Entomopathogenic Fungi and Bacteria in a Veterinary Perspective

Simple Summary Several fungal species are well suited to control arthropods, being able to cause epizootic infection among them and most of them infect their host by direct penetration through the arthropod’s tegument. Most of organisms are related to the biological control of crop pests, but, more recently, have been applied to combat some livestock ectoparasites. Among the entomopathogenic bacteria, Bacillus thuringiensis, innocuous for humans, animals, and plants and isolated from different environments, showed the most relevant activity against arthropods. Its entomopathogenic property is related to the production of highly biodegradable proteins. Entomopathogenic fungi and bacteria are usually employed against agricultural pests, and some studies have focused on their use to control animal arthropods. However, risks of infections in animals and humans are possible; thus, further studies about their activity are necessary. Abstract The present study aimed to review the papers dealing with the biological activity of fungi and bacteria against some mites and ticks of veterinary interest. In particular, the attention was turned to the research regarding acarid species, Dermanyssus gallinae and Psoroptes sp., which are the cause of severe threat in farm animals and, regarding ticks, also pets. Their impact on animal and human health has been stressed, examining the weaknesses and strengths of conventional treatments. Bacillus thuringiensis, Beauveria bassiana and Metarhizium anisopliae are the most widely employed agents. Their activities have been reviewed, considering the feasibility of an in-field application and the effectiveness of the administration alone or combined with conventional and alternative drugs is reported.


Introduction
Biological control has been defined as "the intentional introduction of an exotic biological agent for permanent establishment and long-term pest control" [1].
The present study aimed to review the papers dealing with the biological activity of fungi and bacteria against some mites and ticks of veterinary interest. In particular, attention was turned to the research regarding acarid species, Dermanyssus gallinae and Psoroptes sp., which cause severe threat in farm animals and, regarding ticks, also pets. Furthermore, some agents can involve human health, too.

Entomopathogenic Fungi
Several fungal species are well suited to control arthropods, being able to cause epizootic infection and most of them infect their host by direct penetration through the arthropod's tegument [2]. Most of the organisms are related to biological control of crop pests, but, more recently, have been applied to combat some livestock ectoparasites.
Spiroplasma sp. are bacteria responsible for sexual determination in insects. Tinsley and Majerus [46] demonstrated that Spiroplasma sp. are male-killing bacteria causing a female-biased offspring ratio in female ladybirds Anisosticta novemdecimpunctata. Although Spiroplasma sp. are usually considered to be pathogens, they have also been reported to be symbionts in some insects and the potential role of mosquito spiroplasmas as vector control agents has been discussed [47].
Bacteria of the genus Cardinium have been associated to the parthenogenesis of parasitoid wasps and recognized as a symbiont of the phytophagous mite Tetranychus pueraricola [41,48].
Schineria sp. bacteria have been previously isolated from the larvae of Wohlfahrtia magnifica (Diptera: Sarcophagidae), a myiasis-causing fly species for most domestic animals [49,50]. Toth and colleagues [50] suggested that Schineria has a strong chitinase activity and may contribute to the development of fly larvae and influence the metamorphosis of W. magnifica.
All species are highly fastidious intracellular pathogens and typically target the fat body and hemolymph cells of the host. The infective cells are typically small, dense rods ingested during feeding which traverse the midgut epithelium and enter the hemocoel, where they gain entry to host cells through endocytosis. Once within the cell, pleiomorphic forms develop within the cytoplasmic vacuoles, varying from bacteria-like secondary cells to large, round rickettsogenic stroma. As the disease develops, characteristic protein crystals form and cells revert to small rickettsia. Eventually, infected cells undergo lysis, releasing masses of rickettsia and crystals into the hemolymph.
Concern has been raised over potential inflammation and infection induced by entomopathogenic Rickettsiella in vertebrates [53], so care should be taken when working with these organisms [54].
Wolbachia pipientis was first detected in the common household species Culex pipiens by Hertig and Wolbach in 1924 [55]. This is the most widespread bacterial endosymbiont infecting terrestrial arthropods, mainly insects, but some arachnids, freshwater crustaceans, and filarial nematodes too [56]. It has been observed that Wolbachia may confer in infected Diptera species some resistance versus insect pathogens [57].
