New MacrochelePratums species (Acari, Mesostigmata, Macrochelidae) associated with burying beetles (Silphidae, Nicrophorus) in North America

Abstract Burying beetles (Silphidae, Nicrophorus) are hosts to a broad diversity of mites (Acari), including several species of Macrocheles Latreille, 1829 (Mesostigmata, Macrochelidae). The macrochelid fauna associated with silphids primarily in North America was surveyed; in total, 1659 macrochelids representing seven species were collected from 112 Nicrophorus beetles representing nine host species. Three new species of Macrocheles were discovered during the survey and described as Macrocheles willowae sp. n., M. pratum sp. n., and M. kaiju sp. n. The barcode region of cytochrome oxidase subunit I (COI) was amplified from the three new described species, as well as M. nataliae and M. praedafimetorum, and analysed in a small phylogeny.


Introduction
Carrion-feeding beetles (Silphidae) are associated with a diverse assemblage of mites, nematodes, and fungi. Nicrophorus (Silphidae) species are large-bodied beetles, that breed and feed on decaying organic matter, most often vertebrate carcasses (Anderson and Peck 1985). There are at least 60 extant species of Nicrophorus worldwide, 22 of which occur in the New World (Sikes et al. 2008, Sikes et al. 2016. Nicrophorus beetles are unique amongst insects because most species provide biparental care and they bury small vertebrate carcasses in subterranean crypts (see Anderson and Peck (1985) for a summary of their life cycle). Nicrophorus beetles are associated with a broad diversity of mites that can occur at high prevalences and abundances, with at least 14 species of mites representing four families collected off 95% of beetles in a given population (Wilson and Knollenberg 1987). The symbiotic relationship between silphids and their associated mites are poorly understood; however, the relationship may be a blend of commensalism and mutualism, as some mite species actively prey on eggs of carrionfeeding flies that compete with Nicrophorus (Wilson and Knollenberg 1987).
The Macrochelidae (Mesostigmata) are a cosmopolitan family of predaceous mites with at least 480 described species from 20 genera, occurring in a wide variety of organic substrates where they feed on nematodes and other microinvertebrates (Krantz 1998, Lindquist et al. 2009, Emberson 2010. There are about 320 described species of Macrocheles Latreille, 1829 (Macrochelidae) worldwide (Emberson 2010), many of which are phoretic as adult females on insects, including nine species which are associated with silphids (M. agilis, M. glaber, M. kurosai, M. lisae, M. merdarius, M. muscaedomesticae, M. nataliae, M. praedafimetorum, M. vespillo) (Halliday 2000, Mašán 2003, Niogret et al. 2007, Perotti and Braig 2009. Macrocheles associated with silphids attach with their chelicerae to beetles dispersing to and from carcasses, and they generally feed on nematodes, insect eggs and larvae, and other invertebrates on carrion (Wilson andKnollenberg 1987, Schwarz et al. 1998). Macrochelids phoretic on burying beetles are often overlooked and unstudied, resulting in a scarcity of information about their life history and novel species that remain to be described. A recent survey of tortoise mites (Uropodina, Uroobovella) on Nicrophorus beetles (Knee et al. 2012) also uncovered three new species of Macrocheles associated with burying beetles. Herein, I propose and describe Macrocheles willowae sp. n., M. pratum sp. n., and M. kaiju sp. n., include a small phylogeny based on the barcode region of COI, and describe the diversity, abundance and host range of Macrocheles species found on Nicrophorus throughout this survey.

Biological collections
Silphids were collected by various researchers across eight countries and 21 provinces or states (see acknowledgments). In Canada, most silphids were collected as bycatch from xylophagous beetle trapping by W.K. Specimens from other countries were collected primarily in pitfall traps, and others were hand-collected. Beetle specimens preserved in ethanol were shipped to Carleton University, and upon receipt specimens were placed in 95% ethanol and stored at -20°C. Using a dissecting microscope, silphids were identified to species using keys from Anderson and Peck (1985). The presence, abundance, and attachment location of mesostigmatic mites was recorded. All mesostigmatic mites were removed and placed in a 0.5 ml microfuge tube with 95% ethanol and stored at -20°C for later identification and/or molecular analysis. Mites were slide-mounted in polyvinyl alcohol medium (6371A, BioQuip Products, Rancho Dominguez, California, United States of America (USA)) and cured on a slide warmer at 40°C for 3-4 days. Slide-mounted specimens were examined using a compound microscope (Leica DM 2500) with differential interference contrast illumination (DIC), and identified to species using the primary literature. Initial drawings of mites were made with pencil on paper using a camera lucida. Illustrations were later merged in Adobe Photoshop CS5 and redrawn in Adobe Illustrator CS5 using an Intuos 3 Graphics Tablet from WACOM Co., Ltd. (Saitama, Japan). Leg chaetotaxy is based on the system proposed by Evans (1963) and Evans and Till (1965). Idiosomal chaetotaxy follows the system of Lindquist and Evans (1965) as applied to macrochelids by Hyatt and Emberson (1988). Notation for glandular openings and poroids (proprioreceptors or lyrifissures) follows the system developed by Athias-Henriot (1969, 1971, 1975 and Johnston and Moraza (1991), as reviewed by Kazemi et al. (2014). Measurements were made from at least eight female specimens, all measurements are in micrometres (µm), and lengths presented with mean followed by the range in parenthesis. Type specimens are deposited in the Canadian National Collection of Insects, Arachnids, and Nematodes (CNC), at Agriculture and Agri-Food Canada, Ottawa, Ontario, Canada, and the Smithsonian Institution National Museum of Natural History.

