A New Termitophilous Genus of Paederinae Rove Beetles (Coleoptera, Staphylinidae) from the Neotropics and Its Phylogenetic Position

We describe a new genus and species of Paederinae rove beetles, Ruptor cordatus gen. et sp. nov., which lives in the arboreal nests of the termite Labiotermes labralis (Holmgren, 1906) in the Amazon lowlands of Peru. The morphology of Ruptor gen. nov. is highly derived, apparently due to its close association with the termite host, and thus, morphologically, the genus cannot be classified further than Lathrobiini incertae sedis. In order to address the sister-group relationships of Ruptor gen. nov., we conducted a molecular phylogenetic analysis based on seven gene fragments. The analysis indeed resolved the genus as a member of the tribe Lathrobiini and placed it nested within the informal clade of ‘Medonina and allied taxa’. We provide a morphological comparison of the new genus with all known myrmeco- and termitophilous representatives of the subfamily, and to the extent possible, we illustrate other relevant and poorly known Neotropical Paederinae inquilines. Supplementary Information The online version contains supplementary material available at 10.1007/s13744-022-00946-x.


Introduction
Paederinae is one of the most diverse and abundant subfamilies of the family Staphylinidae (rove beetles), comprising more than 220 genera and 7600 species (Żyła et al. 2021). It is a globally distributed subfamily, well represented in higher latitudes but much more diverse in the tropics. Notably, 99 genera and 1684 described species are confined to the Neotropical region alone (Asenjo et al. 2019). The true Paederinae diversity is expected to be even higher in this region since the main body of the taxonomic work hitherto conducted on that group was limited to the Holarctic and, to a lesser degree, the Neotropical, Indomalayan and Afrotropical fauna.
Outdated subtribes and many poorly delimited genera in Paederinae, i.e. taxonomic groupings not coinciding with phylogenetic lineages, greatly hinder the taxonomic work on this subfamily in the Neotropics and other poorly studied regions. The situation gets even more complicated when one considers highly specialised myrmecophiles or termitophiles, in which their peculiar morphological adaptations to social parasitism lead to odd habitus and drastic modifications that mask diagnostically or phylogenetically valuable morphological characters.
Within Staphylinidae, Paederinae is one of a few subfamilies where social parasitism, in particular myrmeco-and termitophily, has evolved. Inquiline paederines are especially diverse in the Neotropics (Seevers 1957;Kistner 1979). Thus, we faced a complex task in which we had to identify four specimens of an odd-looking termitophilous paederine species, not matching any known genus, which we (A.K.H., J.K. and A.S.) collected in the nests of the termite Labiotermes labralis (Holmgren, 1906) (Fig. 1) in the Amazon lowlands of Peru (Fig. 2).
The objective of this paper is to describe a new genus and species of Paederinae rove beetles. Given the emerging backbone phylogeny of Paederinae based on total-evidence data (Żyła et al. 2019, 2021), as well as the availability of the DNA-grade material for this newly collected species, we decided to conduct a phylogenetic analysis to determine the lineage where this new taxon belongs and complement its taxonomic treatment.

Deposition of taxa
All specimens of the new species were acquired by the authors during a field trip to the Peruvian Amazon under the permit number RDG 0328-2017-SERFOR-DGGSPFFS/ RDG 356-2017-SERFOR-DGGSPFFS. Following the conditions of the permit, the male holotype and a female paratype are deposited at the Museo de Historia Natural de la Universidad Nacional Mayor de San Marcos in Peru (MHN-UNMSM). Two remaining female paratypes are kept at the Natural History Museum of Denmark in Copenhagen (NHMD). Information on the deposition of taxa used for phylogenetic analysis is given in Supplementary Table 1.

Microscopy and illustrations
The specimens were examined using Leica M205 C and Leica M125 stereoscopes. All photos were taken with either a Canon EOS 5D Mark III digital camera with a Canon MP-E 65 mm Ff2.8 1-5 × macro lens or Canon EOS 6D fitted on a Zeiss Axioskop slide microscope at 10 × both using the remote shooting function of EOS Utility v3.4.30.0. Photostacking was achieved using Zerene Stacker (Zerene Systems LLC, 2012) and the photos were edited in Adobe Photoshop 2021. Line drawings were made using images or camera lucida sketches that were digitally inked in Adobe Illustrator 2021.

Taxon sampling and outgroup for phylogenetic analysis
In total, 51 taxa were included in the dataset (Supplementary  Table 1). We sampled all currently recognised tribes and subtribes of Paederinae with a focus on Lathrobiini where the newly described genus was suspected to belong based on morphological observations. We added representatives of closely related subfamilies, namely Staphylininae and Xantholininae (sensu Żyła and Solodovnikov 2020) as the closer related outgroup and representatives of Oxyporinae, Tachyporinae and Mycetoporinae as the more distantly related outgroup. Unfortunately, we were unable to include many taxa which would ideally be represented in such an analysis based on the morphological examination of our target termitophilous species. This especially applies to the many species and genera currently placed in Medonina, because they are unavailable as DNA-grade material.