Different mosquito trials demonstrated that Wolbachia must be carefully assessed for use as a biological control agent; however, the effect depends on the mosquito species investigated, as well as on the Wolbachia strain [58].
This characteristic is a reproductive incompatibility between infected males and females that are either uninfected or infected with different strains of the endosymbiont [60,62]. The symbiotic bacteria spread in an arthropod population, causing a reduction in fitness through failed mating [63].
It has been supposed that the presence of Wolbachia may have direct consequences on the development of other pathogens in the same arthropod vector, and also indirect effects on the epidemiology of pathogens through impacts on the dynamics and genetic diversity of the vector [64,65].

Ticks
Ticks are large-bodied bloodsucking, nonpermanent parasitic Acari, feeding exclusively on vertebrates. They are divided into three families, among which Ixodidae (hard ticks) represent an important concern for mammalian health, although some of them also feed on birds, and can be carried between continents. The life cycle includes eggs, one larva and one nymphal instar, adult male and female. Life cycles are classified based on the number of times the stages change hosts. The ticks start to feed as larvae, then as nymphs, and finally as adults, even if, in some species, males do not ingest blood. Each generation may be 1 or 2 years, although some species may take 3 to 6 years [66].
Most Ixodid species change three hosts, and molts of juvenile stages occur on the ground. A few of them are referred to as one-host ticks, spending most of their life on a unique host and dropping to the ground for oviposition (i. e. Rhipicephalus microplus). Two-host ticks feed on the first host, molt in nymph and feed again. Engorged nymphs fall to ground and molt into adults that feed onto the second host, mate, then females drop and lay thousands of eggs, which are left among the decaying vegetation at protected sites, where a high relative humidity will ensure their survival [67]. Ticks can be more (R. microplus) or less (Ixodes ricinus) host specific, depending on their species [66].
There has been a shift of ticks to elevated latitudes and altitudes, would be due to climate change, along with the host abundance [68][69][70], so the area of distribution of some ixodid species has expanded in the last few years. Among the causes of tick introduction and spread, the uncontrolled movements of domestic or wild animals, climate trends, and changes in the use of land resources that allow hosts to increase have been reported. Ticks introduced into a region where there is no competition with other species of ticks, are, in fact, able to colonize the complete range of abiotic conditions compatible with their biology [71]. However, the effects from human activities appear more important in modifying biotopes, influencing the infection by pathogens, of ticks [67].
Ticks exert a direct damage, feeding on their host. Saliva and/or mouthpart penetration can induce a toxic reaction in hosts, such as tick paralysis [72,73], or allergic state in human patients [74]. Heavy tick infestation can cause severe anemia, considering that an adult female tick can feed up to 2.0 mL of blood from the vertebrate host [75]. Conversely, many tick species have a role in the transmission of several pathogens, zoonotic too, causing an indirect damage. Pathogens transmitted by ticks include the greater part of the agents of vector-borne diseases in temperate areas, with a public health impact mostly unquantified [76]. Tick-borne diseases affect about the 80% of the world's cattle population, with a strong economic impact, mostly in developing countries [71].
As obligate hematophagous ectoparasites, ticks can ingest a huge amount of blood (up to 100-fold their body weight) [77], so they can easily transmit bacteria such as Anaplasma, Ehrlichia, Borrelia and Coxiella. Piroplasms (Babesia and Theileria), Cytauxzoon and Hepatozoon spp. complete their life cycle in hard ticks. One hundred and sixty tick-borne viruses are known, among which are tick-borne encephalitis and Crimean Congo hemorrhagic fever [70].
The current conventional acaricide treatment consists in the administration of different chemicals. The most employed drug classes are synthetic pyrethroids, organophosphates, amitraz, fipronil, insect growth regulators, macrocyclic lactones and isoxazoline. The compounds can be administered systemically or by direct application on the coat, alone or in mixture, with differences among the countries. Moreover, in companion animals, isoxazoline is administered per os. Anyway, the massive administration of acaricide drugs has made a resistant tick population [78]. To the best of our knowledge, the first report of the acaricide (organochlorine) resistance of cattle tick dates back to the half of the past century [79]. Acaricide resistance increased after the development of other compounds and will take place after exposure to any new molecules.
