A Mysterious Asian Firefly Genus, Oculogryphus Jeng, Engel & Yang (Coleoptera, Lampyridae): The First Complete Mitochondrial Genome and Its Phylogenetic Implications

Simple Summary The family Lampyridae, commonly known as fireflies, is a cosmopolitan group comprising approximately 100 genera and 2200 species. Oculogryphus Jeng, Engel & Yang, 2007 is a small firefly genus endemic to Asia with only four known species and the phylogenetic position of it remains unresolved. It has been considered enigmatic because of its characteristic morphology intermingling different subfamilies, while its systematic status has never been rigorously tested due to a lack of molecular data. In this study, we obtained a large series of O. chenghoiyanae Yiu & Jeng, 2018 from southeastern China. We successfully sequenced and annotated its complete mitochondrial genome and clarified the systematic status of Oculogryphus. Abstract The firefly genus Oculogryphus Jeng, Engel & Yang, 2007 is a rare-species group endemic to Asia. Since its establishment, its position has been controversial but never rigorously tested. To address this perplexing issue, we are the first to present the complete mitochondrial sequence of Oculogryphus, using the material of O. chenghoiyanae Yiu & Jeng, 2018 determined through a comprehensive morphological identification. Our analyses demonstrate that its mitogenome exhibits similar characteristics to that of Stenocladius, including a rearranged gene order between trnC and trnW, and a long intergenic spacer (702 bp) between the two rearranged genes, within which six remnants (29 bp) of trnW were identified. Further, we incorporated this sequence into phylogenetic analyses of Lampyridae based on different molecular markers and datasets using ML and BI analyses. The results consistently place Oculogryphus within the same clade as Stenocladius in all topologies, and the gene rearrangement is a synapomorphy for this clade. It suggests that Oculogryphus should be classified together with Stenocladius in the subfamily Ototretinae at the moment. This study provides molecular evidence confirming the close relationship between Oculogryphus and Stenocladius and discovers a new phylogenetic marker helpful in clarifying the monophyly of Ototretinae, which also sheds a new light on firefly evolution.

One of the incertae sedis groups of Lampyridae is Oculogryphus Jeng, Engel & Yang, 2007 [18].The genus was erected based on O. fulvus Jeng, 2007, located in Vietnam, by monotypic and original designation.Subsequently, three more species were added from Vietnam and China [19][20][21].This genus is morphologically distinctive with males having large compound eyes that are significantly emarginate posteriorly and approximate ventrally [18,20], while females are neotenic and completely larviform [21].Originally, Oculogryphus was thought to be a mysterious taxon with features intermingling those of Rhagophthalmidae, Luciolinae, Lampyrinae, or the Ototretine-Ototretadriline complex [18].It was later revealed to be closely related to the ototretine genus Stenocladius Fairmaire in Deyrolle and Fairmaire, 1878 based on a morphology-based phylogenetic study of Lampyridae [13,19].However, Janisova and Bocakova [22] did not consider it as a member of Ototretinae when they synonymized the subfamily Ototretadrilinae Crowson, 1972 with the latter and redefined the limits of the group based on adult morphology.After that, with further discoveries of additional Oculogryphus species, some additional evidence was found to support its close relationship to Stenocladius in the morphological characters of males and neotenic females, as well as bioluminescence behavior for attracting mates [20,21].Despite these great efforts, the systematic status of Oculogryphus remains controversial at the moment and is placed as Lampyridae incertae sedis in the latest classification [16].When there remains controversy over the taxonomic status of some morphospecies, it becomes necessary to use molecular data to test phylogenetic relationships [23][24][25].However, no molecular data has been available for Oculogryphus until now, thereby preventing us from rigorously testing its phylogenetic position.
During our recent study, we conducted extensive insect collections in southeastern China.Fortunately, we were able to obtain a large series of Oculogryphus materials among these samples.The availability of these valuable materials provided us with the opportunity to investigate its phylogenetic position based on the molecular data of Lampyridae.Mitochondrial genomes (mitogenomes) and nuclear ribosomal DNA (rDNA) repeat units are widely used in insect phylogenies [26,27].Furthermore, alongside raw sequence data for phylogenetic comparison [28], the sequences and structural characteristics of mitogenomes, including the secondary structure of RNA genes, base content, and gene arrangement, can reflect differences in the functions and evolutionary patterns of diverse taxa [29,30].As the number of mitogenome reference sequences has increased, comparative feature analysis among and within certain groups has become more informative [30].Gene rearrangement is one of the most frequently investigated characteristics of metazoan mitogenomes [30][31][32][33][34][35], and it has been an extremely useful phylogenetic marker [31,36,37].The gene rearrangement between trnC and trnW was recently detected in Stenocladius [38], prompting us to thoroughly explore whether similar characteristics exist in Oculogryphus or not.Moreover, three regions of the nuclear rDNA repeat unit (e.g., 18S, 5.8S, and 28S) are classically conserved molecular markers with very low rates of genetic saturation (defined as multiple substitutions at the same site) [39], which are commonly used to construct phylogenies in insects [15,[40][41][42].
In the present study, we will analyze the characteristics of the mitochondrial genome of Oculogryphus for the first time and include it in the reconstruction of Lampyridae's molecular phylogeny.Our aims are to (1) make a comparison between Oculogryphus and Stenocladius in the mitochondrial genes arrangement, (2) preliminarily investigate the phylogenetic position of Oculogryphus using a comprehensive molecular dataset following Chen et al. [15], and (3) further investigate the phylogenetic position of Oculogryphus with all representative taxa with complete mitogenomes.Through these results, we will test the hypothesis of whether Oculogryphus is a sister to Stenocladius or not.