GenBank data
To construct our molecular matrix, we used seven gene fragments: the nuclear protein-encoding genes carbamoylphosphate synthetase (CADA and CADC), topoisomerase I (TP), arginine kinase (ArgK), wingless (Wg), the mitochondrial protein-encoding cytochrome c oxidase I (COI) and the nuclear ribosomal 28S. The GenBank accession numbers of all sequences are given in Supplementary Table 1. All sequences, except for the new genus, were already used in Żyła et al. (2021).

DNA extraction, amplification and sequencing
For the DNA extraction of Ruptor gen. nov., we used the Qiagen DNeasy Blood and Tissue kit (Venlo, Netherlands) using the protocol for animal tissue with a prolonged lysis time (12-16 h). The abdominal apex was removed from the specimen and used for the non-destructive extraction (including segment VIII, the genital segment and the aedeagus). The extract was stored in a -20 °C freezer for the period of the project, while the physical voucher (holotype) was pinned on a card mount with glycerin filled microvial for the genital segments and aedeagus which is deposited at MHN-UNMSM. We used the same protocols as in Żyła et al. (2021) for amplification, sequencing, sequence editing and assembly. Polymerase chain reactions were performed in 25 μL reactions. The reaction consisted of 2 μL of DNA extract, 4 μL of 5 × HOT FIREPol Blend Master Mix Ready to Load With 10 mM MgCl2 (Solis BioDyne), 0.5 μM of each primer and 15 μL distilled water. In the case of a second reaction for the genes using nested PCRs, only 1 μL of DNA extract was used, with the analogous increase in distilled water. Sanger sequencing and purification of target genes was done by Macrogen (Amsterdam, Netherlands).

Sequence alignment
Sequences were newly aligned in Geneious v9.1.7 (Biomatters Ltd, Auckland, New Zealand) using the MAFFT plugin v1.3.6, based on MAFFT (Katoh et al. 2002). 28S was aligned using the E-INS-i algorithm of MAFFT and ambiguously aligned regions were identified and removed with the server version of Gblocks (Talavera and Castresana 2007). We allowed gap positions within the final blocks and less strict flanking positions but did not allow many contiguous nonconserved positions. The resulting 28S alignment was 795 bp and had very few, scattered and usually single-nucleotide gaps. Individual gene alignments were concatenated with the 'concatenate' function of Geneious. The concatenated sequence alignment is provided in Supplementary File 1 in fasta format.

Phylogenetic analysis
A matrix of molecular data (4566 bp) for the total number of taxa under study (51) was analysed using the Bayesian Inference method. The alignment was initially partitioned by gene and, for protein-encoding genes, by position. The optimal partitioning scheme and the corresponding models of nucleotide evolution were determined by PartitionFinder v. 2.1.1 (Lanfear et al. 2016) using the Bayesian Information Criterion. All models were considered, branch lengths were unlinked and the search was set to the 'greedy' algorithm (Lanfear et al. 2012). Bayesian inference (BI) was conducted in MrBayes v3.2.6 (Ronquist et al. 2012) running on CIP-RES Science Gateway v3.3. (phylo.org). All analyses used four chains (one cold and three heated) and two runs. The analysis was run for 10 million generations. The script used for the analysis is given in Supplementary File 2 in.nex format. Convergence of the runs was visualised in Tracer v1.7 (Rambaut et al. 2018), and by examining potential scale D. Żyła et al. 284 reduction factor (PSRF) values and the average standard deviation of split frequencies in the MrBayes output. Nodes with (BI) posterior probability (PP) > 0.95 were considered well supported, nodes with PP = 0.90-0.94 moderately supported, those with PP = 0.80-0.89 weakly supported and nodes with PP < 0.79 were considered to be unsupported.