R. sanguineus (brown dog tick), conversely, is a three-host tick, and dogs are the primary host species, although infection of different animal species can occur in certain areas [94]. It can infest homes and is able to complete its life cycle indoors [95]. Acaricide resistance largely occurs in this species, too [96][97][98][99].
The rise of acaricide resistance in ticks is a serious concern, whose real extent is unknown [94]. However, several studies dealing with the mechanisms of resistance have been accomplished in the last few years [100][101][102][103][104], indicating that such resistance is due to genetic changes of tick populations, whose mechanisms would be linked to modifications to the target site, metabolism of compound alterations, or a decrease in the ability of the drug to cross the outer protective layers of the tick's cuticle [105].

Fungi
Natural infections caused by Aspergillus flavus, A. fumigatus and P. insectivorum in all stages (mostly adults) and eggs of examined ticks have been described since 1964 [106]. Afterwards, 17 different fungal species were reported to occur in diseased I. ricinus, Dermacentor marginatus and D. reticulatus. Engorged females of I. ricinus appeared more prone to fungal infection in summer seasons [107]. The most pathogenic molds were A. parasiticus, B. bassiana, Beauveria tenella, V. lecanii and P. fumosoroseus. Similarly, B. bassiana, B. brogniarti, P. farinosus, P. fumosoroseus, V. lecanii and Verticillium aranearum was able to colonize I. ricinus free in the environment from Denmark [108], mostly being engorged females in autumn, suggesting a possible activity of EPFs as regulators of these populations. A. ochraceus, Curvularia lunata, and Rhizopus arrhizus were isolated from naturally infected R. sanguineus and were found able to kill them [109,110]. Recently, A.parasiticus, along with Penicillium steckii and Scopulariopsis brevicaulis, was found to contaminate a laboratory-reared colony of I. ricinus [111].
The control of ticks by entomopathogenic fungi has been widely studied and, differently from insects, tick eggs are sensitive [112]. Tick species differ in their behavior, range of hosts and life cycle, so also, their sensitivity in comparison to a fungal species is not the same [113]. Furthermore, ticks are reported to be more tolerant to EPFs than other arthropods, so the amount of the inocula for tick control purposes should be larger. Different stages of ticks would exhibit differences in sensitivity versus EPFs. R. sanguineus engorged females and unfed other stages appeared more prone to fungal infection with M. anisopliae and M. flavoviride [114]. Nymphs were reported as less sensitive when compared with other stages [113,115]. A slight difference of sensitivity to M. brunneum, between adults and nymphs of I. scapularis was also reported [116,117] and larvae are considered the most susceptible stage to EPFs [118].
Metarhizium and Beauveria, when cultured in a liquid medium, can produce yeast-like propagules, known as blastospores. These fungal stages have also been checked for their entomopathogenic action, being able to easily penetrate cuticles [119].
EPFs recognize their target host, then conidia adhere and germinate on its cuticle, developing hyphae and appressoria. Such fungal structures exert a mechanical pressure along with enzyme secretion, allowing the fungi to cross the cuticle, invade the host's whole body, causing the death and colonizing the cadaver with their mycelium, emerging from the cuticle to continue the vegetative cycle.
EPFs, to start the host invasion, must overcome the epicuticle (outer layer of cuticle), mostly composed by esterified lipids, different among the hosts' species [120,121]. The ability to colonize more host species based on the recognition of such lipids makes the fungi specialist or generalist. Furthermore, the adherence of conidia seems to be mediated by several proteins and lipolytic enzymes [122]. Lipolytic enzymes secreted by EPFs also seem to cause alterations in the lipid balance of ticks, hampering their survival and decreasing their reproductive capacity [123,124].
The pattern of invasion of M. anisopliae in ticks is characterized by a simultaneous internal and cuticular fungal growth, unlike insect colonization by the same mold. This can be also due to the different composition of the cuticle between insects and ticks. The latter have a lower proportion of chitin and differences in the binding of proteins in female alloscutum (to allow a rapid expansion during engorgement), so this different composition makes ticks more prone to fungal attacks. This rapid and extensive cuticle degradations would hasten the tick death, following the water loss [125]. Furthermore, a serious cytotoxic impact of M. robertsii on R. microplus hemocytes has been recently reported [126,127].
