Clitellata

Mercan, Deniz, Horenkamp, Austin, Aydemir, Merve Nur, Aydemir, Habeş Bilal, Neubig, Kurt, Arslan, Naime & Anderson, Frank E., 2025, Comparative mitogenomics of Clitellata reveals mitogenome organization can affect the mode of phylogeny, Turkish Journal of Zoology 49 (4), pp. 197-206 : 201-203

3.2. Comparative analysis of Clitellata mitogenomes

The size of complete Clitellata mitogenomes varied from 14,407 bp in Erpobdella octoculata (Hirudinea: Erpobdellidae ) to 18,528 bp in Acanthobdella peledina (Hirudinea: Acanthobdellidae ), with an average of 15,096 bp in length. Total mitogenomes of Clitellata species were biased to A+T and generally have negative AT- and GCskew values. Exceptionally, 13 species from Hirudinea and two species from Oligochaeta had moderately positive GC-skew values, with most of Oligochaeta (32 of 46) mildly biased to positive AT- skewness (Table S3). Average nucleotide compositions were as follows: T = 35.3%, A = 33.4%, C = 17.5%, and G = 13.8%.

Total PCG size varied from 10,979 bp ( Haemadipsa tianmushana ) to 11,244 bp ( Theromyzon tessulatum ), with an average of 11,102.1 bp long. Like the total mitogenome data, total PCGs of Clitellata species were biased to A+T and generally had negative AT- and GC-skew values.

Amino acid preferences of Clitellata were represented mostly by hydrophobic amino acids as expected with an A+T nucleotide bias: leucine (14.75%), serine (9.80%), isoleucine (8.86%), and methionine (8.37%). Cysteine (1.04%), arginine (1.60%), and glutamine (1.74%) amino acid preferences had the lowest levels (Table S4).

RSCU values from whole sampling are given in

Figure 3 . All stop codons were removed to avoid the bias of incomplete stop codons. All 4-fold degenerate codons preferred A in the third codon position. All 2-fold degenerate codons were biased to keep A or T nucleotides consistent with the A+T bias of the total mitogenome. On the contrary, anticodons of complementary tRNAs are conserved among Clitellata and are biased to G and T nucleotides in the first position. Anticodons of 9 tRNAs start with G or C nucleotides and therefore some interactions do not exhibit exact matches between the tRNA anticodon and the mRNA codon, as previously documented in the trnP interaction in hymenopteran insects ( Aydemir and Korkmaz, 2020). The situation in Clitellata appears more dramatic. This pattern is explained by the fact that the transcription of mitogenomes can be partially restricted by mitochondrial tRNAs (Shtolz and Mishmar, 2023). On the other hand, this is one reason why the most conserved interaction position is the second codon (anticodon) position ( Błażej et al., 2018).

The average rrnS gene length is 763.4 bp and ranges from 688 ( Hirudinaria manillensis and Poecilobdella manillensis ) to 801 bp ( Perionyx excavatus ). Average nucleotide compositions of rrnS are as follows: T = 29.7%, A = 38.5%, C = 16.1%, and G = 15.7%. The rrnS gene has a positive AT-skew (except for Poecilobdella javanica ) but variable GC-skew (32 of 78 have positive values, and 45 of 78 have negative values) (Table S5). The size of the rrnL gene varied from 1027 bp in Enchytraeus irregularis to 12,812 bp in Amynthas deogyusanensis , with an average of 1206.4 bp in length. Average nucleotide compositions of rrnL are calculated as T = 31.0%, A = 39.8%, C = 15.1%, and G = 14.1% (Table S6). As seen with the rrnS gene, the rrnL gene of Clitellata generally has positive AT- and variable GC-skew values.

The length of tRNA genes varied from 41 (trnR of W. pigra ) to 83 (trnS of some Hirudinea organisms) with an average of 63.4 bp in length (Table S7). Average nucleotide compositions of tRNAs are calculated as follows: T = 33.3%, A = 35.9%, C = 14.7%, and G = 16.1%.

The results of dN/dS calculations indicated that all mitochondrial PCGs are subjected to negative selection (dN/dS <0.5) in Clitellata ( Figure 4 ). Mutations that may occur in the cox1 gene are strongly and rapidly purified and following that, the cox3, cytB, nd1 and cox2 genes are subjected to densified negative selection. It is expected for cox and cytB genes because of their critical roles in OXPHOS ( Castellana et al., 2011), but purifying selection on the nd1 gene is unexpected, especially despite the relaxation of negative selection observed in other nd subunits. The most likely reason may be that the nd1 gene product functions as a binding site for the hydrophobic quinone during OXPHOS. This is essential as it interfaces between hydrophobic and hydrophilic subunits, and it is crucial to eliminate mutations in the nd1 gene to stabilize this interaction ( Garvin et al., 2015). On the other hand, the atp8, nd6, nd2, and nd4L genes had the highest dN/dS values and seem to have evolved relatively rapidly among mitochondrial PCGs ( Figure 4 ).