Identification of the termite host
For identification of the termite, DNA was extracted from the leg of a soldier specimen and from the body of a worker specimen using the above-described extraction protocol. The two specimens were both with the following data 'PER17-20a, PERU: Amazonia, Loreto region, Maynas Province, Allpahuayo-Mishana NP, 20 km S of Iquitos, 4.IX.2017, 100-200 m, 3°58.725′S 73°25.497′W, rainforest, in termite nest, leg. A. Hansen, J. Kypke, A. Solodovnikov'. The Pat and Jerry region of the COI gene was amplified and sequenced from both specimens using the same protocol as described above. The consensus of two sequences (100% identity, GenBank: OK310885 and OK310886) was compared to the GenBank repository resulting in the best match of 98.0% to a full mitochondrion of Labiotermes labralis (GenBank: KY436201). To confirm this identification, these and various other specimens of the termites (both worker and soldiers) were taken through the key in Constantino et al. (2006) resulting in the same determination, Labiotermes labralis. Given that this is the only species in the genus that builds arboreal nests and that several methods (genetic and morphological) produced similar results, we feel confident in our identification. can be recognised among all Paederinae by the following combination of characters: compact, dorso-ventrally flattened body with dense even cover of setiferous punctures with short pale setae; trapezoid-shaped head with broad frontal area, pronounced posterior angles and small notch on straight posterior margin of head above moderately narrow (ca. 1/3 of head width) neck; in maxillary palps penultimate (third) palpomere slightly longer and wider than second, apical (fourth) palpomere small and acicular; strongly transverse pronotum and notably shortened antennae and legs; protarsi wide, protarsomere 4 simple, not bilobed; protibial combs placed longitudinally ( Supplementary Fig. 1a); ctenidium present only on posterior side of hind tibia ( Supplementary Fig. 1b); aedeagus without parameres.

Subfamily
Description. Medium size beetle (total body length ca. 5 mm); body robust and compact, somewhat flattened; body surface with dense but fine irregular punctation. Appendages shortened and robust.
For distribution and bionomics see below species description.
Comparison. Ruptor gen. nov. is rather distinct among all known genera of Paederinae, including specialised termitophilous and myrmecophilous forms (Fig. 5), by its habitus alone. Its compact nite VIII; f male tergite VIII; g male sternite IX surrounded by visible parts of tergite IX; h male tergites IX and X; i female sternite IX (divided into basal and apical gonocoxites) and visible parts of tergite IX. a, b and c Scale = 5 mm; d, e, f, g and i scale = 2 mm. Abbreviations: agcx, apical gonocoxite; bgcx, basal gonocoxite; bstr, basisternum; fcstr, furcasternum; gs, gular suture; lgl, ligula; msstp, mesosternal process; mt, mentum; prst, prostheca; psts, prosternal suture; sbmt, submentum; sp, spiracle; sIX, sternite IX; sX, sternite X; tIX, tergite IX; tX, tergite X D. Żyła et al. 286 dorso-ventrally flattened body with short appendages, lack of tuberculose sculpture or carinae on forebody, and head with sharp posterior angles (in dorsal view) combined with transverse pronotum easily tell it apart from other, often poorly known, inquiline (or presumably so) Lathrobiini incertae sedis genera, namely Bolbophites Fauvel, 1904, Ecitobium Wasmann, 1923, Ecitonides Wasmann, 1894 (Fig. 5a), Ecitosaurus Fischer, 1943, (Fig. 5b) Ecitotropis Borgmeier, 1936, Labidophites Borgmeier, 1956, Mimophites Fauvel, 1904 (Fig. 5c), Synecitonides Reichensperger, 1936 (Fig. 5d) and Monista Sharp, 1876 (Fig. 5e) However, it readily differs from them by a somewhat rugose (not glabrous) disc of head and pronotum, by the presence of ctenidium only on the posterior side of hind tibia, as well as by the longitudinal placement of the protibial combs. Also, Ruptor may distantly resemble members of the subgenus Eurysunius Reitter, 1909 of the genus Astenus (Astenina). It differs from all Astenina at least by its simple not bilobed protarsomere 4, by its prosternum not being expanded under the front coxae and not fused with the pronotal hypomera, and by its short and stout mandibles. The new genus differs from all myrmecoand termitophilous Pinophilini and Paederini by having the typical for Lathrobiini small and unmodified apical (fourth) maxillary palpomere; it differs from the myrmecophilous members of Scopaeina by its larger labrum and lack of trichobothrium on the head; it differs from myrmecophilous Stilicina by the wider neck and the prosternum not being expanded under the front coxae and not being fused with the pronotal hypomera.
Etymology: The genus name is a Latin noun of masculine gender meaning burglar or housebreaker. It refers to the social parasitism biology of this taxon which was found in the termite's 'house'.

Description
Body colouration varies from pale brown to dark brown; head and elytra always darker; pro-and mesotibia darker than metatibia. Anterior and posterior margins of head, pronotum and elytra evenly covered with short golden setae; rest of head and abdomen mainly with grey setae; disc of head appearing dull due to microsculpture between fine punctation. Antennae: antennomere 1 as long as two following antennomeres; 2 and 3 elongate; 4 to 8 clearly transverse, 9 and 10 slightly transverse, 11 elongate, 1.5 × longer than antennomere 10. Antennomeres 3 to 7 appearing bicoloured, apical third of each antennomere distinctly darkened. Pronotum slightly longer than head. Elytra slightly longer than, and about as wide as pronotum; punctation with punctures larger and more densely positioned than those on pronotum. Aedeagus with narrowly pointed median lobe and characteristically elongate, bifurcate sclerite of internal sac.

