The First Mitochondrial Genome of Family Xylococcidae (Hemiptera, Coccomorpha) and Its Phylogenetic Implications

Abstract

Background: The mitochondrial genome has been used for understanding higher-level phylogenetic relationships within Coccomorpha. we sequenced and analyzed the first mitochondrial genome of Xylococcus castanopsis Wu & Huang, 2017 to elucidate its genomic features and phylogenetic position. Methods: The complete mitogenome was assembled using NOVOPlasty and annotated with MITOS. We analyzed genome organization, codon usage, and tRNA structures. Phylogenetic relationships were reconstructed using 13 protein-coding genes from 19 scale insect species with Bayesian Inference and Maximum Likelihood methods. Result: The mitochondrial genome is 16,363 bp in size and contains the typical 37 mitochondrial genes, with an A + T content of 89.2%. All protein-coding genes start with the ATN and end with TAA/TAG or a single T- residue. Sixteen tRNAs exhibit the typical cloverleaf structure, while the remaining six lack either the dihydrouridine (DHU) or TΨC (T) arm. Gene rearrangements occur only in individual tRNAs and transpositions between the gene clusters trnS2-ND1 and trnL1-rrnL-trnV-rrnS. Phylogenetic analysis consistently place Xylococcidae as a sister group to all scale insects except Matsucoccidae. Conclusions: This study provides the first complete mitogenome for Xylococcidae, revealing characteristic gene rearrangements. Phylogenetic reconstruction resolves the phylogenetic position of Xylococcidae as a distinct lineage sister to all scale insects except Matsucoccidae, providing critical evolutionary insights.

1. Introduction

Scale insects (Hemiptera: Coccomorpha) are small and live hidden, causing plant weakening and pathogen transmission through feeding on phloem sap, which in severe cases can even lead to plant death [1]. Currently, there are more than 8500 described species in 1238 genera, belonging to 35 extant and 21 extinct families [2]. They were divided into two informal groups: archaeococcoids and neococcoids. The archaeococcoids typically feature abdominal spiracles, at least in females. However, neococcoids lack the abdominal spiracles and multifaceted eyes [3,4].

The taxonomic status of the subfamily Xylococcinae has undergone multiple revisions. It was considered a part of the family Margarodidae sensu lato by various authors, such as Morrison [5], who recognized it as the tribe Xylococcini. Zahradnik elevated Xylococcinae to family rank [6], while Koteja restricted Xylococcidae to two genera [7]: Xylococcus and Xylococculus. Therefore, the relationships within Margarodidae sensu lato, particularly regarding the systematic position of Xylococcidae, remain unresolved. These conflicting classifications demonstrate the persistent difficulties in establishing a satisfactory system for Coccomorpha. For consistency, we follow Koteja’s [7,8] classification, which treats Xylococcidae as a distinct family.

The family Xylococcidae currently includes four genera: Baisococcus Koteja, 1989; Mesophthirus Gao, Shih, Rasnitsyn & Ren, 2019; Xylococculus Morrison, 1927; and Xylococcus Lôw, 1882. However, only seven extant species remain within Xylococcus (Xylococcus castanopsis Wu & Huang, 2017; Xylococcus filiferus Löw, 1882; Xylococcus japonicus Oguma, 1926; Xylococcus quercicola Danzig, 1980); Xylococculus (Xylococculus betulae (Pergande, 1898); Xylococculus macrocarpae (Coleman, 1908); and Xylococculus quercus (Ehrhorn, 1900), while the other two genera are extinct [2]. In China, only X. castanopsis has been recorded [9]. Remaining extant species are distributed in North America, Northeast Asia and Europe [2,10].

Mitochondria, essential organelles present in nearly all eukaryotic cells, are crucial for energy production [11]. Mitochondrial genomes of insects are now widely used in studies on species identification, population genetic structure, biogeography, and phylogenetics [12,13]. However, the current research on mitochondrial genomes of scale insects remains relatively sparse. Only a handful of cocciologists have published studies on their mitochondrial genomes, with the studies primarily focusing on their structure and characteristics and exploring phylogenetic relationships based on existing mitochondrial genomes [14,15,16,17]. Despite the high diversity of scale insects, only mitogenomes of 24 species have been recorded in NCBI.

