A similar effect is seen in a comparison of codon usage for the amino acids leucine and serine (Tab. 4). For the coding of leucine, UUR-codons are more frequently used in all chelicerate taxa than CUN-codons, but the difference is by far higher in those taxa, which show the highest AT-contents (The acarid taxa Haemaphysalis flava, Amblyomma triguttatum, Varroa destructor, Rhipicephalus sanguineus, and the pycnogonid Nymphon gracile). In contrast AGN-codons and UCN-codons for serine show only moderate variance between the taxa.

Table 4.

Codon usage for leucine and serine codons of chelicerate arthropods. See Fig. 3 for full species names.

Codon Amino acid Xiph. Pycn. Scor. Araneae Acari
L.pol. N.g. C.l. H.h. O.h. H.o. L.pal. I.h. C.c. H.f. A.t. V.d. R.s. O.m.
UUA L 226 363 154 321 212 301 245 265 314 314 328 380 355 242
UUG L 84 51 163 49 118 70 44 52 56 45 28 29 24 84
UUR L 310 414 317 370 330 371 289 317 370 359 356 409 379 326
CUA L 94 27 87 81 62 81 59 70 73 50 33 40 42 70
CUC L 55 1 62 16 13 8 28 37 30 9 10 6 6 16
CUG L 4 2 28 16 7 4 7 7 2 7 1 1 1 5
CUU L 106 43 110 78 57 31 120 67 64 52 70 58 59 95
CUN L 259 73 287 191 139 124 214 181 169 118 114 105 108 186
AGA S 66 67 40 83 99 91 89 64 63 74 60 84 72 71
AGC S 8 1 28 8 5 6 2 2 12 5 3 6 2 8
AGG S 12 18 17 1 12 7 7 15 2 7 6 2 7 10
AGU S 22 45 44 22 16 30 29 32 30 26 22 61 30 29
AGN S 108 131 129 114 132 134 127 113 107 112 91 153 111 118
UCA S 98 60 43 85 97 95 101 99 100 119 127 39 129 82
UCC S 65 8 47 49 40 31 49 58 46 26 28 16 25 23
UCG S 6 7 12 7 15 8 14 6 4 8 8 4 1 8
UCU S 110 163 145 121 119 118 141 110 86 98 98 108 89 89
UCN S 279 238 247 262 271 252 305 273 236 251 261 167 244 202
Open in a new tab

Thus the noticed derivations in amino acid usage seem to be directly linked to AT-content: the higher the proportion of adenine and thymine, the stronger the differences in amino acid usage. And in fact, for some amino acids Nymphon gracile together with some Acari (the taxa showing the highest AT contents as mentioned above) show the strongest differences to Limulus polyphemus (Ala, Gly, Phe, Trp). This fact may have further promoted the clustering of Acari and Nymphon gracile in our first phylogenetic analysis (Fig. 3) and in [13]. In contrast, a web spider (Habronattus oregonensis) which also shows high AT content does not cluster with Acari and Nymphon gracile, probably due to a balancing effect of the other two web spider species, which show a comparably moderate AT content.

Conclusion

Ten individually translocated tRNA genes, a large inversion of a segment covering three protein-coding genes and five tRNA genes, and translocation of the control region lead to a derived gene order in Nymphon gracile compared to other arthropods, including another species of Pycnogonida (Endeis spinosa). If sequence data from more pycnogonid species becomes available, gene translocations may serve as phylogenetic markers, which probably resolve relationships between pycnogonid subtaxa. Phylogenetic analysis of chelicerate relationships using mitochondrial protein-coding genes supported a clade consisting of Pycnogonida and Acari. These results contradict other analyses performed with nuclear genes or morphological characters. Omiting some of the long-branching acarid species from the analysis led to a tree with unresolved relationships between Myriapoda, Pycnogonida, Xiphosura and two arachnid clades. We hypothesize that phylogenetic analyses of chelicerate interrelationships based on mitochondrial protein-coding genes is biased by three misleading factors: (a) long-branch attraction, (b) derived AT bias in all chelicerate taxa except Limulus polyphemus, and (c) reversed strand bias in Scorpiones, two species of Araneae and the mite Varroa destructor. Thus from a mitogenomic point of view the exact phylogenetic position of Pycnogonida remains an open question, but a sister group relationship between Acari and Pycnogonida as suggested by Hassanin [13] is rather caused by long-branch attraction and higher AT content than to an underlying phylogenetic signal.