Materials
We conducted extensive insect collections in southeastern China from 2020 to 2021 and obtained a large series of Oculogryphus materials among these samples.Some Oculogryphus materials were located from Nanling National Natural Reserve, Guangdong Province, China, 24.911547-24.937461C for long-term storage prior to DNA extraction.All of them were deposited at the Museum of Hebei University, Baoding, China (MHBU).The material for the molecular study was attached with a voucher number (MHBU, 3CA0034).

Morphological Technique
The specimens were initially softened in water, followed by dissection of the genitalia and hind wings.Subsequently, the male genitalia were cleared in a 10% NaOH solution, examined and photographed in glycerol, and finally affixed onto a paper card for permanent preservation.The left hind wing of each specimen was removed from the body and mounted in neutral balsam between a microscope slide and a cover slip.At least one specimen was dissected for each locality, with additional specimens treated if any damage occurred during dissection.Images of the genitalia and hind wings were captured using a Leica M205A stereo microscope.Multiple layers were stacked using Helicon Focus 7. Post-processing of images was conducted using Adobe Photoshop 2020.
The distribution information was gathered from the original publications [21] and the newly collected material of the present study.

DNA Extraction, Mitochondrial Genome Sequencing, and Assembly
Total genomic DNA was extracted from chest muscle using a DNeasy Blood and Tissue kit (QIAGEN, Beijing, China), following the manufacturer's protocol, and the extracted DNA was stored at −20 • C for further molecular studies.Libraries were constructed and sequenced using the Nova Seq 6000 platform (Illumina, Alameda, CA, USA) at Berry Genomics (Beijing, China).Paired-end reads with 150 bp length and an insert size of 350 bp were sequenced.Approximately 6G of data were produced for the sample to ensure the minimum sequencing coverage of 10×.The mitogenome was assembled using NOVOPlasty version 3.8.3[43], with variable K-mer sizes (21,24,27,30,33,36), until a consistent result was achieved.

