Pseudolebinthus lunipterus sp. nov.: a striking deaf and mute new cricket from Malawi (Orthoptera, Gryllidae, Eneopterinae)

This article presents an intriguing new cricket species of the tribe Xenogryllini discovered in Northern Malawi. This is the first case of mute and deaf species in the subfamily Eneopterinae; it shows no stridulatory apparatus on short male forewings and no tympana on either side of fore tibiae in both sexes. We introduce the new species and its complete mitogenome and assess phylogenetic relationships based on molecular data obtained from next-generation sequencing genome skimming method. Phylogenetic analyses place the new species within the genus Pseudolebinthus in Xenogryllini, as the sister species of Pseudolebinthus gorochovi Robillard. We describe Pseudolebinthus lunipterus sp. nov., provide illustrations of main morphology, male and female genitalia, photographs of living specimens and information about habitat and update the identification key for species of genus Pseudolebinthus. We discuss the differences between the new species and related taxa and the striking loss of acoustic communication in this cricket.


INTRODUCTION
Crickets have long been studied for their capacity to communicate with sound (Horch et al., 2017). Their mechanism of sound production by wing stridulation and their hearing system has been extensively studied during the last 50 years (Michelsen, 1998;Bennet-Clark, 1989;Gerhardt & Huber, 2002). What is less known is that many lineages within the cricket clade have independently lost their capacity to produce sound, and sometimes their hearing capacity too. Numerous examples of mute lineages are spread in the taxonomic literature about crickets (Otte & Alexander, 1983;Otte, 1992;Wang et al., 2018), some species being completely wingless and some retaining the wings while losing stridulatory structures at different degrees (Zuk, Rotenberry & Tinghitella, 2006;Pascoal et al., 2014). Recent studies have demonstrated that the loss of sound production structures on male forewings (FWs) could occur convergently and very rapidly in populations of Teleogryllus oceanicus (Le Guillou) as a result to strong selective pressures by a parasitoid fly (Zuk, Rotenberry & Tinghitella, 2006;Pascoal et al., 2014).
In many mute lineages of crickets, auditory tympana are retained after the tegminal stridulatory mechanism is lost (Otte, 1992), which could be linked with avoidance of bat predation. Species still able to fly but in which males have lost the stridulum usually retain the tympana (Otte & Alexander, 1983;Otte, Alexander & Cade, 1987). Species becoming both mute and deaf are relatively less common, even among the diversity of situations presented by crickets.
In this study, we describe the species Pseudolebinthus lunipterus sp. nov., a new eneopterine from Northern Malawi being both mute and deaf (Fig. 1). The new species is the first member of Eneopterinae showing no stridulatory apparatus on short male FWs and no tympana on either side of fore tibiae in both sexes. We provide illustrations about main morphology, male and female genitalia, photographs of living specimens and information about its natural habitat. We describe its complete mitogenome and assess phylogenetic relationships based on molecular data obtained by next-generation sequencing using the genome skimming method (Straub et al., 2012). We discuss the differences between the new species and related taxa, their phylogenetic relationships and the possible origins of muteness and deafness of these crickets.

