Phylogeny of the Chinese Subgenera of the Genus Homoneura (Diptera, Lauxaniidae, Homoneurinae) Based on Morphological Characters

Simple Summary The genus Homoneura is the most speciose genus of Lauxaniidae. However, no phylogenetic study of Homoneura has been published using morphological data, despite the high diversity and ecological significance. Therefore, we present the first morphological phylogeny of Homoneura. The monophyly of the genus Homoneura and the subgenus Homoneura is not supported. The monophyly of the subgenera Euhomoneura and Neohomoneura is supported, as well as the sister relationship between the subgenera Chaetohomoneura and Neohomoneura. These results provide a comprehensive framework and information toward future research of this genus. Abstract The genus Homoneura comprises over 700 described species in eight known subgenera distributed worldwide and has the highest species richness of Lauxaniidae. Five subgenera and more than 200 species have currently been recorded from China. Despite its high diversity, the monophyly of Homoneura and its subgenera, and the phylogenetic relationships among its subgenera remain to be investigated. One maximum-parsimony tree was generated based on 105 morphological characters scored from 24 species, representing all five subgenera of Homoneura recorded from China. The results did not support the monophyly of the genus Homoneura and subgenus Homoneura. The subgenus Chaetohomoneura is a sister to subgenus Neohomoneura. The monophyly of the subgenera Euhomoneura and Neohomoneura is supported. Much of the current classification of the genus Homoneura needs a revision before taxonomy can reflect natural groupings.


Introduction
The family Lauxaniidae is species-rich and globally distributed, with three subfamilies, more than 170 genera, and nearly 2100 described species [1]. Lauxaniids have a variety of habitats [2,3], mainly scavenging and fungus feeding [4,5], and have the habit of visiting flowers [6,7]. They play a very important role in the ecological system, degrading organic matter, plant pollination, and maintaining ecological balance [8][9][10]. In addition, lauxaniids are sensitive to environmental change. They have been used as an indicator organism for environmental change assessments of farmland ecosystems in Europe and have also become one of the research hotspots of terrestrial ecosystem assessment indicators [11].
The genus Homoneura is the most speciose genus of Lauxaniidae. Currently, it contains eight subgenera and more than 700 described species worldwide, among which five subgenera and more than 200 species are recorded from China [1]. It is also one of the most diverse genera of the Acalyptratae. Homoneura is widely distributed in all major animal geographical areas except for the Neotropical region.
Stuckenberg hypothesized the relationship among 19 genera of the subfamily Homoneurinae based on morphological characters and divided 19 genera into three groups [12]. Kim used numerical methods to cluster Australian members of the genera Homoneura, Trypeisoma, and allied genera by their similarities in morphology [13]. In the only existing molecular phylogenetic study of the generic level relationships of Lauxaniidae, Shi et al. explored the phylogeny of the subgenus Homoneura based on two mitochondrial and two nuclear genes [14]. The monophyly of the subgenus Homoneura was not recovered. No phylogenetic study of the genus Homoneura has been published using morphological data. Despite the high diversity and ecological significance, the monophyly of the genus Homoneura and its subgenera, and the phylogenetic relationships among the subgenera remain to be investigated.
In this study, we use 105 morphological characters to reconstruct the phylogeny based on all five subgenera of Homoneura that are recorded from China, aiming to (1) test the monophyly of the genus Homoneura; (2) test the monophyly of Chinese subgenera of Homoneura; (3) investigate relationships among five Chinese subgenera of Homoneura.

Morphological Study and Terminology
General terminology follows Cumming & Wood and Gaimari & Silva [15,16]. Genitalia preparations were made by removing and macerating the apical portion of the abdomen in cold saturated NaOH for six hours, then rinsing and neutralizing them for dissection and study. After examination in glycerine, they were transferred to fresh glycerine and stored in a microvial pinned below the specimen or moved to an ethanol tube together with the wet specimens. Most characters were illustrated using photographs and line drawings. Photographs were taken using a Canon EOS6D microscope (Canon, Tokyo, Japan) and stacked using HELICO FOCUS v7.0.2.0 (Helicon Soft, Kharkiv, Ukraine). Line drawings were drawn with Adobe Illustrator 2021 v25.2.1 (Adobe, San Jose, CA, USA).
The studied specimens are deposited in the Insect Collection of Henan University of Science and Technology (HAUST).

Cladistic Analysis
The phylogenetic reconstruction was conducted using maximum-parsimony analysis using NONA v2.0 (Goloboff, P.A., Tucuman, Argentina) with a heuristic search by 1000 replications, and TNT with 1000 random-addition traditional searches [17,18]. All characters were initially equally weighted. Branch support values were verified through bootstrap analyses on NONA 2.0 with 100 replications. The Bremer support value or decay index for the resultant strict consensus tree was calculated using TNT [19]. The unambiguous characters were mapped on the tree using WinClada version v1.00.08 (Nixon, K.C., Ithaca, NY, USA) [20].