Dataset Assembly and Phylogenetic Analyses
In our study, we analyzed three different datasets including Oculogryphus.First, we investigated the position of this genus within Lampyridae using the most comprehensive DNA datasets (dataset-1: both nuclear and mitochondrial genes, including 10 mitochondrial genes: COX1-2, ATP6&8, ND1, ND4-5, CYTB, 12S, 16S, and three rDNA regions: 18S, 5.8S, and 28S by Chen et al. [15].In this analysis, a total of 53 Lampyridae species served as ingroups, and eight species of Elateroidea were used as outgroups due to the inconsistency of the sister group [16] (Table S1).Second, because not all molecular markers of the representative species are available for us in this analysis, which may influence the estimation of the phylogenetic relationships, we created another two datasets that included only complete mitogenomes (dataset-2: PCGRNA matrix, including 13 PCGs and two rRNAs; and dataset-3: PCG12RNA matrix, including the 1st and 2nd codon positions of the 13 PCGs and two rRNAs).In this second analysis, a total of 43 species of Lampyridae were chosen as the ingroups, and eight species of Elateroidea were also used as outgroups (Table S2).Except for the newly sequenced mitogenome and three rDNA regions (18S, 5.8S, and 28S) of Oculogryphus, all other sequences were downloaded from GenBank [54].
The individual genes of PCGs and nuclear sequences were aligned using the MAFFT algorithm [55] implemented in PhyloSuite [56] with the L-INS-I strategy.The alignments were trimmed using G-blocks [57] and then concatenated into the aforementioned different datasets.The pre-defined partitions for different datasets follow a consistent pattern, with protein-coding genes divided according to codon positions, while other genes are each placed into a separate partition.PartitionFinder [58] and ModelFinder [59] were utilized to search for the optimal partitioning scheme and models for each dataset.Maximum likelihood (ML) trees were constructed using IQ-Tree version 2.0.7 [60] with 1000 SH-aLRT replicates.Bayesian inference (BI) analyses were conducted in MrBayes v3.2.6 [61] until the average standard deviation of splitting frequency was less than 0.01.Additionally, the second datasets were analyzed under the default CAT+GTR model using PhyloBayes-MPI version 1.8 [62].Two independent MCMC chains were run until they satisfactorily converged (maxdiff < 0.3).Interactive Tree of Life (iTOL, http://itol.embl.de,assessed on 25 January 2024) [63] was used to visualize the phylogenetic tree.Distribution.China (Jiangxi, Guangdong, Hong Kong).Descriptive notes.In the hind wing of the male (Figure 1), the MP3+4 is vestigial, with its trace being more or less clear along the whole length (e.g., Figure 1A,B).Sometimes it is branched terminally into MP3 and MP4, respectively (e.g., Figure 1C,D).In the aedeagus (Figure 2), the median lobe is thinned apically in the dorsal view (Figure 2A-H), and strongly curved dorsally in the lateral view, which looks nearly even in width along the whole length and feebly expanded at the apex (Figure 2a-h).The parameres are elongate and narrowly rounded at apices, but they can never reach the apex of the median lobe (Figure 2A-H).They are usually diverging from each other apically, rarely nearly parallel (Figure 2F).The basal piece is roughly a U-shaped band (Figure 2A-H), sometimes presenting with a more or less distinct median notch at the caudal margin (e.g., Figure 2B,E-G).It is shorter than (e.g., Figure 2D,F-H) or approximately as long as (e.g., Figure 2A-C,E) the median lobe.Remarks.Considering the rarity, or little current understanding of this taxon, we conducted a thorough and comprehensive morphological observation to ensure accurate specific identification.

Class
The external morphology of this species was thoroughly described in the original manuscript [21].Since there is little variation within these materials, it is unnecessary to reiterate the descriptions of the same characters herein.Distribution.China (Jiangxi, Guangdong, Hong Kong).Descriptive notes.In the hind wing of the male (Figure 1), the MP 3+4 is vestigial, with its trace being more or less clear along the whole length (e.g., Figure 1A,B).Sometimes it is branched terminally into MP 3 and MP 4 , respectively (e.g., Figure 1C,D).
In the aedeagus (Figure 2), the median lobe is thinned apically in the dorsal view (Figure 2A-H), and strongly curved dorsally in the lateral view, which looks nearly even in width along the whole length and feebly expanded at the apex (Figure 2a-h).The parameres are elongate and narrowly rounded at apices, but they can never reach the apex of the median lobe (Figure 2A-H).They are usually diverging from each other apically, rarely nearly parallel (Figure 2F).The basal piece is roughly a U-shaped band (Figure 2A-H), sometimes presenting with a more or less distinct median notch at the caudal margin (e.g., Figure 2B,E-G).It is shorter than (e.g., Figure 2D,F-H) or approximately as long as (e.g., Figure 2A-C,E) the median lobe.
Remarks.Considering the rarity, or little current understanding of this taxon, we conducted a thorough and comprehensive morphological observation to ensure accurate specific identification.
The external morphology of this species was thoroughly described in the original manuscript [21].Since there is little variation within these materials, it is unnecessary to reiterate the descriptions of the same characters herein.