Material and taxonomy
The new collected material comes from a field expedition in Malawi in The electronic version of this article in Portable Document Format will represent a published work according to the International Commission on Zoological Nomenclature (ICZN), and hence the new names contained in the electronic version are effectively published under that Code from the electronic edition alone. This published work and the nomenclatural acts it contains have been registered in ZooBank, the online registration system for the ICZN. The ZooBank Life Science Identifiers (LSIDs) can be resolved and the associated information viewed through any standard web browser by appending the LSID to the prefix http://zoobank.org/. The LSID for this publication is: urn:lsid:zoobank. org:pub:DA17C29A-B265-4819-A9E6-55FCB137E3F0. The online version of this work is archived and available from the following digital repositories: PeerJ, PubMed Central and CLOCKSS. Description follows terminologies as proposed by Robillard et al. (2014). Observations of external morphological characters and dissection of male and female genitalia were performed using a Leica stereomicroscopes MZ16. Terminologies for male FW venation follow Ragge (1955) and Robillard & Desutter-Grandcolas (2004a). Male and female genitalia were dissected from freshly killed specimens. Male genitalia were dissected by making a small slit between paraproct and subgenital plate. Female copulatory papilla was dissected out by cutting the membrane between ovipositor and subgenital plate. Dissected genitalia were cleared in 10% cold KOH solution and preserved in glass vials containing glycerine. Imaging of male and female genitalia were made using Canon EOS 60D Digital SLR camera on a Nikon stereomicroscope SMZ1500. To highlight the structural components of male and female genitalia, water solution containing a drop of JBL Punktol was used. To fix orientation and stabilization of genitalia for photography, a clear and viscous Power Plast Hand Sanitizer was used following Su (2016). The dry mounted adults were photographed with a Canon EOS 6D Digital SLR camera.
Measurements: (in mm, except for spine numbers) BL, body length in dorsal view, from fastigium to apex of abdomen; FIIIL, length of FIII; FIIIW, width of FIII; TIIIL, length of TIII; FWL, forewing length; FWW, forewing width (at the level of maximal width at about 1/3 of FWL); Ias, inner spines on TIII dorsal side above the spurs; Ibs, inner spines on TIII dorsal side between the spurs; Oas, outer spines on TIII dorsal side above the spurs; Obs, outer spines on TIII dorsal side between the spurs; OL, ovipositor length; PronL, pronotum length; PronW, pronotum width; TaIIIs, spines of third hind tarsomere, not including the apical spines: Ids, inner dorsal spines; Ods, outer dorsal spines; Ols, outer spines on lateral side of TaIII.

Laboratory methods
DNA extraction, PCR amplification and bank preparations were carried out at Service de Systématique Moléculaire of the MNHN. We extracted DNA from ethanol-preserved median legs for four newly collected specimens (two P. lunipterus sp. nov. and two Pseudolebinthus gorochovi Robillard, 2018) (see Table 1 for details about specimen vouchers). Total genomic DNA was extracted using a DNeasy Blood and Tisue Kit (Qiagen Inc., Venlo, Netherlands and Germany) following the manufacturer's instructions. For each newly generated extract, we amplified the mitochondrial gene maker 12S (12S rRNA gene, amplicon~400 bp) with the protocols described in Nattier et al. (2012) with the following primers and annealing temperatures: 12SF 5′-TACTATGT TACGACTTAT′3′, 12sr 5′-AAACTAGGATTAGATACCC-3′ at 48 C (Kambhampati, 1995). PCR products were sequenced with the Ion Torrent PGM platform in MNHN. Assembling and annotations were executed with Geneious Prime 2019.1.3 (Biomatter Ltd., Auckland, New Zealand, Oceania, www.geneious.com, Kearse et al., 2012).
One extract of the new species P. lunipterus (X28, MNHN-EO-ENSIF10720) was used for library preparation in a Genome Skimming approach (Straub et al., 2012). We assessed total DNA with a Qubit TM dsDNA High-Sensitivity Assay Kit (Life Technologies, Paisley, UK) with a Fluorescence Microplate Reader in 1.0 µL of sample. Prior to library preparation, we fragmented the DNA by sonication using BioRuptor Ò UCD-200 (Life Technologies and Invitrogen, Carlsbad, CA, USA) using 50 µL DNA sample and 50 µL TE buffer 0.1X. The molecular weight of the fragmented DNA sample was analyzed in agarose gel electrophoreses (3.0 µL DNA sample plus BG 1.0 µL; gel agarose 1% in TAE buffer (Tris-acetate-EDTA) 1.0X; migration buffer TAE 0.5X; migration time 20 min) before and after the sonification of the DNA. We then used the NEBNext Ò Ultra TM II DNA Library Prep Kit for Illumina (New England BioLabs, Ipswich, MA, USA; dsDNA protocol) with a modified version of the protocol based on Meyer & Kircher (2010). After library preparation, total DNA was quantified with a Qubit TM dsDNA (HS) Assay Kit using Qubit TM Fluorometer (Life Technologies, Carlsbad, CA, USA) in 1.0 µL of sample. Libraries were then analyzed with a Bioanalyzer 2100 DNA 1000 series II chip (Agilent Technologies, Santa Clara, CA, USA) (High Sensitivity DNA Assay). Pooled libraries were sequenced as paired-end reads (150 bp) on an Illumina HiSeq 3000 HWI-J0015 at the Genome and Transcriptome Platform of Toulouse (Genotoul, Toulouse, France).