Whether based on phylogenetic trees reconstructed from the morphological characteristics of adult males or those combining molecular data with morphological traits, the position of Xylococcidae remains unresolved, and its mitochondrial genome has yet to be documented. In this paper, the mitogenome of X. castanopsis was sequenced and analyzed in detail. We then performed a phylogeny analysis of Coccomorpha mitogenomes and tried to explore the systematic position of Xylococcidae. This contributes to a better understanding of the phylogenetic relationships within Coccomorpha.

2. Materials and Methods

2.1. Sampling, Genomic DNA Extraction and Sequencing

The sample for DNA extraction was collected from the bark of Castanea henryi (Skan) Rehd. et Wils. (Fagaceae) from Guangdong Province, China, and was preserved in 95% ethanol under −20 °C at the Beijing Forestry University, Beijing, China (BFUC). The total genomic DNA was extracted using the TIANamp Genomic DNA Kit, following the manufacturer’s protocol. Voucher specimens are deposited at Beijing Forestry University, Beijing, China (BFUC). The COX1 sequences were amplified using the primers C1-1554F and C1-2342R [18] and sequenced via the Sanger method on an ABI 3730xl sequencer to provide seed sequences for subsequent mitochondrial genome assembly. We sequenced the genome of X. castanopsis using next-generation sequencing on the Illuminia Hiseq 2500 platform at Berry Genomics, Beijing, China. The sequencing achieved a depth of 6× with 150 bp paired-end reads. A single-species Illumina TruSeq library was prepared from total genomic DNA, featuring an average insert size of 150 bp. The sequence data of phylogenetic analysis were obtained from the National Center for Biotechnology Information at https://www.ncbi.nlm.nih.gov (refer to

Table 1. Species involved in the phylogenetic analysis.

Order	Family	Species	NCBI Accession Number
Thysanoptera	Aeolothripidae	Aeolothrips albicinctus	PP932666
	Thripidae	Thrips palmi	MN127983
		Megalurothrips usitatus	ON815612
Hemiptera	Aleyrodidae	Aleyrodes shizuokensis	MT880225
		Paraleyrodes minei	PP727237
		Trialeurodes vaporariorum	AY521265
	Psyllidae	Psyllopsis discrepans	OR608273
		Trioza erytreae	MG989238
	Aphididae	Aphis citricidus	MK540501
		Aphis glycines	MK111111
		Aphis gossypii	OR449278
		Myzus persicae	KU236024
	Matsucoccidae	Matsucoccus matsumurae	PP103290
	Monophlebidae	Icerya purchasi	OX7316821
		Coronaproctus castanopsis	SRR26067561
	Putoidae	Puto sinensis	PQ727056
	Xylococcidae	Xylococcus castanopsis	PV227316
	Pseudococcidae	Paracoccus marginatus	PP060471
		Phenacoccus aceris	SRR14087799
		Phenacoccus manihoti	MZ958983
	Eriococcidae	Acanthococcus coriaceus	OP351525
		Apiomorpha munita	OP351523
	Kerriidae	Albotachardina sinensis	OP351521
	Cerococcidae	Antecerococcus theydoni	OP351522
	Aclerdidae	Nipponaclerda biwakoensis	OP351526
		Aclerda takahashii	MW839575
	Coccidae	Diensmococcus koreanus	MW302211
		Saissetia coffeae	MN863803
		Parasaissetia nigra	OK040656
		Ceroplastes japonicus	OP351524
		Ceroplastes floridensis	OK040657

).

2.2. Mitogenome Assembly and Annotation

Following the removal of adapters and low-quality reads, we obtained approximately 12 GB of clean data and assembled it for reference-guided assembly using NOVOPlasty v4.3.3 [19], with the mitochondrial genome of Matsucoccus matsumurae as the reference and the COX1 sequence of X. castanopsis serving as a bait. This process successfully yielded a mitogenome of 16,363 bp. Then, annotated using the MITOS v2.1.9 on the Galaxy platform, the accurate boundaries of protein-coding genes (PCGs) were confirmed by comparison with other scale insects using MEGA 7.0.26 [20,21]. Six tRNA genes (trnA, trnQ, trnL1, trnS2, trnW, and trnV) were not identified by MITOS v2.1.9; however, they were successfully detected using the manual method and ARWEN v1.2 [22]. The rrnS was detected by MITOS v2.1.9, and the rrnL was determined by the upstream and downstream of tRNAs and alignment with the other scale insects.

