Mite fauna (Acari) associated with the poultry industry in different laying hen management systems in Southern Brazil: a species key

This study is based on results from a large project on the mite diversity in commercial laying hen systems in Brazil. It aims to identify potential biocontrol agents amongst mites present in poultry farms, with a focus on one feather mite pest Megninia ginglymura. A secondary goal is to provide an operational identification tool to help the development of biological control in poultry farms. Sampling was conducted from August 2013 to August 2014, with 43 sampling events, in three different laying hen management systems. We used traps, laying hen feathers, and abandoned wild bird nests to collect the mites. A total of 38,862 mites belonging to 23 families and 39 species were found, including 17 species exclusively found in traps, six in wild bird nests, and one in feathers. The most abundant and frequent ectoparasitic species was M. ginglymura (76.3%; observed in all management systems). Amongst other mites, Tydeidae showed the highest richness with five species (Brachytydeus argentinensis (Baker, 1970), Brachytydeus australensis (Baker, 1970), Brachytydeus obnoxia (Kuznetzov and Zapletina, 1972), Brachytydeus oregonensis (Baker, 1970), and Brachytydeus tuttlei (Baker, 1965)), followed by Cheyletidae with four species (Chelacheles bipanus Summers and Price, 1970, Cheyletus eruditus (Schrank, 1781), Cheyletus malaccensis (Oudemans, 1903), and Cheletomimus (Hemicheyletia) wellsi (Baker, 1949)). Potential predatory species for biological control were C. malaccensis (9.4%), Typhlodromus transvaalensis (Nesbitt, 1951) (0.8%), Blattisocius keegani (Fox, 1947) (0.7%), and Blattisocius dentriticus (Berlese, 1918) (0.4%). A dichotomous key is provided to identify 48 species dwelling in Brazilian layer farms.

The repeated use of acaricides at high concentrations to control infestations could lead firstly to the development of acaricide-resistant and more importantly to the accumulation of acaricides in chickens' organs and also in eggs. Various synthetic chemical pesticides, for example carbaryl and permethrin, can accumulate in different levels in organs/tissues, skin, fat and muscle (Marangi et al. 2012).
This study, which complements the study by , provides additional data aiming to identify the most promising biocontrol agent species, with a focus on the feather mite M. ginglymura, and to provide a key to species derived from poultry farms in the South of Brazil.

Materials and methods
This study was conducted in different commercial laying hen systems between August 2013 and August 2014, in Lajeado County, Vale do Taquari, state of Rio Grande do Sul, Brazil. Six poultry houses were sampled with their characteristics as reported in .
Two additional methods of sampling mites associated to poultry farms were proposed in this study as complementary to trap data by : laying hen feathersand wild bird nests. For feather mites evaluations, ten laying hens were evaluated for each laying hen house. Five feathers per laying hen were collected, totaling 50 feathers per sample in each laying hen house. The feathers were placed individually in a plastic container with 70% alcohol during a minimum of 24 hours before the screening. The plastic containers were taken to the laboratory in paper box with styrofoam inside. The screening was performed by filtering the alcohol in qualitative filter paper of diameter 12.5 cm and weight of 80 g/m 2 . This method primarily checks actual health and economic mite species. In addition, potential predatory mites captured in the feathers may indicate a possible predation relationship due to foraging for food.
The wild bird nests abandoned were found in the lateral roof side of the laying hens houses were also evaluated for the presence of mites. The species birds identified were Columbina picui (Temminck, 1813), C. talpacoti (Temminck, 1810) (Columbidae), Turdus sp. (Turdidae) and Zenaida auriculata (Des Murs, 1847) (Columbidae). The bird nests were stored in individual plastic bags and in laboratory, they were exposed in Berlese funnel for five to seven days and mites stored in plastic container with 70% alcohol. This sampling method can highlight a possible migration relationship between acarofauna of wild birds and confined laying hens as well as help identify potential predators for biological control (Lesna et al. 2009).
All mites were collected with a fine-tipped paintbrush and mounted with Hoyer's medium on microscope slides (Walter and Krantz 2009). The slides were kept for up to 10 days at 50-60°C to dry the medium, to extend the legs and for diaphanization of the specimens.