Molecular methods
Genomic DNA was extracted from whole specimens for 24 hours using a DNeasy Tissue kit (Qiagen, Inc., Santa Clara, California, USA). Following extraction, mites were removed from the extraction buffer, vouchers were-slide mounted, and genomic DNA was purified following the DNeasy Tissue kit protocol. PCR amplifications were performed in a total volume of 25 µl, with 14.7 µl ddH2O, 2.5 µl 10× ExTaq buffer, 0.65 µl 25 mM MgCl2, 1.0 µl of each 10 µM primer, 2.0 µl 10 mM dNTPs, 0.15 µl ExTaq DNA polymerase, and 3 µl genomic DNA template. Primer pairs LCO1490 + HCO2198 (Folmer et al. 1994) were used to amplify a 689 bp fragment of the 5'end of COI. PCR amplification was performed on an Eppendorf ep Gradient S Mastercycler (Eppendorf AG, Hamburg, Germany), using the following protocol: initial denaturation cycle at 94 °C for 3 min, followed by 45 cycles of 94 °C for 45 s, primer annealing at 45 °C for 45 s, 72 °C for 1 min, and a final extension at 72 °C for 5 min. Amplified products and negative controls were visualized on 1% agarose electrophoresis gels, and purified using pre-cast E-Gel CloneWell 0.8% SYBR Safe agarose gels (Invitrogen, Carlsbad, California, USA). Sequencing reactions followed the protocol of Knee et al. (2012), and sequencing was performed at the Agriculture and Agri-Food Canada, Eastern Cereal and Oilseed Research Centre Core Sequencing Facility (Ottawa, Ontario, Canada). Sequence chromatograms were edited and contiguous sequences were assembled using Sequencher v5.3 (Gene Codes Corp., Ann Arbor, Michigan, USA). COI sequences were aligned manually in Mesquite v3.10 (Maddison and Maddison, 2016) according to the translated amino acid sequence. COI sequences from Macrocheles subbadius (MBIOE1677-13, MBIOE1699-13) generated by the Barcode of Life Data Systems (BOLD) were included in the phylogeny. COI sequences on GenBank from two Macronyssidae (Mesostigmata) species, Ornithonyssus bacoti and O. sylviarum (FM179677, KR103486), were used as outgroup sequences. Sequences generated during this study have been submitted to GenBank (Table 1). Pairwise distances were calculated using neighbour-joining (NJ) analyses with the Kimura-2-parameter (K2P) model in PAUP* v4.0b10 (Swofford 2003). Phylogenetic reconstructions of the COI dataset was performed using Bayesian inference (BI) in MrBayes v3.2.6 (Huelsenbeck and Ronquist 2001;Ronquist and Huelsenbeck 2003). Each specimen in the phylogeny is labeled with the mite species and the beetle number, followed by the host species and abbreviated state, province or country (Fig. 14).
MrModeltest v2.3 (Nylander 2004) was used to determine the best-fit model of molecular evolution for each molecular marker, which was determined to be GTR+I+G. Bayesian analysis was performed in MrBayes using the Markov Chain Monte Carlo (MCMC) method, two independent runs, with nucmodel = 4by4, N st = 6, rates = invgamma, samplefreq = 1000, four chains = one cold and three heated. The COI dataset ran for 10 million generations, producing 19502 trees after a burn-in of 250 trees. The remaining trees in Mesquite, excluding the burn-in, were used to generate a majority-rule consensus tree displaying the posterior probability supports for each node. Bayesian analyses were performed using the on-line Computational Biology Service Unit at Cornell University, and at the Cyberinfrastructure for Phylogenetic Research (CIPRES) portal (Miller et al. 2010).  Hyatt and Emberson 1988). All dorsal and ventral setae smooth and spinose, except J5 barbed and slightly shorter than Z5. Seta j1 simple with rounded tip, j1 slightly longer than z1. Dorsal hexagonal setae (j5, z5, j6) nearly as long as marginal and submarginal setae (R and UR). Dorsal shield with moderate reticulations throughout, except smooth in dorsal hexagonal area and between j4 setae, without well-defined procurved line, sigillary rami absent. Sternal shield more than twice as wide as long, punctures small, posterior margin concave. Well defined linea media transversa (l.m.t.) and linea oblique anteriores (l.o.a.), l.o.a. contacts l.m.t. Linea arcuata (l.arc.) well defined and contacts l.o.a. Linea angulata (l.ang.) and linea oblique posteriore (l.o.p.) well defined laterally but faint medially. Area punctata laterale (a.p.l.) well defined, but area punctata posteriore (a.p.p.) not well defined. Ventrianal shield longer than wide (ratio 1.3). Arthrodial brush as long as movable digit. Genu IV with six setae. Femur IV setae ad2, pd1 prominent spikes with flattened forked tip.
Macrocheles nemerdarius was described from the nest of a mouse, Peromyscus in Maryland and the nest of the eastern woodrat Neotoma floridana in Florida, USA, and this species is also phoretic on coprophilous beetles (Krantz and Whitaker 1988). Female M. willowae sp. n. differs from that of M. nemerdarius in having marginal or submarginal setae slightly longer than dorsal hexagonal setae, posterior margin of sternal shield more concave, pon seta smooth not weakly pilose, j1 only slightly longer than z1 not 1.5 times as long, and J5 slightly shorter than Z5 not half as long as Z5. Comparisons were made using the species description for M. nemerdarius and examination of the holotype specimen loaned from the National Museum of Natural History, Smithsonian Institution.
Female M. willowae sp. n. differs from that of M. pratum sp. n. in having marginal and submarginal setae slightly longer than dorsal hexagonal setae. Genu and tibia IV with slight ridge on anterolateral and posterolateral surfaces in M. willowae sp. n., while M. pratum sp. n. only has a ridge on the posterolateral surface. Seta J5 is slightly shorter than Z5 and more spinose in M. willowae sp. n., J5 is less than half as long as Z5 in M. pratum sp. n. Punctures on the sternal shield are smaller and less prominent in M. willowae sp. n. than in M. pratum sp. n. The ventrianal shield is longer than wide for both species, but the shield is slightly narrower in M. pratum sp. n., ratio of 1.4 compared to 1.3 for M. willowae sp. n.
In Krantz and Whitaker (1988), Dr. W. Yoder provided a short diagnosis and partial illustrations of the female and male of an undescribed and unnamed Macrocheles species collected from Nicrophorus beetles and three mammal species (Tamiasciurus hudsonius, American red squirrel in Michigan; Tamias striatus, eastern chipmunk in Maryland; and Zapus hudsonius, meadow jumping mouse in Prince Edward Island). This undescribed species was a common associate of Nicrophorus beetles, but it was also found frequently enough on live rodents to suggest an association with small mammals (Krantz and Whitaker 1988). Dr. W. Yoder reportedly intended to formally describe and illustrate this new species of Macrocheles; however, to date this species has not been described. Macrocheles willowae sp. n. is likely the same species that Dr. W. Yoder was intending to describe. Over several years, repeated attempts were made to contact Dr. W. Yoder about the status of the description, but contact was unsuccessful.  N. hybridus, N. marginatus, N. obscurus, and N. pustulatus (Table 1).