General Features of the Mitochondrial Genome
The complete mitochondrial genome of O. chenghoiyanae is 16,546 bp in size, with a nucleotide composition biased towards A and T (47.1% for A, 33.4% for T, 12.1% for C, and 7.4% for G).It has a positive AT-skew (0.17) and a negative GC-skew (−0.24) (Table S3).In addition, the value of non-synonymous substitutions is 0.20957 (Figure S1).The mitogenome is a typical double-strand circular molecule containing 13 protein-coding genes (PCGs), 22 transfer RNA genes (tRNAs), 2 ribosomal RNA genes (rRNAs), and an A+T-rich region, of which 14 genes (8 tRNAs, 4 PCGs, and 2 rRNAs) were transcribed from the minority strand (N-strand) while the others (14 tRNAs and 9 PCGs) were encoded on the majority strand (J-strand) (Figure 3a, Table S4).Gene arrangement differs from most fireflies, e.g., as in [64][65][66] (Figure 3b, as ND2-trnW-trnC-trnY), with a rearrangement between trnC and trnW genes observed in O. chenghoiyanae (as ND2-trnC-trnW-trnY).In addition, all tRNA genes fold into the standard clover-leaf structure except trnS1, which is missing the dihydrouridine (DHU) arm (Figure S2).Moreover, there are 12 gene overlaps ranging from 1 to 8 bp in length, as well as five intergenic spacer regions between genes, with the longest between trnC and trnW (702 bp) (Table S4).Notably, the intergenic spacer between trnC and trnW genes contains six tandemly repeated units (TRU, 111 bp) along with a partial 44 bp repeat unit.Each TRU is composed of two parts (Figure 3b).One part corresponds to a portion of the nucleotide sequence of trnW (29 bp), which is thought to be a remnant of trnW and relatively conserved except for only one base mutation from A to G in the fifth repeat unit.The other part of the TRU is 82 bp, which is identical to one another.composed of two parts (Figure 3b).One part corresponds to a portion of the nucleotide sequence of trnW (29 bp), which is thought to be a remnant of trnW and relatively conserved except for only one base mutation from A to G in the fifth repeat unit.The other part of the TRU is 82 bp, which is identical to one another.

Phylogenetic Analyses
Using the ML and BI analyses, almost congruent topologies were produced for the mitochondrial and nuclear genes (Figures 4a-d Like the above, the phylogenetic analyses of the complete mitogenomes (concatenated into two different datasets) showed that Ototretinae was a paraphyly, consisting of two branches (Figures 4e and S5-S9).One branch included only Drilaster (BS = 100, PP = 1), and the other grouped with Oculogryphus and Stenocladius (BS = 100, PP = 1).Drilaster was always in the basal clade sister to the rest of the Lampyridae (BS = 98-100, PP = 1).And Oculogryphus and Stenocladius were further sisters to the others (PP = 1) using the BI analysis, while they were grouped with either Lampyrinae or Luciolinae under the ML analysis albeit with lower supporting values (BS = 38-59).

Phylogenetic Analyses
Using the ML and BI analyses, almost congruent topologies were produced for the mitochondrial and nuclear genes (Figures 4a-d Like the above, the phylogenetic analyses of the complete mitogenomes (concatenated into two different datasets) showed that Ototretinae was a paraphyly, consisting of two branches (Figures 4e and S5-S9).One branch included only Drilaster (BS = 100, PP = 1), and the other grouped with Oculogryphus and Stenocladius (BS = 100, PP = 1).Drilaster was always in the basal clade sister to the rest of the Lampyridae (BS = 98-100, PP = 1).And Oculogryphus and Stenocladius were further sisters to the others (PP = 1) using the BI analysis, while they were grouped with either Lampyrinae or Luciolinae under the ML analysis albeit with lower supporting values (BS = 38-59).

Species Identification
We identified all materials as O. chenghoiyanae following the key by Yiu & Jeng [21].A comprehensive study of these available materials indicates some variations in their aedeagi and hind wings, which are important for identifying the Oculogryphus species [18][19][20][21].