Methods

Samples and DNA extraction

Specimens of Nymphon gracile were sampled at Concarneau (France) and immediately preserved in pure ethanol (99.8%). DNA from the legs of a single specimen was extracted using the DNeasy Kit (Qiagen, Germany) following the manufacturers protocol.

PCR, sequencing and gene annotation

Fragments of six mitochondrial genes (cox1, nad4, nad5, rrnl, rrns) were PCR amplified using primers especially designed for this purpose. Primer sequences were as follows: cox1f: 5'-ACTAATCACA ARGAYATTGG-3'/cox1r: 5'-TAGTCTGAGT ANCGTCGWGG-3' (annealing temperature: 45°C); nad4f: 5'-TTGAGGTTAY CAGCCYG-3'/nad4r: 5'-ATATGAGCYA CAGAAGARTA AGC-3' (45°C); nad5f: 5'-AGAATTCACT AGGDTGRGAT GG-3'/nad5r: 5'-AAAGAGCCTT AAATAAAGCA TG-3' (45°C); 16Sf: 5'-GCGACCTCGA TGTTGGATTA A-3'/16Sr: 5'-CCGGTCTGAA CTCAYATC-3' (48°C); 12Sf: 5'-CAGCAKYCGC GGTTAKAC-3'/12Sr: 5'-ACACCTACTW TGTTACGACT TATCTC-3' (52°C). Primer design was performed on conserved regions of alignments using mitochondrial genes of various arthropod species [32]. All PCR experiments were done on Mastercycler and Mastercycler gradient (Eppendorf, Hamburg, Germany) using the HotMasterTaq Kit (Eppendorf). PCR reaction volumes were 50 μl (42 μl sterilized destilled water, 5 μl 10× reaction buffer 1μl dNTP mix, 1 μl primer mix (10 μM each), 1 μl DNA template, 0,2 μl = 1 u HotMasterTaq polymerase). Cycling protocol includes an initial denaturation step for 2 min at 94°C, 40 cycles of 30 sec at 94°C, 1 min at the appropriate annealing temperature (see above) and 90 sec at 68°C; in the end a final extension step for 1 min at 68°C. PCR products were gel purified (Qiaquick Gel purification kit, Qiagen, Hilden, Germany) and subsequently used for sequencing. Sequencing reactions were done using the DCTS quick start kit (Beckman-Coulter) and the CEQ 8000 capillary sequencer (Beckman-Coulter). Sequence information from these PCR fragments was used to design PCR primer pairs to amplify missing parts between them. The PCR protocol described above was therefore modified with extension steps of 7 min, and a final extension step of 3 min.

Successful amplification of PCR products was obtained with the following primer pairs (annealing temperature and approximate length in brackets): Ng-12sr: 5'-AAAAAGAATA CTAGGGTCTC TAATC C-3'/Ng-cox1f: 5'- AGCGGGTTTT ACTAATTGGT ATCC-3' (54°C, 2000 bp); Ng-cox1r: 5'- AAGAAGTTAC TAACAATATT AAAGCAGGAG G-3'/Ng-nad5f: 5'-TTTAACTATA TTCTTAGCTA GAGTATGTGC TTC-3' (58°C, 5000 bp); Ng-nad5r: 5'-ATAAAACATA AACCCCAGCA G-3'/Ng-nad4f: 5'-GATTATAGGT TGAGGAAAAT CTC-3' (49°C, 1700 bp); Ng-16sr: 5'-CGGTCTGAAC TCAGATCATG TAA-3'/Ng-12sf: 5'-TTAAAGGATA AGATGGGCTA C-3' (48°C, 1500 bp). In addition for amplification of the part spanning from nad4 to rrnl a primer pair already published in [33] was used successfully used: N4: 5'-GGAGCTTCAA CATGAGCTTT-3'/16S2: 5'-GCGACCTCGA TGTTGGATTA A-3' (50°C, 3900 bp). Sequencing of PCR products larger than 1000 bp was done using a primer walking strategy.