From the CREx analysis, nine different gene order patterns were observed within Clitellata ( Figure 5 ). A minimum of three and a maximum of 20 rearrangement breakpoints were detected between the reference mitogenome and the clitellate mitogenomes (Table S8). Accordingly, the transposition and tandem-duplicationrandom-loss (TDRL) event of trnG-trnY- atp8, the TDRL event of trnM-trnD gene clusters and the transposition of trnC were characterised. The most divergent gene order was determined in Olavius algarvensis and generally four different patterns were determined in Naididae . Except for Naididae (with five representatives) and Lumbriculidae ( Lumbriculus variegatus ), the oligochaete species shared a common mitochondrial gene order. Within Hirudinea, two unique gene orders were generally detected in two main groups (Erpobdelliformes + Oceanobdelliformes and Hirudiniformes) with a few exceptions. Interestingly, the mitochondrial gene order of P. hammoniensis was identical to Theromyzon tessulatum (Hirudinea: Glossiphoniidae ) ( Figure 5 ).

3.3. Phylogenetic analysis of Clitellata

In the phylogenetic analysis of the newly generated mitogenomes combined with all available clitellate mitogenomes on GenBank, St. lacustris , St. fossularis , C. diaphanus , C. diaphanus sp. B, and Sl. appendiculata clustered with the other naidines with high support ( Figure 5 ). These findings are congruent with previous studies, with Naidinae forming a sister clade to Tubificinae , with near maximum support on every branch. In previous studies, within the species used in this study, Naididae is recovered as a sister clade to most other clitellate taxa, but the mitogenome tree recovers Enchytraeidae (represented by E. irregularis ) in that position ( Erséus et al., 2020). Mitogenomes from representatives of Phreodrilidae and Propappidae could aid in recovering a similar tree to the phylogenomic tree in Erseus et al., 2020. In this analysis, the leeches are recovered as the sister taxon to Lumbriculidae , which matches a previous phylogenomic study ( Erséus et al., 2020). The unusual relationships recovered in our mitogenomic tree may reflect phylogenetic reality, but they could also be a result of long-branch attraction between Hirudinida and Naididae , which itself may be due in part to the relatively poor sampling of mitogenomes from Clitellata to date—there are no published mitogenomes available for several major lineages represented in Erséus et al. (2020). Levels of bootstrap support for many taxa, particularly within Megascolecidae , are low, so relationships between members of this family and other families are less trustworthy.

This is the most complete mitogenome study using all available clitellate mitogenomes to date, with six newly created mitogenomes for a total of 90 species. Previous studies of clitellate evolution examined the evolution of carnivory within the group, and the timing and nature of transitions among different habitat types (i.e. marine, freshwater, and terrestrial) ( Erséus et al., 2020; Horn et al., 2019; Mack et al., 2023; Rousset et al., 2008). The mitogenome data here could aid future studies of clitellate phylogeny and evolution, but accurate phylogenetic inferences will likely require sampling of mitogenomes from additional clitellate lineages, including Branchiobdellidae , Capilloventridae , Haplotaxidae , Parvidrillidae, Phreodrillidae and Propappidae . To conduct a more thorough mitogenome study, additional specimens from the following clades should also be obtained: Alluroididae , Capilloventridae , Crassiclitellata , Haplotaxidae , Moniligastridae , Narapidae , Parvidrillidae, and Randiellidae , as well as additional members of Naididae .

Figures

Figure 3. RSCU values and preferred anticodon pattern among Clitellata. Colors indicate preferred nucleotides at the 3rd codon position. Red-marked tRNA anticodons emphasize that do not exhibit an exact-match between related tRNA anticodon and preferred mRNA codons.

Figure 4. Comparison of dN/dS values of mitochondrial PCGs ofClitellata.For eachgene, the bottom and top of the line indicates the minimum and maximum values respectively. All mitochondrial genes seem to be under negative purifying selection (dN/dS <1).

Figure 5. ML tree created in IQTREEdisplaying the phylogenetic relationships betweendifferent families within Clitellata.Bootstrapsupportvalues are shown at the base of each clade with>80% showing high support. Species names in red represent the newly sequenced clitellate mitogenomes. Nine different gene order patterns observed in Clitellata were given at the end of the figure.