Species Identification
We identified all materials as O. chenghoiyanae following the key by Yiu & Jeng [21].A comprehensive study of these available materials indicates some variations in their aedeagi and hind wings, which are important for identifying the Oculogryphus species [18][19][20][21].
In the original description of O. chenghoiyanae ([21]: Figure 2), the hind wing venation is similar to some of our material (Figure 1A), in which the MP 3+4 is vestigial, and its vein trace is disrupted.But in some other cases, the MP 3+4 vein trace is complete (Figure 1B), or may also be bifurcate (Figure 1C,D).Despite this variability, the vestigial nature of MP 3+4 is consistent across specimens of O. chenghoiyanae, unlike all others of Oculogryphus having well-developed ones ( [18]: Figure 8; [19]: Figure 3; [20]: Figure 3).Also, the aedeagus of O. chenghoiyanae is variable in the shapes of its component structures.As originally described ( [21]: Figure 4A), the basal piece is usually a Ushaped band in the dorsal view (e.g., Figure 2A,C,D), sometimes with a median notch at the caudal margin (e.g., Figure 2B,E,H).It is always bent dorsally, and abruptly (e.g., Figure 2a,c,f,h; [21]: Figure 4B) or progressively (e.g., Figure 2b,d,e) narrowed basally in the lateral view.Parameres are generally elongate and tapered apically, but their length varies.Some are short, less than half the length of the median lobe (e.g., Figure 2B,G), while others are relatively long, approximately 2/3 the length of the median lobe (e.g., Figure 2F,H), and the rest fall between these lengths.The median lobe varies in stoutness in the dorsal view (e.g., Figure 2A,D; [21]: Figure 4A), with some being slender (e.g., Figure 2B,E).However, its shape is quite consistent in the lateral view (Figure 2a-h).While the male genitalia are generally considered to be the most stable and reliable structure in insect taxonomy e.g., [67], their shape may vary more or less within the species [68].This is not an individual case of fireflies, such as Microphotus LeConte, 1866 [69], but it typically exhibits only quantitative variation.Therefore, it is essential to thoroughly examine a large number of individuals when describing a new species, in order to comprehensively and accurately recognize its defining characteristics.
Although there are some variations within O. chenghoiyanae, its males could be well differentiated from all other Oculogryphus species by a combination of the following characters: a pronotum and elytra of highly contrasting colors, with a reddish-brown pronotum, and black elytra except for brown at the humeri; the elytra are widest near the middle part, with a narrow epipleura; a hind wing with vestigial MP 3+4 ; an aedeagus with parameres less than 2/3 the length of the median lobe, which is uniform in width along the whole length in the lateral view.The consistency of these characters, coupled with their close geographical distribution (or the sympatric range), leads us to conclude that all specimens should be classified as O. chenghoiyanae.

Comparisons of the Mitogenomes between Oculogryphus and Stenocladius
With O. chenghoiyanae as the studied material, this study presented the first complete mitogenome record for Oculogryphus.Similar to the mitogenomes of Stenocladius [38], a rearrangement between the trnW and trnC was detected in Oculogryphus, with a long intergenic spacer existing between the two rearranged genes.Within this long intergenic spacer, some remnant genes of trnW were also found (Figure 3a).However, there are some detailed differences between Oculogryphus and Stenocladius in the intergenic spacer between the trnW and trnC genes (Table S4).This non-coding region of Oculogryphus (702 bp) is much longer than that of Stenocladius (241 to 376 bp) [38].Additionally, the remnant gene of trnW in Oculogryphus is longer (29 bp) and repeated more times (six) than that of Stenocladius (23 bp, one or two repeats) [38].Furthermore, the sequences between the remnant genes are identical to one another (82 bp) in Oculogryphus, while they are different in Stenocladius [38].From another perspective, the intergenic spacer of Oculogryphus is composed of six TRUs, while only repeat sequence units (RSUs) or remnant genes exist in Stenocladius.
The presence of the intergenic spacers located in positions involved in rearrangements, combined with the presence of gene remnants that changed positions within intergenic spacers, provides evidence for identifying plausible rearrangement pathways [70,71].Similar to what occurred in all Stenocladius mitogenomes [38], the rearrangement of trnW and trnC in Oculogryphus also occurred through transposition, with the genes moving to different places on the same strand [71].Transposition can be explained by a tandem duplication and random loss model (TDRL) [72,73].TDRL involves a tandem duplication of a continuous segment of genes such that the original segment and its copy are placed consecutively, fol-lowed by the loss of one copy of each redundant gene [74].Further assumptions regarding rearrangement processes were made based on the principle of parsimony for Oculogryphus (Figure 3b).First, gene duplication of the gene cluster trnW-trnC with intergenic spacer in ancestral order resulted in a tandemly repeated structure with repeated duplication: (trnW-trnC)-(trnW-trnC) 6 -(trnW-trnC).Subsequently, a random loss of genes occurred, including the upstream trnW gene, middle gene cluster (trnW-trnC) 6 , and downstream trnC.Finally, a new gene order was produced: trnC-trnW with an intergenic spacer consisting of six tandemly repeated units.
What is particularly noteworthy is that numerous studies have demonstrated a correlation between high rates of mitochondrial gene rearrangement and elevated nucleotide replacement rates in animal lineages [75].The non-synonymous substitutions (Ka) serve as a valuable parameter for assessing the evolutionary rate [76].Indeed, the Ka values for Oculogryphus and Stenocladius exceed 0.2, significantly surpassing those of Drilaster within Ototretinae (Figure S1), suggesting that a high rate of nucleotide substitution may be one contributing factor to gene rearrangement.However, it should be noted that higher Ka values (>0.2) do not necessarily result in gene rearrangement, as evidenced by Lampyrinae and Photurinae (Figure S1).