Sequence analyses and mitogenome annotation
Sequencing reads from both paired-end libraries were imported in Geneious Prime 2019.1.3, then filtered and trimmed by quality using the BBDuk plugin (minimum quality score of 30 and minimum length of reads of 30 bp). Quality and length distribution of the sequences were inspected using FastQC v. 0.11.8 (Andrews, 2010) under the open-source application Galaxy (http://galaxyproject.org/) (Afgan et al., 2018). We then extracted sequences of interest from the total read using the Map to reference option in Geneious (Custom sensibility, fine tuning: iterate up to 10 times; Maximum Mismatches Per Read 30). The Mitochondrial genomes of Xenogryllus marmoratus (Haan, 1842) (Ma, Zhang & Li, 2019; GenBank Accession MK033622) and Cardiodactylus muiri Otte, 2007; GenBank Accession MG680938) were used as references. After removing the reference sequence from the resulting contig, a step of De Novo assemble (Sensibility: High sensibility/Medium) was performed in Geneious. The longest resultant contig (Number of reads 14,817; Sequence length 16,075 bp) was chosen as a seed and mapped with the filtered reads again (Custom sensibility, fine-tuning: iterate up to 25 times; max. Mismatches Per Read 10).
The consensus sequence (Threshold: Highest quality; Assign quality: Highest) was then circularized and annotated with Geneious with X. marmoratus as reference. Genome annotation based on sequence similarity was performed independently using Geneious and with MITOS (Bernt et al., 2013), available at http://mitos.bioinf.uni-leipzig.de/index.py using the invertebrate mitogenome genetic code.
A similar process was used to extract sequences of three nuclear genes: histone H3 (H3,~330 bp) and the sequences of the non-protein-coding genes corresponding to nuclear small ribosomal subunit (18S rRNA, 18S,~650 bp) and of the nuclear large ribosomal subunit (28S rRNA, 28S,~400 bp). Nuclear genes were assembled using references from the species P. gorochovi from . All the newly generated sequences are available on GenBank (Table 1).

Phylogenetic analysis
The cricket tribe Xenogryllini is composed of three genera and 13 valid species (including the new species): Xenogryllus Bolívar (eight species), Pseudolebinthus Robillard (four species) and Indigryllus Robillard & Jaiswara (one species). To infer the phylogenetic position of the new species, we refer to the recent molecular phylogeny of Xenogryllini  which included eight Xenogryllini species representing the three genera and eight species representing all four other tribes of the Eneopterinae subfamily, plus two more distant species belonging to the subfamily Gryllinae. We used DNA markers from eight genes, five from the mitochondrial and three from the nuclear genome based on Jaiswara,  and previous studies (Robillard & Desutter-Grandcolas, 2006;Nattier et al., 2012). The mitochondrial markers were partial sequences of the small subunit rRNA gene (12S, amplicon~400 bp), the large subunit rRNA gene (16S,~500 bp), of the cytochrome b gene (Cytb,~400 bp), and of the cytochrome c oxidase subunit 1 (CO1,~750 bp) and subunit 2 (CO2,~400 bp). Nuclear markers were partial sequences of protein coding histone H3 gene (H3,~330 bp), and partial sequences of two non-protein-coding genes corresponding to nuclear ribosomal subunits 18S rRNA (18S,~650 bp) and 28S rRNA (28S,~400 bp).
Newly generated sequences were added to the previous data set. We extracted the mitochondrial markers from the mitogenome of P. lunipterus sp. nov. See Table 1 for detailed information about taxon and molecular sampling.

Mitogenome of P. lunipterus
The mitogenome of P. lunipterus is 16,075 bp in length and has a typical circular structure (Fig. 2). The nucleotide composition of this genome has a GC content of 24.2%. The identity and position of 13 PCGs, 22 tRNA and 2 rRNA genes is detailed in Table 2.