Primary analysis of nucleotide sequences was done using the Beckman CEQ 8000 software. Sequences were then aligned and assembled using Bioedit [34]. Protein-coding genes and ribosomal RNA genes were identified by blasting on NCBI entrez databases and by comparison with other arthropod mitochondrial genomes. Transfer RNA genes were identified using tRNAscan-SE 1.21 [19] and DOGMA [35].

Phylogenetic analysis

Phylogenetic analysis was performed using nucleotide sequences from mitchondrial protein-coding genes. Amino acid sequences from single genes were aligned by Clustal X [36] with default settings. After retranslation to nucleotides, ambigously aligned parts were omitted from the analysis by making use of Gblocks 0.91b [37], using the "codons" option and default block parameters. Due to the results of a saturation analysis [38] on single codon positions, implemented in DAMBE, version 4.2.13 [39], third codon positions were eliminated from the alignment. The final alignment consisted of 5,711 bp for 20 taxa. A second alignment was obtained by the same procedure but omitting the two species of Acari with longest branches (Varroa destructor and Leptotrombidium pallidum), leading to an alignment consisting of 5,996 bp for 18 taxa.

Two different analyses were performed on both alignments. (1) A maximum likelihood tree was computed using PAUP* ver. 4.0b10 [40]. The model (GTR+I+gamma) and model parameters were chosen according to the AIC with modeltest 3.7 [41]. In addition 100 bootstrap replicates were performed. (2) Bayesian analysis was performed with MrBayes 3.1.2 [42]. 1,000,000 generations were run under the GTR model, with gamma distribution and a proportion of invariant sites. The first 100 out of 1000 trees were discarded as burn-in. Mitochondrial genome data from other species than Nymphon gracile was obtained from the OGRE database [43]. NCBI GenBank accession numbers: Squilla mantis, [Genbank NC_006081]; Triops cancriformis, [Genbank:NC_004465]; Tricholepidion gertschi, [Genbank:NC_005437]; Locusta migratoria, [Genbank:NC_001712]; Narceus annularius, [Genbank:NC_003343]; Lithobius forficatus, [Genbank:NC_002629]; Limulus polyphemus, [Genbank:NC_003057]; Centruroides limpidus, [Genbank:NC_006896]; Heptathela hangzhouensis, [Genbank:NC_005924]; Ornithoctonus huwena, [Genbank:NC_005925]; Habronattus oregonensis, [Genbank:NC_005942]; Leptotrombidium pallidum, [Genbank:NC_007601]; Varroa destructor, [Genbank:NC_004454]; Ornithodoros moubata, [Genbank:NC_004357]; Carios capensis, [Genbank:NC_005291]; Ixodes hexagonus, [Genbank:NC_002010]; Haemaphysalis flava, [Genbank:NC_005292]; Rhipicephalus sanguineus, [Genbank:NC_002074]; Amblyomma triguttatum, [Genbank:NC_005963].

Abbreviations

Mitochondrial genes: atp6, atp8, ATP synthase subunits 6 and 8; cob, cytochrome oxidase b; cox1-3, cytochrome oxidase subunit I-III; nad1-6, nad4L, NADH dehydrogenase subunits 1–6, 4L; rrns, rrnl, small (12S) and large (16S) subunit ribosomal RNA; transfer RNA (tRNA) genes are listed as trnX, where X is replaced by the one letter amino acid code of the corresponding amino acid; CR, mitochondrial control region. EF-1α/EF-2, elongation factor-1α/-2; RNA-Pol II, RNA polymerase II. BI, Bayesian inference; ML, maximum likelihood; bp, base pairs.

Authors' contributions

LP was primarily responsible for design and coordination of the study and conducted all the laboratory work. Analyses and manuscript draft was done in equal parts by both authors.

Acknowledgments

Acknowledgements

We thank T. Bartolomaeus (Freie Universität Berlin) and G. Scholtz (Humboldt Universität Berlin) for lab space and support of our study through all stages. Many thanks to C. Bleidorn (Universität Potsdam) and L. Vogt (Freie Universität Berlin) for helpful discussions regarding methodology and interpretation of results, as well as to two anonymous reviewers for improvement of the manuscript. This study was supported by German science foundation grants DFG Ba 1520/10-1 and DFG Scho 442/8-2 (SPP Deep Metazoan Phylogeny).