Phylogenetic Analysis
Maximum-parsimony analyses yielded one maximum-parsimonious tree [branch length = 337, consistency index (CI) = 0.36, retention index (RI) = 0.56]. The maximumparsimony tree is shown in Figure 10, Bootstrap values (BS) and Bremer support (B) values are presented next to the nodes.

Cladistic Analysis
The phylogenetic reconstruction was conducted using maximum-parsimony analysis using NONA v2.0(Goloboff, P.A., Tucuman, Argentina) with a heuristic search by 1000 replications, and TNT with 1000 random-addition traditional searches [17,18]. All characters were initially equally weighted. Branch support values were verified through bootstrap analyses on NONA 2.0 with 100 replications. The Bremer support value or decay index for the resultant strict consensus tree was calculated using TNT [19]. The unambiguous characters were mapped on the tree using WinClada version v1.00.08(Nixon, K.C., Ithaca, NY, USA) [20].

Phylogenetic Analysis
Maximum-parsimony analyses yielded one maximum-parsimonious tree [branch length = 337, consistency index (CI) = 0.36, retention index (RI) = 0.56]. The maximumparsimony tree is shown in Figure 10, Bootstrap values (BS) and Bremer support (B) values are presented next to the nodes.

Monophyly and Relationships among Chinese Subgenera of Homoneura
The monophyly of the subgenera Neohomoneura and Euhomoneura was supported, while the subgenus Homoneura was recovered as polyphyletic (clades 4, 5, and 6). Due to the limited taxon sampling, the monophyly of the subgenera Minettioides and Chaetohomoneura could not be tested.

Discussion
Five non-Homoneura Homoneurinae genera were included in the present study intended to be an outgroup, but deeply nested into Homoneura based on four synapomorphies on the wing: spot on crossvein r-m absent (53:1), wing 2nd (between R 1 and R 2+3 ) section/3rd (between R 2+3 and R 4+5 ) section < 3X (60:1), area of transparent area or light-yellow area of wing/area of wing spot area ≤ 1 (49:1), and area of transparent or light-yellow area above wing R 4+5 /area of wing spot above R 4+5 ≤ 1 (50:1). The monophyly of genus Homoneura, therefore, was not supported by our analysis.
The non-monophyly of the genus Homoneura and the subgenus Homoneura was also detected based on molecular data. Shi et al. obtained two mitochondrial genes (COI, 16S-rRNA) and two nuclear genes (Elongation factor 1-α, 28S-rRNA), and reconstructed gene trees using four different methods (NJ, ME, MP, and ML). Among 16 resulted gene trees, the monophyly of genus Homoneura and subgenus Homoneura was never recovered (they always had Minettia, or Sapromyza, or both nested) [14].
The sister relationship between the subgenera Chaetohomoneura and Neohomoneura was also suggested by Sasakawa [21]. Sasakawa proposed two potential 'plesiomorphic' characters to support Chaetohomoneura + Neohomoneura, among which mid tibia with posterior seta (44:1) was found to be the synapomorphy by our current analysis. Another character, acrostichal seta no less than seven rows (34:0) was found to be homoplastic in multiple species of the subgenera Minettioides and Homoneura. To alleviate the difficulty of distinguishing these two subgenera, Shi et al. summarized three characters: Chaetohomoneura with two supra-alar setae, four strong apical ventral setae on mid tibia, and posterior ventral setae often present in mid femur; Neohomoneura with one supra-alar seta, three strong apical ventral setae on mid tibia, and posterior ventral setae often absent from mid femur [22]. We found that additional characters could help distinguish these two subgenera: Chaetohomoneura with posterior ventral seta in mid femur present, and Neohomoneura with wing with four spots, separately on tip of R 2+3 , R 4+5 , M 1, and crossvein dm-cu.
Sasakawa suggested intra-alar seta absent as a synapomorphy to support the monophyly of subgenus Euhomoneura + subgenus Homoneura [21]. Based on the present analysis, we found that the character is plesiomorphic on our tree and present in all species of non-Minettioides Chinese subgenera of the genus Homoneura. The synapomorphies support the monophyly of subgenera Euhomoneura, which is consistent with Sasakawa (one pre-sutural dorsocentral setae (32:1) and two post-sutural dorsocentral setae (33:2)).

Conclusions
This study presents the first morphological phylogeny of Homoneura, based on 105 characters of adults and 24 species representing all five subgenera of Homoneura recorded from China, underpinning our understanding of the phylogenetic relationships in the group.
Our results show that the monophyly of the genus Homoneura and subgenus Homoneura is not supported. Additionally, the results show that the monophyly of the subgenera Euhomoneura and Neohomoneura is supported, as well as the sister relationship between the subgenera Chaetohomoneura and Neohomoneura. Due to our regional taxon sampling, our result are premature to propose a new classification for genus or subgenus Homoneura, but we discovered the urgent need to revise this diverse group. Future studies with global taxon sampling, morphological evidence from multiple life stages, and molecular data are needed to reconstruct the phylogeny of Homoneurinae and revise the classification.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/insects13080665/s1, Table S1: The species studied, Table S2: Morphological dataset used for the analysis of the phylogeny.