Phylogenetic relationships
The alignment of all eight markers consists of 3,684 aligned base pairs (bp) for 34 terminals: 416 bp for 12S, 521 bp for 16S, 707 bp for CO1, 335 bp for CO2, 346 bp for Cytb, 328 bp for H3, 652 bp for 18S and 379 bp for 28S. The ML phylogenetic tree inferred from this data is shown in Fig. 3.
Most nodes of the tree have high bootstrap supports and are consistent with the recent study of Jaiswara, . All species represented by two or more terminals are found monophyletic. The subfamily Eneopterinae and the tribe Xenogryllini    Type locality. North Malawi, Mount Uzumara, S10 52′19,3″ E34 07′44,7″, 1,941 m.
Distribution. The species is only known from the type locality in Northern Malawi (Fig. 1C).
Etymology. The species name refers to the whitish wings, rounded in males and crescent-shaped in females, which look like tiny moons on the back of the dark body of these crickets when encountered at night.

Diagnosis
Size small, mostly dark brown with pale wings (Figs. 1, 3 and 4). Among Eneopterinae genera, the new species presents the characteristics of Pseudolebinthus: large lateral eyes (Figs. 5A-5C); brachypterous FWs barely reaching quarter of abdomen length in males ; male genitalia with long sclerotized lophi, close to that of P. gorochovi (Figs. 8 and 9); female ovipositor little differentiated but less pointed and thicker than in P. gorochovi (Fig. 10A). The new species is characterized by complete absence of tympana (unique feature among eneopterines) (Figs. 7A and 7B), absence of stridulatory apparatus on male FWs (Fig. 6), abdomen ventrally yellow with a wide black stripe (Fig. 5F), thick and short female ovipositor (Fig. 4D), and differences in male genitalia, including shape of pseudepiphallic parameres, shape of sclerite in ectophallic fold and endophallic apodeme with anterior lateral expansions. Description In addition to the characters of the genus: body mostly dark brown with pale wings (Figs. 1, 3 and 4). Head rounded dorsally, with five black bands separated by thin yellow lines, including two thin bands posterior to eyes and three wider median bands, sometimes partly faded in median area (Figs. 5A-5C). Fastigium black, setose, with thin yellow margins, its apex with a wide yellow band surrounding median ocellus. Eyes occupying ca. 40% of total head width in dorsal view. Ocelli forming a wide triangle. Scapes short, wider than long, black with a yellow spot on upper inner side; rest of antennae brown. Face  (Fig. 5C), almost entirely black except: a median vertical line connected to the yellow band at fastigium apex, two yellow spots above epistomal suture, a yellow spot below each eye; mouthparts black, mottled with yellow. Lateral part of head entirely black (Fig. 5B). Pronotum dark brown dorsally, its lateral margins thinly underlined with yellow (Figs. 5B and 6A); lateral field black (Fig. 5B). TI with no tympana (Figs. 7A and 7B). FWs very short in both sexes (Fig. 4), hind wings absent. Fore and median legs with black femora, dark brown tibiae (Fig. 4); base of tarsomeres yellow, apex darker. FIII external face bicolor: dorsal half black, ventral half yellow; knees black ( Fig. 7C); TIII dark brown; tarsomere III yellow basally, apex dark brown. Abdomen rather long and fusiform (Fig. 4), yellow ventrally with a wide black stripe, including subgenital plate (Fig. 5F). Subgenital plate slightly indented apically in both sexes (Figs. 5F and 10B). longitudinal veins, including R, Sc and M, the latter separating dorsal and lateral fields; one more ventral vein sometimes near FW base. Hind wings vestigial. Metanotum with glandular structures (Fig. 5E); gland morphology close to that of other Pseudolebinthus, with a bunch of long setae on basal margin, a wide median process on scutum, and basal edge of scutellum raised medially and carrying a bunch of setae orientated anteriorly; posterior part of mesonotum setose and extended posteriorly, covering anterior part of metanotal scutum. Male subgenital plate elongate (Fig. 5F). Male genitalia (Figs. 8 and 9). Pseudepiphallic sclerite as long as rami, widened laterally near base of rami. Pseudepiphallic lophi thin and parallel, longer than in P. gorochovi, twisted ventro-apically; their apex with a small dorsal expansion. Pseudepiphallic parameres dorsal lobe triangular, longer than ventral lobe; ventral lobe oriented anteriorly forming an apical fold. Ectophallic fold with a strong ventro-lateral sclerotization forming a transversal bridge (Figs. 8B, 9C and 9D) slightly extended posteriorly within pointed membranous apex. Ectophallic apodemes long and parallel. Endophallic sclerite with a small median area, with wide lateral arms; endophallic apodeme including two small dorso-anterior arms at anterior apex of endophallic sclerite and a narrow apical transverse crest extended laterally, underlying arms of endophallic sclerite.
Habitat and life history. Pseudolebinthus lunipterus lives on low vegetation in herbaceous areas near forest hedge or in open areas along trails in forest (Figs. 1A and 1B). Adults and juveniles have been found active at night on top of vegetation, but can also be found lower within vegetation during the day. Remarkably, the species lives in syntopy with P. gorochovi in the type locality, where adults and juveniles of both species are quite abundant. One juvenile specimen of P. lunipterus has been observed eating a dead insect on a low leaf on vegetation (Fig. 11B). Females maintained in controlled laboratory conditions (20-22 C, 14-10 day-night cycle) with a single male produced 46-50 offspring (n = 2) during their life; first hatchings started 42-49 days after first mating and occurred on a period of 35-66 days.