Contributor Information

Lars Podsiadlowski, Email: lars@podsiadlowski.de.

Anke Braband, Email: phylogenetics@t-online.de.

References

  1. Adams KL, Palmer JD. Evolution of mitochondrial gene content: gene loss and transfer to the nucleus. Mol Phylogenet Evol. 2003;29:380–395. doi: 10.1016/S1055-7903(03)00194-5. [DOI] [PubMed] [Google Scholar]
  2. Wolstenholme DR. Genetic novelties in mitochondrial genomes of multicellular animals. Curr Opin Genet Dev. 1992;2:918–925. doi: 10.1016/S0959-437X(05)80116-9. [DOI] [PubMed] [Google Scholar]
  3. Moritz C, Dowling TE, Brown WM. Evolution of animal mitochondrial-DNA – relevance for population biology and systematics. Annu Rev Ecol Syst. 1987;18:269–292. doi: 10.1146/annurev.es.18.110187.001413. [DOI] [Google Scholar]
  4. Bensasson D, Zhang D, Hartl DL, Hewitt GM. Mitochondrial pseudogenes: evolution's misplaced witnesses. Trends Ecol Evol. 2001;16:314–321. doi: 10.1016/S0169-5347(01)02151-6. [DOI] [PubMed] [Google Scholar]
  5. Hassanin A, Leger N, Deutsch J. Evidence for multiple reversals of asymmetric mutational constraints during the evolution of the mitochondrial genome of Metazoa, and consequences for phylogenetic inferences. Syst Biol. 2005;54:277–298. doi: 10.1080/10635150590947843. [DOI] [PubMed] [Google Scholar]
  6. Boore JL, Lavrov DV, Brown WM. Gene translocation links insects and crustaceans. Nature. 1998;392:667–668. doi: 10.1038/33577. [DOI] [PubMed] [Google Scholar]
  7. Cook CE, Yue Q, Akam M. Mitochondrial genomes suggest that hexapods and crustaceans are mutually paraphyletic. Proc R Soc Lond B Biol Sci. 2005;272:1295–1304. doi: 10.1098/rspb.2004.3042. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Nardi F, Spinsanti G, Boore JL, Carapelli A, Dallai R, Frati F. Hexapod origins: Monophyletic or paraphyletic? Science. 2003;299:1887–1889. doi: 10.1126/science.1078607. [DOI] [PubMed] [Google Scholar]
  9. Mallatt JM, Garey JR, Shultz JW. Ecdysozoan phylogeny and Bayesian inference: first use of nearly complete 28S and 18S rRNA gene sequences to classify the arthropods and their kin. Mol Phylogenet Evol. 2004;31:178–191. doi: 10.1016/j.ympev.2003.07.013. [DOI] [PubMed] [Google Scholar]
  10. Regier JC, Shultz JW, Kambic RE. Pancrustacean phylogeny: exapods are terrestrial crustaceans and maxillopods are not monophyletic. Proc R Soc Lond B Biol Sci. 2005;272:395–401. doi: 10.1098/rspb.2004.2917. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Giribet G, Richter S, Edgecombe GD, Wheeler WC. The position of crustaceans within Arthropoda – evidence from nine molecular loci and morphology. Crustacean Issues. 2005;16:307–330. [Google Scholar]
  12. Lavrov DV, Brown WM, Boore JL. Phylogenetic position of the Pentastomida and (pan)crustacean relationships. Proc R Soc Lond B Biol Sci. 2004;271:537–544. doi: 10.1098/rspb.2003.2631. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Hassanin A. Phylogeny of Arthropoda inferred from mitochondrial sequences: strategies for limiting the misleading effects of multiple changes in pattern and rates of substitution. Mol Phylogenet Evol. 2006;38:100–116. doi: 10.1016/j.ympev.2005.09.012. [DOI] [PubMed] [Google Scholar]
  14. Dunlop JA, Arango CP. Pycnogonid affinities: a review. J Zool Syst Evol Res. 2005;43:8–21. doi: 10.1111/j.1439-0469.2004.00284.x. [DOI] [Google Scholar]