Distribution and bionomics
Ruptor cordatus sp. nov. was found inside the chambers of arboreal termite nests (Fig. 1) in the lowland forest of the North-Eastern Peru (Fig. 2). The termite host was confidently identified as a widespread Labiotermes labralis ( Figs. 1 and 2), the only species in the genus that builds an arboreal nest (Constantino et al. 2006). Based on our sampling, we were able to find Ruptor cordatus sp. nov. in roughly every fourth termite nest that was searched. All nests where Ruptor cordatus sp. nov. was found were with winged termites, where also a number of other inquiline staphylinids (mostly Aleocharinae) were always encountered.

Etymology
The scientific name of the new species is a Latin adjective that refers to the cordate shape of the head of the new species due to its characteristic notch in the middle of the hind margin of the head.

Phylogenetic analyses
PartitionFinder found the following four partitions: (1) 28S + COI2 + ArgK2 + Wg2 + TP2 + CADC2 + CADA2 + COI1 + ArgK1 + CADC1 + CADA1 + Wg1 + TP1; (2) Wg3 + TP3 + ArgK3; (3) CADA3 + CADC3; (4) COI3. For partition 1, SYM + I + G was found to be the best-supported model, for partitions 2 and 3-GTR + I + G, and for partition 4-HKY + G. We moved 28S to a separate partition as this is the only non-protein-coding gene in our dataset and analysed it under the SYM + I + G model. The third codon positions of COI were excluded as it has been suggested that they suffer saturation for deep divergences, which can potentially bias phylogenetic analyses (e.g. Swofford et al. 1996;Lin and Danforth 2004). Since COI3 was excluded, no partition was analysed under the HKY + G model. The BI analysis reached convergence, with a standard deviation of split frequencies well below 0.01 after 10 million generations. Mixing of the Markov chain Monte Carlo chains was good, effective sample size (ESS) values were greater than 200 for all parameters indicating good mixing of the chains and the observed PSRF was 1.00. Convergence was also visualised in Tracer v1.7.

Discussion
Our molecular phylogeny confirmed the placement of a new taxon as a new genus in the tribe Lathrobiini, which could be suspected from its morphology. It was resolved within the informal 'Medonina and allied taxa' clade, which can be additionally confirmed by the following morphological characters that Ruptor gen. nov. shares with all other members of this clade: small and acicular apical maxillary palpomere and the presence of apical ctenidium on the one side of metatibia only. The position of the new genus within this clade is rather isolated, which, given the currently available subtribal classification, would require erecting a new subtribe for placing Ruptor gen. nov. in the classification. However, as shown by the phylogeny here and earlier (Żyła et al. 2019, 2021), the current supra-generic classification of Paederinae needs a revision in order to align it with the phylogenetic data. To avoid the creation of a new family-group name while pending for such a revision, here the new genus is placed in Paederinae as Lathrobiini incertae sedis. Based on the distribution of its host and difficulty to collect even a few specimens of Ruptor that we have experienced, we presume it may have a broader distribution, potentially with more species to be discovered.
The derived morphology of Ruptor gen. nov., with its general compact appearance, highly reduced eyes, short, stout legs and short antennae suggest a degree of co-evolution with L. labralis. The fact that Ruptor gen. nov. was hitherto not found in any other situation typical of rove beetles (e.g. in leaf litter or in flight intercept traps) further supports its close association and some level of co-evolution with the termite host, in whose nests apparently it spends most of the time.
The overall tree topology in Fig. 6 was consistent with the previous analysis of Żyła et al. (2021), which served as a basis for the molecular-based phylogenetic analysis that we performed here. The most important differences were the position of the tribe Pinophilini which was resolved as sister to Lathrobiini + Paederini in Żyła et al. (2021), but as sister to Lathrobiini in our analysis, and the position of the Pseudolathra + Cylindroxystina clade resolved here as sister to Lathrobiina + 'Medonina and allied taxa', while it was recovered as sister to 'Medonina and allied taxa' clade in Żyła et al. (2021). These differences are most likely a result of the reduced taxon  sampling here and the lack of the morphological partition that was included in the previous study. While there were also some local differences in the tree topology within the 'Medonina and allied taxa' clade itself, consistently with other studies (Żyła et al. 2019, 2021Bogri et al. 2020), we recovered Medonina, the largest subtribe within this lineage, as non-monophyletic.