Multiple losses of acoustic communication
In this article we described the species P. lunipterus sp. nov., a new eneopterine cricket from Northern Malawi being both mute and deaf. This new species is the first reported case showing complete absence of stridulatory apparatus (no stridulatory file, harp and mirror) in this cricket clade, associated with absence of tympana on both sides of fore tibiae.
Our taxonomic study suggests that P. lunipterus belongs to the tribe Xenogryllini despites all its special morphological features. The phylogenetic analysis shows that it is closely related with at least one other species of the genus Pseudolebinthus (Fig. 3). This phylogenetic position has two interesting consequences: First, this is the first case of muteness documented in the clade Xenogryllini. Two other cases of loss of acoustic communication were previously reported in eneopterines (Table 4): one occurred in the tribe Nisitrini and concerns the apterous species of the genus Paranisitra, which has diverged from its sister genus Nisitrus ca. 55 Ma, before diversifying in the Philippines after 12.5 Myrs Baroga-Barbecho et al., 2019). The second other mute lineage is the genus Swezwilderia in the tribe Lebinthini; species of Swezwilderia possess long wings, but lack stridulatory structures. This genus has diverged from its sister group ca. 48.4 Ma, and has diversified in Fiji and Samoa after 21.6 Myrs . In addition to their phylogenetic independence, these three losses of calling abilities are structurally different, one occurring through the loss of complete wings and the two others consisting of losses of stridulatory structures either on fully formed wings (Swezwilderia) or on reduced wings (P. lunipterus). These different combinations of traits support the hypothesis that these three losses of acoustic communication are convergent in Eneopterinae. The phylogenetic context of each loss may explain these different configurations in relation with two other functions of the wings of insects: flight and protection. In Swezwilderia, the wings might have been retained in association with keeping flying capacities, while short wings might have been  (Jaiswara, Dong & Robillard, 2018, 2019. Interestingly, the short wings in the new species are shorter than that of other Pseudolebinthus species, but remain just long enough to cover the glands beneath (Figs. 5B, 5E and 6). In contrast, Paranisitra is apterous but has also lost the metanotal glands while diverging from Nisitrus.
The second interesting observation about the phylogenetic position of P. lunipterus is that this is the only mute species occurring within a genus, while others mute cases concern clear-cut genera which show strong divergence from their sister lineages. Even if genera do not represent evolutionary units, it is interesting to notice that P. lunipterus and P. gorochovi are separated by very short branches in the phylogenetic tree (Fig. 3). A molecular dating analysis of Pseudolebinthus will be necessary with a better taxonomic sampling, but this situation suggests that the loss of acoustic communication in P. lunipterus is likely a recent event, which is recalling the loss of sound production structures occuring convergently and very rapidly within some populations of T. oceanicus as a result to strong selective pressures by a parasitoid fly (Zuk, Rotenberry & Tinghitella, 2006;Pascoal et al., 2014). Analogous selective pressures might be responsible for the loss of sound production in P. lunipterus.
The most unique feature of P. lunipterus at the scale of the subfamily is the deafness of the species. Other cases of deaf crickets have been documented in other clades, but this is the only case known in eneopterines. In many mute lineages of crickets, auditory tympana are retained after the stridulatory mechanism is lost (Otte, 1992). In species that are still able to fly, but in which males have lost the stridulum (such as in species of Swezwilderia), the tympanum is usually retained, which is supposed to be linked with avoidance of bat predation in flight (Otte & Alexander, 1983;Otte, Alexander & Cade, 1987). Species becoming both mute and deaf, such as P. lunipterus, are less common. This combination of traits might be explained by predator avoidance selecting for mute crickets in lineages having ancestrally lost their flying capacities (all Pseudolebinthus). In such cases, maintaining tympana might not be necessary. Interestingly, this hypothesis does not hold with the case of Paranisitra, which lost the wings while retaining hearing capacities (or at least external organs, since the hearing capacities of Paranisitra have never been evaluated).