  15. Maxmen A, Browne WE, Martindale MQ, Giribet G. Neuroanatomy of sea spiders implies an appendicular origin of the protocerebral segment. Nature. 2005;437:1144–1148. doi: 10.1038/nature03984. [DOI] [PubMed] [Google Scholar]
  16. Jager M, Murienne J, Clabaut C, Deutsch J, Le GH, Manuel M. Homology of arthropod anterior appendages revealed by Hox gene expression in a sea spider. Nature. 2006;441:506–508. doi: 10.1038/nature04591. [DOI] [PubMed] [Google Scholar]
  17. Ojala D, Montoya J, Attardi G. tRNA punctuation model of RNA processing in human mitochondria. Nature. 1981;290:470–474. doi: 10.1038/290470a0. [DOI] [PubMed] [Google Scholar]
  18. Staton JL, Daehler LL, Brown WM. Mitochondrial gene arrangement of the horseshoe crab Limulus polyphemus L.: conservation of major features among arthropod classes. Mol Biol Evol. 1997;14:867–874. doi: 10.1093/oxfordjournals.molbev.a025828. [DOI] [PubMed] [Google Scholar]
  19. Lowe TM, Eddy SR. tRNAscan-SE: a program for improved detection of transfer RNA genes in genomic sequence. Nucleic Acids Res. 1997;25:955–964. doi: 10.1093/nar/25.5.955. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Wolstenholme DR, Macfarlane JL, Okimoto R, Clary DO, Wahleithner JA. Bizarre tRNAs inferred from DNA sequences of mitochondrial genomes of nematode worms. P Natl Acad Sci USA. 1987;84:1324–328. doi: 10.1073/pnas.84.5.1324. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Masta SE, Boore JL. The complete mitochondrial genome sequence of the spider Habronattus oregonensis reveals rearranged and extremely truncated tRNAs. Mol Biol Evol. 2004;21:893–902. doi: 10.1093/molbev/msh096. [DOI] [PubMed] [Google Scholar]
  22. Qiu Y, Song D, Zhou K, Sun H. The mitochondrial sequences of Heptathela hangzhouensis and Ornithoctonus huwena reveal unique gene arrangements and atypical tRNAs. J Mol Evol. 2005;60:57–71. doi: 10.1007/s00239-004-0010-2. [DOI] [PubMed] [Google Scholar]
  23. Masta SE. Mitochondrial sequence evolution in spiders: intraspecific variation in tRNAs lacking the TPsiC Arm. Mol Biol Evol. 2000;17:1091–1100. doi: 10.1093/oxfordjournals.molbev.a026390. [DOI] [PubMed] [Google Scholar]
  24. Lavrov DV, Boore JL, Brown WM. The complete mitochondrial DNA sequence of the horseshoe crab Limulus polyphemus. Mol Biol Evol. 2000;17:813–824. doi: 10.1093/oxfordjournals.molbev.a026360. [DOI] [PubMed] [Google Scholar]
  25. Clayton DA. Replication of animal mitochondrial DNA. Cell. 1982;28:693–705. doi: 10.1016/0092-8674(82)90049-6. [DOI] [PubMed] [Google Scholar]
  26. Yang MY, Bowmaker M, Reyes A, Vergani L, Angeli P, Gringeri E, Jacobs HT, Holt IJ. Biased incorporation of ribonucleotides on the mitochondrial L-strand accounts for apparent strand-asymmetric DNA replication. Cell. 2002;111:495–505. doi: 10.1016/S0092-8674(02)01075-9. [DOI] [PubMed] [Google Scholar]
  27. Podsiadlowski L, Bartolomaeus T. Major rearrangements characterize the mitochondrial genome of the isopod. Idotea baltica. 2006;40:893–899. doi: 10.1016/j.ympev.2006.04.008. [DOI] [PubMed] [Google Scholar]
  28. Giribet G, Edgecombe GD, Wheeler WC, Babbitt C. Phylogeny and systematic position of opiliones: A combined analysis of chelicerate relationships using morphological and molecular data. Cladistics. 2002;18:5–70. [PubMed] [Google Scholar]