2.3. Bioinformatic Analysis

Base composition, codon usage of protein-coding genes (PCGs), and relative synonymous codon usage (RSCU) values were calculated using MEGA v7.0.26. Figures were then generated using the ggplot2 package in R v4.3.2 [23,24]. Base composition skew values were computed as follows: AT skew = (A − T)/(A + T) and GC skew = (G − C)/(G + C) [25].

2.4. Phylogenetic Analysis

The phylogenetic analysis included a total of 31 species, comprising 19 species from the Coccomorpha, 4 species from the Aphidomorpha, 3 species from the Aleyrodomorpha, 2 species from the Psyllomorpha, and 3 species of order Thysanoptera. Phylogenetic relationship was inferred using Maximum Likelihood (ML) and Bayesian Inference (BI). A dataset comprising protein-coding genes from 31 species was concatenated in Phylosuite v1.2.2 for phylogenetic analysis [26]. Thirteen sequences were aligned in batches with MAFFT v7.505 using the ‘G-INS-i (accurate)’ strategy and codon alignment mode [27]. Ambiguously aligned fragments of 13 alignments were removed in batches using Gblocks v0.91b [28]. Finally, the 13 protein-coding genes were concatenated in Phylosuite. Before constructing Maximum Likelihood tree and Bayesian Inference tree, the protein-coding dataset was partitioned schemes for evolutionary models were selected using PartitionFinder v2.1.1 [29]. For the Maximum Likelihood tree, the best partitioning scheme and evolutionary models for the 13 protein-coding genes were selected using the greedy algorithm combined with the AICc criterion. However, for the Bayesian tree, the greedy algorithm was applied in conjunction with the BIC criterion. Maximum likelihood phylogenies were inferred using IQ-TREE v2.2.0 under the edge-linked partition model for 5000 ultrafast bootstraps, as well as the Shimodaira–Hasegawa-like approximate likelihood ratio test [30,31,32]. Bayesian Inference phylogenies were inferred using MrBayes v3.2.6 under the partition model (2 parallel runs, 2,000,000 generations) [33], in which the initial 25% of sampled data were discarded as burn-in.

3. Results

3.1. General Features and Nucleotide Composition

The length of the mitogenome sequence of X. castanopsis was 16,363 bp, containing 13 PCGs, 22 tRNAs and 2 rRNAs, amongst which 9 PCGs and 15 tRNAs were encoded on the forward strand (+) and 13 genes were encoded by the reverse strand (-) (Figure 1 and

Table 2. Mitochondrial genome characteristics of X. castanopsis.

Gene	Location	Size	Strand	Start Codon	Stop Codon	Anticodon	Intergenic Length
trnS2	1273–1344	72	+	-	-	CCT	-
ND1	1439–2356	918	-	ATA	TAA	-	94
trnM	2424–2493	70	+	-	-	CAT	67
trnI	2500–2565	66	+	-	-	GAT	6
trnQ	2564–2617	54	-	-	-	TTG	−2
ND2	2617–3564	948	+	ATT	TAA	-	−1
trnY	3565–3625	61	+	-	-	GTA	0
trnW	3631–3703	61	-	-	-	TCA	5
trnC	3705–3773	69	-	-	-	GCA	1
COX1	3775–5304	1530	+	ATG	TAA	-	1
trnL2	5300–5365	66	+	-	-	TAA	−5
COX2	5369–6038	670	+	ATA	T-	-	3
trnK	6039–6104	66	+	-	-	TTT	0
trnD	6114–6178	65	+	-	-	GTC	9
ATP8	6179–6328	150	+	ATA	TAA	-	0
ATP6	6318–6947	630	+	ATA	TAA	-	−11
COX3	6937–7683	750	+	ATG	TAA	-	−14
trnG	7729–7792	64	+	-	-	TCC	45
ND3	7790–8143	354	+	ATA	TAA	-	−3
trnA	8157–8203	47	+	-	-	AGC	13
trnR	8259–8312	54	+	-	-	TCG	55
trnN	8312–8375	64	+	-	-	GTT	−1
trnS1	8377–8439	60	+	-	-	TCT	1
trnE	8447–8512	66	+	-	-	TTC	7
trnF	8512–8568	57	-	-	-	GAA	−1
ND5	8574–10,205	1632	-	ATT	TAG	-	5
trnH	10,227–10,299	73	-	-	-	GTG	21
ND4	10,303–11,610	1308	-	ATG	TAA	-	3
ND4L	11,604–11,897	294	-	ATT	TAA	-	−7
trnT	11,887–11,955	69	+	-	-	TGT	−11
trnP	11,959–12,017	59	-	-	-	TGG	3
ND6	12,021–12,485	465	+	ATA	TAA	-	3
cytb	12,493–13,579	1086	+	ATA	TAA	-	7
trnL1	13,603–13,669	67	-	-	-	TAA	24
rrnL	13,700–15,009	1340	-	-	-	-	0
trnV	15,010–15,071	62	+	-	-	AAC	0
rrnS	15,072–15,739	668	-	-	-	-	0