Macrocheles pratum
Diagnosis female. All dorsal and ventral setae smooth and spinose, except J5 barbed and much shorter than Z5. Seta j1 simple with rounded tip, j1 slightly longer than z1. Dorsal hexagonal setae slightly longer than marginal and submarginal setae. Dorsal shield with moderate reticulations throughout, except smooth in dorsal hexagonal area and between j4 setae, without well-defined procurved line, sigillary rami absent. Sternal shield more than twice as wide as long, punctures moderate size, posterior margin concave. Well defined l.m.t. and l.o.a.; l.o.a. contacts l.m.t. Well defined l.arc. contacts l.o.a., l.ang. and l.o.p. well defined laterally but faint medially. Well defined a.p.l., but a.p.p. not well defined. Ventrianal shield longer than wide (ratio 1.4). Arthrodial brush as long as movable digit. Genu IV with six setae. Femur IV setae ad2, pd1 prominent spikes with flattened forked tip.
Male and immatures. Unknown. Etymology. Pratum (Latin neuter noun) means "meadow". This species was only collected in Kearney County, Nebraska and Onefour, Alberta, which are in the prairies.
Remarks Female M. pratum sp. n. differs from that of M. nemerdarius in having larger more prominent punctures on the sternal shield, the posterior margin of the sternal shield is more concave, pon seta is smooth not weakly pilose, j1 is only slightly longer than z1 not 1.5 times as long, J5 is shorter and broader for M. pratum sp. n. (9) than for M. nemerdarius (13), and the ventrianal shield is narrower, length to width ratio of 1.4 for M. pratum sp. n. and 1.2 for M. nemerdarius. Measurements were made examining M. nemerdarius holotype specimen loaned from the National Museum of Natural History, Smithsonian Institution.