  29. Waloszek D, Dunlop JA. A larval sea spider (Arthropoda: Pycnogonida) from the Upper Cambrian 'Orsten' of Sweden, and the phylogenetic position of pycnogonids. Palaeontology. 2002;45:421–446. doi: 10.1111/1475-4983.00244. [DOI] [Google Scholar]
  30. Sanderson MJ, Shaffer HB. Troubleshooting molecular phylogenetic analyses. Annu Rev Ecol Syst. 2002;33:49–72. doi: 10.1146/annurev.ecolsys.33.010802.150509. [DOI] [Google Scholar]
  31. Tarrio R, Rodriguez TF, Ayala FJ. Shared nucleotide composition biases among species and their impact on phylogenetic reconstructions of the Drosophilidae. Mol Biol Evol. 2001;18:1464–1473. doi: 10.1093/oxfordjournals.molbev.a003932. [DOI] [PubMed] [Google Scholar]
  32. Podsiadlowski L, Bartolomaeus T. Organization of the mitochondrial genome of mantis shrimp Pseudosquilla ciliata (Crustacea: Stomatopoda) Mar Biotechnol. 2005;7:618–624. doi: 10.1007/s10126-005-0017-8. [DOI] [PubMed] [Google Scholar]
  33. Roehrdanz RL, Degrugillier ME, Black WCI. Novel rearrangements of arthropod mitochondrial DNA detected with long-PCR: Applications to arthropod phylogeny and evolution. Mol Biol Evol. 2002;19:841–849. doi: 10.1093/oxfordjournals.molbev.a004141. [DOI] [PubMed] [Google Scholar]
  34. Hall TA. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucl Acids Symp Ser. 1999;41:95–98. [Google Scholar]
  35. Wyman SK, Jansen RK, Boore JL. Automatic annotation of organellar genomes with DOGMA. Bioinformatics. 2004;20:3252–3255. doi: 10.1093/bioinformatics/bth352. [DOI] [PubMed] [Google Scholar]
  36. Jeanmougin F, Thompson JD, Gouy M, Higgins DG, Gibson TJ. Multiple sequence alignment with Clustal X. Trends Biochem Sci. 1998;23:403–405. doi: 10.1016/S0968-0004(98)01285-7. [DOI] [PubMed] [Google Scholar]
  37. Castresana J. Selection of conserved blocks from multiple alignments for their use in phylogenetic analysis. Mol Biol Evol. 2000;17:540–552. doi: 10.1093/oxfordjournals.molbev.a026334. [DOI] [PubMed] [Google Scholar]
  38. Xia X, Xie Z, Salemi M, Chen L, Wang Y. An index of substitution saturation and its application. Mol Phylogenet Evol. 2003;26:1–7. doi: 10.1016/S1055-7903(02)00326-3. [DOI] [PubMed] [Google Scholar]
  39. Xia X, Xie Z. DAMBE: software package for data analysis in molecular biology and evolution. J Hered. 2001;92:371–373. doi: 10.1093/jhered/92.4.371. [DOI] [PubMed] [Google Scholar]
  40. Swofford DL. Paup – A Computer-Program for phylogenetic Inference using maximum parsimony. J Gen Physiol. 1993;102:A9. [Google Scholar]
  41. Posada D, Crandall KA. MODELTEST: testing the model of DNA substitution. Bioinformatics. 1998;14:817–818. doi: 10.1093/bioinformatics/14.9.817. [DOI] [PubMed] [Google Scholar]
  42. Huelsenbeck JP, Ronquist F. MRBAYES: Bayesian inference of phylogenetic trees. Bioinformatics. 2001;17:754–755. doi: 10.1093/bioinformatics/17.8.754. [DOI] [PubMed] [Google Scholar]
  43. Jameson D, Gibson AP, Hudelot C, Higgs PG. OGRe: a relational database for comparative analysis of mitochondrial genomes. Nucleic Acids Res. 2003;31:202–206. doi: 10.1093/nar/gkg077. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Perna NT, Kocher TD. Patterns of nucleotide composition at fourfold degenerate sites of animal mitochondrial genomes. J Mol Evol. 1995;41:353–358. doi: 10.1007/BF01215182. [DOI] [PubMed] [Google Scholar]

Articles from BMC Genomics are provided here courtesy of BMC

RESOURCES