Generic allocation
The morphological study of the new species shows that it shares all the characteristics of the genus Pseudolebinthus in terms of general morphology, body size and main features of male genitalia. On the other hand, the new species differs by important characters such as FW length, absence of stridulatory file and tympana, and by the shape of female copulatory papilla. Such differences suggest that the new species might have been considered as an easily recognized new genus close to Pseudolebinthus. This hypothesis has been tested using the molecular data and the phylogenetic relationships. Although it is too early to conclude that the new species is nested within Pseudolebinthus (only one previously described species was successfully sequenced here), our results clearly show that the new species is very close to P. gorochovi, and the short stem branches in the phylogeny leading to the new species and P. gorochovi strongly support the hypothesis that the new species should be considered as a particular species of Pseudolebinthus.

CONCLUSION: CRICKETS OF MALAWI
The diversity of crickets in Eastern Africa in general, and Malawi in particular, has been underestimated, understudied and undersampled. This is at least the case for the members of the tribe Xenogryllini which were recently revised (Jaiswara, Dong & Robillard, 2018, 2019. Despite the large amount of data considered in these systematic studies (several hundreds of specimens studied across the study of the largest natural history museum collections), they gathered very little information about the species of Pseudolebinthus, known by a few specimens each.
A single recent field trip in Malawi allowed us to re-discover two of the previously described species of the genus, which are in fact common species, and it allowed documenting the acoustic features of their calling songs and their ecology (T. Robillard et al., 2020, in prep.). Interestingly, these findings allowed us to discover P. lunipterus, a completely different new species belonging to the Xenogryllini lineage, but with strikingly new morphological features. This finding reveals that more taxa probably remain unrecorded in the whole Eastern African region, as suggested by the large amount of new species and genera recently discovered in this region for other clades of orthopteran insects (Hemp et al., 2018;Hemp & Heller, 2019). More taxonomic surveys with appropriate collecting methods in regions where there is zero record about these crickets, such as other regions of Malawi, but also Zimbabwe, Zambia, Western Mozambique and Northern South Africa, are thus necessary to explore this part of African biodiversity.
Tony Robillard is an Academic Editor for PeerJ.

Author Contributions
Karen Salazar conceived and designed the experiments, performed the experiments, analyzed the data, prepared figures and/or tables, authored or reviewed drafts of the paper, and approved the final draft. Raymond J. Murphy analyzed the data, authored or reviewed drafts of the paper, and approved the final draft. Marion Guillaume performed the experiments, analyzed the data, and approved the final draft. Romain Nattier conceived and designed the experiments, analyzed the data, prepared figures and/or tables, authored or reviewed drafts of the paper, and approved the final draft. Tony Robillard conceived and designed the experiments, performed the experiments, analyzed the data, prepared figures and/or tables, authored or reviewed drafts of the paper, and approved the final draft.

Field Study Permissions
The following information was supplied relating to field study approvals (i.e., approving body and any reference numbers): The Forestry Research Institute of Malawi (FRIM) granted a collection and export permit for this study (EAD-12-07-087-18-20a).

DNA Deposition
The following information was supplied regarding the deposition of DNA sequences: Sequence data are available at NCBI under GenBank Accession Numbers listed in Table 1.

Data Availability
The following information was supplied regarding data availability: Specimens are deposited in the MNHN specimen data base (https://science.mnhn.fr/ institution/mnhn/search?lang=fr_FR) under number given in the section "type material" as numbers "MNHN-E0-ENSIF ÃÃÃ ".