“+/-” indicates whether the gene is located on the forward or reverse strand, respectively. The minus signs (-) appearing in the “Start codon”, “Stop codon”, and “Anticodon” columns simply denote blank entries or absence of relevant data for those particular fields.

). Ten regions of gene overlap between adjacent genes were detected in the mitogenome, ranging from −1 to −14 bp, and 20 intergenic regions, ranging from 1 to 94 bp. The overall base composition of the mitogenome sequence was 42.9% T, 46.3% A, 7.7% C and 3.1% G, with a strong bias towards A + T (89.2%) (

Table 3. The nucleotide composition of the mitochondrial genome.

Gene	% T	% C	% A	% G	% A + T	AT Skew	GC Skew
ATP6	47.0	7.5	43.0	2.5	90.0	−0.04	−0.49
ATP8	41.3	5.3	52.7	0.7	94.0	0.12	−0.78
COX1	43.9	10.5	38.6	7.1	82.5	−0.06	−0.19
COX2	42.4	8.1	45.2	4.3	87.6	0.03	−0.31
COX3	49.3	6.5	41.2	3.0	90.4	−0.09	−0.37
cytb	45.3	8.9	41.0	4.8	86.3	−0.05	−0.30
ND1	51.7	3.3	36.3	8.7	88.0	−0.18	0.45
ND2	49.4	6.8	42.8	1.0	92.2	−0.07	−0.74
ND3	44.6	5.4	46.9	3.1	91.5	0.02	−0.27
ND4	54.0	3.0	35.7	7.3	89.7	−0.20	0.42
ND4L	53.7	1.0	38.8	6.5	92.5	−0.16	0.73
ND5	53.7	3.0	36.4	6.9	90.1	−0.19	0.39
ND6	48.8	4.7	45.2	1.3	94.0	−0.04	−0.57
PCGs	48.8	6.0	40.0	5.2	88.8	−0.10	−0.07
rrnL	48.1	2.6	42.1	7.2	90.1	−0.07	0.47
rrnS	45.1	3.1	43.6	8.2	88.6	−0.02	0.45
2 rRNA	47.0	2.8	42.6	7.6	89.6	−0.05	0.46
22 tRNA	44.3	4.8	45.9	5.0	90.2	0.02	0.01
Total	42.9	7.7	46.3	3.1	89.2	0.04	−0.43

).

3.2. Protein-Coding Genes

The 13 protein-coding genes (PCGs) span a total of 10,747 bp, comprising 65.7% of the complete mitogenome length of X. castanopsis, with a high A + T content of 88.8% (Table 3). The AT skew values for the 13 PCGs range from −0.20 in ND4 to 0.12 in ATP8, while the GC skew values vary from −0.78 in ATP8 to 0.73 in ND4L. The start codons for all protein-coding genes were ATN, while the stop codons varied: COX2 ended with T-, ND5 ended with TAG, and the remaining genes all terminated with TAA. The relative synonymous codon usage (RSCU) was depicted in Figure 2. The usage of codons for protein-coding genes is presented in descending order of frequency as follows: AAT, ATT, TTT, TAT and ATA (

Table 4. The utilization of codons of the mitogenome of X. castanopsis.