Identifications
The identification of specimens to the species level was done using a phase contrast light microscope and identifications keys. Voucher specimens were stored at the reference Collection of the Natural Sciences Museum of the UNIVATES University Center (ZAUMCN), Lajeado, Rio Grande do Sul, Brazil.
Seventeen species were exclusively found in traps and a single species exclusively found on feathers, Dermatophagoides farinae (Hughes, 1961) (Pyroglyphidae). In bird nests, six exclusively species were found and three of them belong to the genus Brachytydeus (B. argentinensis, B. australensis and B. obnoxia). Furthermore, A. casalis and Oribatid mites were collected from bird nests. Also in abandoned bird nests, the most representative species were C. malaccensis (30.1%), B. keegani (19%), and B. dentriticus (11.7%). Androlaelaps casalis and O. bursa were only recorded using this sampling method. In feathers, there was a high dominance of M. ginglymura (99.7%) and all other species occurred in low numbers. The most representative species in traps were C. malaccensis (35.1%), T. setosus (32.7%), and M. ginglymura (13.3%).
Mite species and the laying hen systems where they were collected from are listed in Appendix, along with sampling month and year and comments on their known habits and habitats. The number of recorded specimens is shown in parentheses.  (1) 15. Chelicerae rarely chelate, fixed digit often regressed and movable digit usually a hook, knife, needle or stylet-like structure; cheliceral bases sometimes fused medially; palpi simple or modified into a thumb-claw process, sometimes reduced; subcapitulum without rutella; ambulacral of at least legs II and III usually with 2 lateral claws and with or rarely without a median empodium that may be padlike or rayed and often armed with tenent hairs, or occasionally claw-or sucker-like; opisthosoma lacking paired lateral glands; opisthosomatic setal row c usually with 2 pairs of setae (cl-c2), rarely with 3 pairs or hypertrichous -Chelicerae typically chelate, usually dentate, rarely attenuate or stylet-like; cheliceral bases always separated; palpi simple, never with thumb-claw process; subcapitulum usually with rutella or pseudorutella; ambulacral of legs I-IV usually with 1 or 3 claws, bidactyl condition rare, empodium clawlike or sucker-like, never pad-like, rarely rayed; opisthosomata usually with pair of lateral glands (these are absent in more primitive taxa); opisthosomatic setal row c usually with 3-4 pairs of setae or hypertrichous; tracheal system absent or, when present, arising from bases of legs or as brachytracheae (relatively short tubes) on various parts of the legs or idiosoma; stigmata and peritremes never present between cheliceral bases or on prodorsum .  (Baker, 1949) 33. Peritremes and stigma absentSTIGMAEIDAE………Storchia pacificus (Summers, 1964) - 43. Genu I with solenidion " no more than three times longer than '; ventral apex of tarsi with proral (p, q) and unguinal (u, v)