Phylogenetic Position of Oculogryphus
In the present study, we reconstructed the phylogeny of Lampyridae using molecular data, including Oculogryphus in the analyses for the first time.Our results indicate that the monophyly of Ototretinae has never been recovered, consistent with the recent molecular phylogenetic analyses [15,38,65].The clade consisting of Oculogryphus and Stenocladius always appeared separate from Drilaster and/or some related genera.Or this clade was grouped together with Ototretadrilus when the latter was included in the comprehensive molecular phylogeny (Figure 4a-d).While this result aligns with most molecular phylogenetic analyses [15], it contradicts morphological classification such as Janisova and Bocakova's work [22], in which Oculogryphus is not considered a member of Ototretinae, and Stenocladius along with Brachypterodrilus Pic, 1918, Baolacus Pic, 1915, and Falsophaeopterus Pic, 1911 belongs to an ototretine subgroup characterized by less prominent lateroposterior angles of the pronotum.In contrast, Drilaster and Ototretadrilus are included in the other subgroup exhibiting contrasting characteristics.
A single rearrangement occurred in the stem lineage leading to all Stenocladius mitogenomes and Oculogryphus (i.e., not a parallel acquisition).Specifically, at least a rearrangement between the trnW and trnC genes is a synapomorphy supporting the sister relationship of Oculogryphus and Stenocladius, but this was not found in the mitogenome of Drilaster [38].This finding is in line with Jeng's (2008) phylogenetic analysis based on morphological data, as well as the opinions held by Jeng and his coauthors [19][20][21].In light of this discovery, Oculogryphus should theoretically be classified in the same category as Stenocladius within the subfamily Ototretinae [16] at the moment.
Unfortunately, not all genera sampling or all molecular markers of Ototretinae are available in the phylogenetic analyses.It is particularly regrettable that the complete mitochondrial genome of Ototretadrilus has been unavailable until now.Therefore, the monophyly of Ototretinae cannot be clarified herein.Additionally, the systematic status of Ototretadrilinae (probably at least included Ototretadrilus, Stenocladius, and Oculogryphus) remains unclear.The determination of whether Oculogryphus is a member of true Ototretinae or Ototretadrilinae can only be made once the monophyly of the newly redefined Ototretinae [22] has been clarified.This requires more samplers representing all related ototretine genera (a total of 21 genera hitherto known) [21] to be included in future phylogenetic analysis.
No matter how, the gene rearrangement between trnW and trnC may serve as a valuable phylogenetic marker for testing the monophyly of Ototretinae, which is currently unsatisfactory in its definition.Thus, it is strongly encouraged that more mitogenomes are accumulated for Ototretinae and that extensive comparative analysis of their gene order is conducted, in order to improve classification and gain deeper insight into firefly evolution, which may be driven by the gene rearrangement.

Conclusions
In the present study, we have successfully sequenced and annotated the complete mitochondrial genome for the rare-species firefly genus Oculogryphus for the first time.The studied material of O. chenghoiyanae was determined through a comprehensive morphological identification.Our analysis showed that the characteristics of this sequence are similar to those of Stenocladius, including a rearranged gene order between trnW and trnC, as well as a large intergenic spacer (702 bp) between the two rearranged genes, within which six remnants of trnW (29 bp) exist.Furthermore, our phylogenetic reconstruction of Lampyridae incorporating this sequence in the cladistics analyses indicated that Oculogryphus is consistently grouped with Stenocladius and that gene rearrangement is a synapomorphy for this monophyletic clade.Based on these results, we suggest Oculogryphus be placed in Ototretinae at the moment.Nevertheless, further sequencing and analysis of additional genera within Ototretinae are needed to verify these obtained results and clarify the monophyly of the subfamily.This will also help us better understand the evolution of fireflies in general.