Amino Acid	Codon	Count	RSCU	Amino Acid	Codon	Count	RSCU
Phe	UUU(F)	447	1.77	Ser	UCU(S)	36	1.33
	UUC(F)	58	0.23		UCC(S)	13	0.48
Leu	UUA(L)	235	4.62		UCA(S)	70	2.59
	UUG(L)	25	0.49		UCG(S)	8	0.3
Leu (c)	CUU(L)	21	0.41	Ser (s)	AGU(S)	24	0.89
	CUC(L)	5	0.1		AGC(S)	5	0.19
	CUA(L)	18	0.35		AGA(S)	38	1.41
	CUG(L)	1	0.02		AGG(S)	22	0.81
Ile	AUU(I)	478	1.83	Thr	ACU(T)	38	1.69
	AUC(I)	44	0.17		ACC(T)	17	0.76
Met	AUA(M)	243	1.81		ACA(T)	27	1.2
	AUG(M)	25	0.19		ACG(T)	8	0.36
Val	GUU(V)	18	3	Ala	GCU(A)	7	2.8
	GUC(V)	1	0.17		GCC(A)	2	0.8
	GUA(V)	5	0.83		GCA(A)	1	0.4
	GUG(V)	0	0		GCG(A)	0	0
Tyr	UAU(Y)	370	1.89	Cys	UGU(C)	14	1.47
	UAC(Y)	22	0.11		UGC(C)	5	0.53
*	UAA(*)	170	1.68	Trp	UGA(W)	68	1.77
	UAG(*)	32	0.32		UGG(W)	9	0.23
His	CAU(H)	31	1.88	Arg	CGU(R)	5	1.82
	CAC(H)	2	0.12		CGC(R)	0	0
Gln	CAA(Q)	23	2		CGA(R)	6	2.18
	CAG(Q)	0	0		CGG(R)	0	0
Asn	AAU(N)	483	1.89	Pro	CCU(P)	24	1.96
	AAC(N)	29	0.11		CCC(P)	2	0.16
Lys	AAA(K)	177	1.71		CCA(P)	23	1.88
	AAG(K)	30	0.29		CCG(P)	0	0
Asp	GAU(D)	38	2	Gly	GGU(G)	8	0.64
	GAC(D)	0	0		GGC(G)	0	0
Glu	GAA(E)	27	1.86		GGA(G)	38	3.04
	GAG(E)	2	0.14		GGG(G)	4	0.32

The “*” denotes stop codon.

).

3.3. tRNA Genes and rRNA Genes

The tRNAs of X. castanopsis had a length ranging from 47 bp (trnA) to 73 bp (trnH). Among 22 tRNAs, 16 tRNAs (trnC, trnD, trnE, trnG, trnH, trnI, trnK, trnL1, trnL2, trnM, trnN, trnQ, trnR, trnS2, trnT, and trnW) exhibited a complete cloverleaf structure, while 3 tRNAs (trnA, trnS1, and trnY) lacked the dihydrouridine (DHU) arm, and 3 other tRNAs (trnF, trnP, and trnV) lacked the TψC (T) arm (Figure 3).

The two rRNAs were both on the reverse strand in the mitogenome of X. castanopsis. The rrnL gene was 1,340 bp with 90.1% A + T content. The rrnS gene was 668 bp with 88.6% A + T content.

3.4. Gene Rearrangements Analysis

The mitochondrial genome of X. castanopsis exhibits significant gene rearrangements when compared to the ancestral sequence and other archeaococcoids (Figure 4). The genes trnS2-ND1 and trnL1-rrnL-trnV-rrnS have undergone transposition (T), and the positions of trnI-trnQ and trnM have been transposed (T). Additionally, the gene trnV has undergone inversion (I), and the rearrangement involving trnY and trnW-trnC is an inverted transposition (iT).