Discussion
Commercial egg production is an important economic activity in Vale do Taquari and in the state of Rio Grande do Sul. In this context, improving the efficiency and sustainability of pest control practices is a major challenge. The development of biological control against pest mites might be of help and requires to know all associated mites, their frequency, damages caused by them ) and possible natural enemies. Among food habits of mites, the huge diversity can be emphasized: there are detritivorous, predatory , parasitic, hematophagous and omnivorous mites, not to mention feather mites (which are more detrivivous-microbivorous, but live as parasites on birds' body). This study provides the first dichotomous key that helps to identify the main species associated with poultry farms in Southern Brazil. It includes all food habits and might assist acarologists as well as farm technicians (agronomists, agricultural technicians, veterinarians, and zoo-technicians) to better understand pest problems and contribute to improve control. It includes species with health and economic importance such as feather mites, hematophagous mites and stored food mites as well as potential predatory mites. The biological control of mite pests by natural predators can help reduce high infestations, preventing ectoparasites from becoming a health and economic problem, and is essential to knowing which laying hens and wild birds are carriers of the pest mites into poultry houses . Megninia ginglymura stood out as a sanitary species associated with all environments and managements evaluated. Currently used synthetic acaricides have limitations, such as unfeasible eradication and presence of residues in eggs (Lesna et al. 2009). The biological control of M. ginglymura by the predators C. malaccencis (Granich et al. 2016) and B. dentriticus ) under laboratory conditions showed higher values of life-table parameters when C. malaccensis was evaluated as natural enemy of M. ginglymura (Ro = 135.6; T = 41.6; λ = 1.13; rm = 0.12). No M. ginglymura specimen was recorded in abandoned bird nests in this study, thus corroborating the findings of . This absence of M. ginglymura from any of the abandoned wild bird nests under test might be considered a very important piece of information in elucidating the relationship between wild birds in this region and mite fauna of laying hen houses. However, this information must be considered with caution as this mite species feeds on the body of the host, and nests should be evaluated soon after abandonment so that M. ginglymura could be captured. Ornithonyssus bursa was recorded only in abandoned bird nests and no specimens were recorded inside laying hen houses. However, several predators and other mites belonged to the same species as those collected in laying hen houses (feathers and traps) and abandoned bird nests, which leads us to conclude that there may be a transit of mite fauna in commercial poultry farms, corroborating the results obtained by .
Among the recorded mite fauna, eight species were captured with all three methods for sampling mites,: five of which were considered as having predatory habits (B. dendriticus, B. keegani, C. (Hemicheyletia) wellsi, C. malaccensis and T. transvaalensis), two were generalists (T. setosus and T. putrescentiae) and the presence of Tetranychus sp. in laying hen houses is yet to be elucidated due to the fact that this genus is largely known as a phytophagous genus (Bolland et al. 1998).
Predatory mite strategies to control feather mites are different from other economic pests as D. gallinae can remain outside the host and be captured by them over this period. For M. ginglymura to be effectively controlled by predatory mites, these must be able to have foraging activity directly on the bird-host body. Moreover, requirement is related to physical constraints: the predatory mite must survive to hight temperature of the body of the host. The fact that predatory mites were captured in feathers might suggest that they were able to live in the host and endure the average body temperature of laying hens (41.5 -42.5°C). The presence of several predatory mite species on feathers strongly suggests that they are not unable to forage on host, which makes biological control against M. ginglymura promising.
Predatory species associated with feathers were observed at very low relative densities ( Figure 1). It is worth noting that some of Cheyletids developed parasitic habits throughout evolution (some Cheyletiella spp.). This suggests that this group (which comprises mostly predatory mites) is intrinsically more adapted to on-bird life than other predatory mites such as soil/litter mites, and therefore, represents a promising candidate to control of M. ginglymura.
The predator Raphignathus sp. and the generalists B. oregonensis and T. granarius were concomitant in abandoned bird nests and traps. Brachytydeus oregonensis was misidentified as Lorryia sp. and observed in hen nests .
Inside hen houses, A. ovatus, M. muscaedomesticae, M. ginglymura, and R. nidorum were common (species present in feathers and traps). Most predatory species were associated to hen nests and traps; B. dentriticus, B. keegani, C. eruditus, C. malaccensis, Cheletomorpha lepidopterorum (Shaw, 1794) (Cheyletidae), and T. transvaalensis Faleiro et al. 2015). A total of 17 species were captured exclusively in traps, with the most recurrent species being B. tuttlei, G. destructor, and C. arcuatus. Species exclusively found in abandoned bird nests were A. casalis, B. argentinensis, B. australensis, B. obnoxia, O. bursa and the soilmite group Oribatida. Androlaelaps casalis is a generalist predator that is capable of feeding and reproducing on D. gallinae (Lesna et al. 2012). This provides a valuable overview of mite species present in poultry farm buildings, and thus a basis for subsequent studies dedicated to improve biocontrol in such contexts. The next step is choosing the natural enemies' candidates amongst present predatory mites, in order to initiate the development of a pest management method without pesticides to control ectoparasites. Preliminary studies at laboratory level indicate potential for biological control of M. ginglymura by natural enemies in poultry systems. New investigative stage should be the application of this bnowledge to the practical level in poultry farms.