Supplementary Materials:
The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/insects15070464/s1, Figure S1: Values of the non-synonymous (Ka) substitution rates of 13 protein-coding genes among 43 species of Lampyridae.Order from left to right corresponds to order of phylogenetic tree in main text;  S1: Species used in dataset 1; Table S2: Species used in datasets 2 and 3; Table S3: Nucleotide composition features within Oculogryphus chenghoiyanae mitochondrial genome; Table S4: Mitochondrial genome organization of Oculogryphus chenghoiyanae.

Data Availability Statement:
The newly sequenced mitochondrial genome in this study has been uploaded to GenBank (ON985401).

Figure 1 .
Figure 1.Hind wings of Oculogryphus chenghoiyanae Yiu & Jeng, 2018, unfolded views.Scale bars: 1.0 mm.Different colors denote different material from different localities or on collecting dates to show variations within species: (A,C,D): specimens from Jiulianshan, Jiangxi, China; (B): specimens from Nanling National Natural Reserve, Guangdong, China (detailed information denoted in Examined Material part).

Figure 1 .
Figure 1.Hind wings of Oculogryphus chenghoiyanae Yiu & Jeng, 2018, unfolded views.Scale bars: 1.0 mm.Different colors denote different material from different localities or on collecting dates to show variations within species: (A,C,D): specimens from Jiulianshan, Jiangxi, China; (B): specimens from Nanling National Natural Reserve, Guangdong, China (detailed information denoted in Examined Material part).

Figure 3 .
Figure 3. (a) A linear map of the complete mitogenome of Oculogryphus chenghoiyanae and the intergenic spacer between trnW and trnC, showing the number, position, and nucleotide sequence of the gene remnant.(b) The putative mechanism of gene rearrangement.The size of the genes is not proportional.W* indicates a 29 bp trnW residue; W, C and Y are trnW, trnC and trnY, respectively; ND2 is the protein-coning gene ND2; NC means intergenic spacer.

Figure 3 .
Figure 3. (a) A linear map of the complete mitogenome of Oculogryphus chenghoiyanae and the intergenic spacer between trnW and trnC, showing the number, position, and nucleotide sequence of the gene remnant.(b) The putative mechanism of gene rearrangement.The size of the genes is not proportional.W* indicates a 29 bp trnW residue; W, C and Y are trnW, trnC and trnY, respectively; ND2 is the protein-coning gene ND2; NC means intergenic spacer.

Figure 4 .
Figure 4. (a) Bayesian inference phylogenetic tree of Lampyridae inferred based on ten mitogenome and three nuclear markers.Ototretinae is highlighted in red.(b-d) Taxon names and statistical support for branches of a; only Ototrtinae clade shown (except Brachylampis blaisdelli), see full trees in Figures S3 and S4.(e) Bayesian inference phylogenetic tree of Lampyridae based on PCGRNA with heterogeneous models CAT+GTR in PhyloBayes.The yellow star denotes the phylogenetic position of Oculogryphus chenghoiyanae.

Figure 4 .
Figure 4. (a) Bayesian inference phylogenetic tree of Lampyridae inferred based on ten mitogenome and three nuclear markers.Ototretinae is highlighted in red.(b-d) Taxon names and statistical support for branches of a; only Ototrtinae clade shown (except Brachylampis blaisdelli), see full trees in Figures S3 and S4.(e) Bayesian inference phylogenetic tree of Lampyridae based on PCGRNA with heterogeneous models CAT+GTR in PhyloBayes.The yellow star denotes the phylogenetic position of Oculogryphus chenghoiyanae.
Figure S2: Secondary structures of 22 tRNA genes of Oculogryphus chenghoiyanae; Figure S3: Bayesian inference phylogenetic tree inferred based on ten mitogenome and three nuclear markers dataset; Figure S4: Maximum-likelihood phylogenetic tree based on ten mitogenome and three nuclear markers dataset; Figure S5: Maximum-likelihood phylogenetic tree based on PCG12RNA dataset.Numbers on nodes represent bootstrap support; Figure S6: Maximum-likelihood phylogenetic tree based on PCGRNA dataset.Numbers on nodes represent bootstrap support; Figure S7: Bayesian inference phylogenetic tree based on PCG12RNA in MrBayes; Figure S8: Bayesian inference phylogenetic tree based on PCGRNA in MrBayes; Figure S9: Bayesian inference phylogenetic tree based on PCG12RNA with heterogeneous models CAT+GTR in PhyloBayes; Table