3.5. Phylogenetic Analysis of Coccomorpha

The phylogenetic trees reconstructed using the Maximum Likelihood and Bayesian Inference methods, based on the dataset of 13 protein-coding genes, were identical and had high support (Figure 5). Four infraorders (Aleyrodomorpha, Psyllomorpha, Aphidomorpha and Coccomorpha) were recovered as monophyletic with strong support. Aleyrodomorpha was sister to other infraorders. The phylogenetic relationship of Aleyrodomorpha + (Psyllomorpha + (Aphidomorpha + Coccomorpha))) was found with high statistical support (bootstrap value = 100%, posterior possibility = 1).

Within Coccomorpha, the Matsucoccidae was a sister to other coccoid families analyzed here. Xylococcidae formed a sister group with all other scale insects excluding Matsucoccidae. The two species of Monophlebidae formed separate clades, strongly supporting their monophyly (bootstrap value = 98%, posterior possibility = 1). Putoidae was found to be the sister group to all neococcids. The Pseudococcidae was grouped with the remaining neococcids. Among the remaining neococcids, the phylogenetic relationship of (Eriococcidae + (Kerriidae + (Cerococcidae + (Aclerdidae + Coccidae)))) was found with high statistical support (bootstrap value = 100%, posterior possibility = 1).

4. Discussion

Rearrangements in mitochondrial genomes are common in scale insects [14,15,16,17,18]. The Matsucoccidae exhibits gene arrangements that are similar to the ancestral sequence, with only individual tRNA transpositions (T). However, due to morphological similarities between xylococcids and matsucoccids, they are often placed by cocciologists as tribe Xylococcini and Matsucoccini within Margarodidae sensu lato, or alternatively [5,34], genera such as Xylococcus and Matsucoccus are classified under Xylococcidae [6]. Our results show that, compared to the ancestral arrangement, Matsucoccidae exhibits tRNA transpositions, while Xylococcidae exhibits more rearrangements, such as tRNA transpositions (T), inversions (I), inverted transpositions (iT), and transpositions (T) between the gene clusters trnS2-ND1 and trnL1-rrnL-trnV-rrnS.

The Coccomorpha is divided into two informal groups: archaeococcids and neococcoids. Archaeococcids are characterized by the presence of abdominal spiracles, and adult males possess compound eyes or a row of unicorneal eyes encircling the head, while the neococcoids lack the abdominal spiracles and compound eyes (except for Kermesidae and some Coccidae) [4,35]. Within Coccomorpha, Matsucoccidae is considered to be more morphologically similar to Orthezia and Aphis (Aphidomorpha) than to Xylococcidae [36]. However, Koteja proposed a potential relationship between Matsucoccidae and Xylococcidae based on mouthpart morphology, while suggesting a less likely affinity with Ortheziidae [7]. Due to the lack of data on Ortheziidae, our results recognize Matsucoccidae as the sister with other scale insects, showing closer phylogenetic proximity to Xylococcidae, consistent with previous findings [34,37,38].

Although the Xylococcidae and Matsucoccidae were included in Margarodidae sensu lato, the morphology of adult males and females of Xylococcidae is distinct from that of Matsucoccidae. The pores are present in the atrium of the abdominal spiracles of Xylococcidae, while absent in those of Matsucoccidae. Additionally, bitubular ducts are absent in Xylococcidae but present in Matsucoccidae [39]. In males, Matsucoccidae possess only a cluster of cupolae at the wing base, whereas Xylococcidae exhibit both a row of cupolae along the subcostal complex and a cluster at the wing base [8]. Morphological analysis of adult males reveals close phylogenetic relationships between Matsucoccidae and Xylococcidae, evidenced by shared derived characteristics including (1) the presence of a dorsal waxy tuft, (2) a group of very small setae on the dorsal surface of antennal pedicel and coxal bases, and (3) the development of alar folds [40,41]. In our study, Xylococcidae was grouped with all other scale insects except for Matsucoccidae, with strong support (98% bootstrap support and one posterior probability). This mainly supports the phylogenetic result obtained by Hodgson and Hardy based on male morphological characteristics [38]. Monophlebidae has consistently shown strong monophyly, whether based on molecular or morphological data [37,42,43]. However, there is still no satisfactory classification system [44]. Our results suggest that Monophlebidae forms a clade. It forms a sister with all other scale insects.