Male and immatures. Unknown.
Etymology. Kaiju, 怪獣, from Japanese means strange beast, and refers to giant monsters such as Godzilla or Mothra. Female M. kaiju sp. n. is relatively unique morphologically when compared to other Macrocheles species associated with beetles, it is relatively large, has a unique dorsal shield shape, and bears numerous setae with distinct forms.
Remarks. Female M. kaiju sp. n. is different from that of any other described Macrocheles species; however, it does fit the Macrocheles generic description (see Hy-  att and Emberson 1988). Female M. kaiju sp. n. has two character states that are irregular for Macrocheles but somewhat similar to Holostaspella (Macrochelidae) species: the anterior margin of the dorsal shield wraps around onto the ventral surface such that j1 is on a slight projection on the venter; and a slight ridge is present on femur II. Holostaspella species are characterised in part by females having a spur on femur II, and j1 being on a tuberculate anterior extension of the dorsal shield but not wrapped around onto the venter. Macrocheles kaiju sp. n. differs from that of Holostaspella species in having weak ornamentation on the dorsal shield, its lateral regions without a series of depressions; j1 is smooth and not pectinate, j1 is on a slight projection and not a prominent tuberculate extension; sternal shield weakly ornamented and without strong median ridge; and metasternal shields small and always free of endopodal shields.
The majority rule consensus tree from the BI analysis of COI was well supported, with all nodes having high posterior probabilities, and eight nodes with 100% support (Fig. 14). Macrocheles willowae sp. n. was divided into two well supported clades, one with mites from N. defodiens and the other with mites from N. orbicollis and N. carolinus. Macrocheles willowae sp. n. did not appear to diverge based on geographic location (Fig. 14). Macrocheles willowae sp. n. mites from N. defodiens and those from N. orbicollis and N. carolinus were morphologically indistinguishable, despite the higher than average intraspecific divergence between these two well supported clades. The phylogenetic relationships between Macrocheles species, and the genetic structure of these newly described species, requires further analysis with better taxon sampling, specimens from more host species and localities, and additional molecular markers.

Distribution and biology
Seven species of macrochelids were collected from 112 Nicrophorus beetles representing nine species from three countries (Canada, USA, Germany) and 10 provinces or states (Table 1). Nicrophorus carolinus was associated with the most Macrocheles species (3), three host species had only two macrochelid species, and five had only one macrochelid species. Mites were usually found under the elytra, either clasping onto the integument near the prospiracle (68%) or on the ventral surface of the elytra (22%), and sometimes they were on the coxae (10%). Mites attached to the outer surface of the beetle could have been dislodged into the preservative.
A  low abundances (32 mites total on 10 beetles) (Table 1). Macrocheles pratum sp. n. was collected from five host species and had the greatest host range of all species collected. Macrocheles willowae sp. n. collected from three host species had the second broadest host range. Macrocheles nataliae, M. praedafimetorum, and M. kaiju sp. n. were each collected from a single host species.
The species with the greatest geographic range was M. willowae sp. n., collected from 22 sites, across seven provinces/states in Canada and USA. Macrocheles pratum sp. n. was collected from a single site in Alberta (Canada) and from another site in Nebraska (USA). Macrocheles kaiju sp. n. was collected from one site in Florida and another site in Nebraska, USA. The four other macrochelid species collected were each found in a single locale (Table 1).
COI sequences generated in this study were compared against those on GenBank and BOLD, and M. willowae sp. n. was the only species to have high level (100%) matches on BOLD. These matching sequences belonged to generic level identified specimens from Alberta, Ontario, Saskatchewan, Nova Scotia, and Florida. The species briefly diagnosed by Dr. W. Yoder was collected from three species of rodents in Maryland, Michigan and Prince Edward Island. Combined together, results from this study, BOLD and Dr. W. Yoder's findings, the geographic distribution of M. willowae sp. n. may cover 11 provinces or states in Canada and USA.