M. anisopliae is the most studied fungal species. Its most prominent features on R. microplus have been recently reviewed [128]. Striking differences of entomopathogenic activity among the isolates have been reported [129], probably due to genetic differences [128]. Furthermore, a trial on cattle did not yield the same results as in-field studies [130]. The mold acted as active, along with B. bassiana [131] as well as a blastospore suspension, together with B. bassiana and M. robertsii [132]. R. microplus was sensitive to commercial conidial suspensions in vegetable oils [133] and both in vegetal and mineral oils, applied on the grass, killing 100% of ticks and lasting in the environment up to 60 days [134]. A similar good efficacy against R. microplus was reported for a formulation containing both microsclerotia and blastospores of M. robertsii [135].
M. anisopliae appeared active against Rhipicephalus variegatus, R. sanguineus and Ixodes scapularis, but led to a limited mortality in Dermacentor variabilis, which was sensitive to B. bassiana [116]. It was proven to be effective against Amblyomma parvum [136] and Haemaphysalis qinghaiensis, such as B. bassiana [137]. In a trial on Dermacentor albipictus larvae, a spray application of M. anisopliae was more effective and active in a shorter time, such as M. brunneum, when compared to B. bassiana [138].
M. anisopliae appeared more active than B. bassiana against I. ricinus and to a lesser extent against D. reticulatus [139]. Strains of M. anisopliae and B. bassiana were able to kill engorged females of Hyalomma anatolicum, yielding better results in comparison with Paecilomyces lilacinus [140], while strains of M. anisopliae and P. lilacinus were more pathogenic than B. bassiana against R. microplus [129]. Similarly, M. anisopliae was reported to be very pathogenic against Haemaphysalis longicornis, unlike B. bassiana [141], while this latter fungal species was reported to be very effective against the same tick in another study [142]. These remarks would indicate huge differences among fungal populations and the usefulness of testing fungi isolated from the same environment where selected ticks occur. M. brunneum was successfully administrated to I. scapularis in a granular formulation [117,143]. This fungal species showed marked differences in activity against Rhipicephalus annulatus under field conditions [144].
B. bassiana was assayed on R. microplus in vitro and in vivo on affected cattle, with very promising results [145,146], as well as on Hyalomma lusitanicum both in vitro and on field, applied inside wild rabbit burrows [147,148]. The mold was active against unfed adults of Amblyomma americanum [149], as well as against amitraz-resistant and amitraz-sensitive Rhipicephalus decoloratus, without detecting any significant difference [150].
M. anisopliae was reported to inhibit 92% Rhipicephalus appendiculatus, while B. bassiana inhibited 80% [151]. Furthermore, a synergistic effect of both EPFs on Rhipicephalus appendiculatus and Amblyomma variegatum [152,153] was referred, too. In a recent comparative study on the efficacy of the autodispersion of commercially available strains of M. anisopliae and B. bassiana against nymphal stages of R. sanguineus, the latter fungal species was faster in killing ticks and more efficient in sporulating on cadavers, allowing a mold dissemination [154].
However, strong differences in sensitivity to M. anisopliae and B. bassiana among different populations of R. microplus are reported [155][156][157][158][159][160]. For these reasons, preliminary in vitro assays of different fungal strains against the selected tick population are mandatory, also considering the sublethal effects on reproduction, then on future population size in the environment [161].
Isaria fumosorosea, Isaria farinosa and Purpureocillium lilacinum (formerly included within the genus Paecylomyces) are considered as important EPFs and scored active against R. microplus, acting also in reducing the egg production from treated adult females. I. fumosorosea was able to reduce the hatching percentage of treated eggs but was scarcely effective on larvae [162]. I. fumosorosea induced a low mortality of larvae in R. sanguineus [114], in D. reticulatus and I. ricinus [136]. For these reasons, its use as mycoacaricide against ticks is not recommended [139].
S. brevicaulis was recovered from D. variabilis [163] and, although this genus is considered as a minor EPF [164,165], it was assessed as capable of protecting this tick species from the desiccation induced by M. anisopliae [166]. It is currently considered as a pathobiont, transstadially transmitted in winter ticks, able to kill experimentally infected ticks, but without significantly affecting eggs and larvae [167].
The activity of EPFs can be affected by environmental conditions (high temperature, desiccation, strong solar radiation) [113]. For these reasons, the formulation is essential. Oily formulations are effective in protecting conidia from solar UUVV irradiation and from the loss of humidity [128], and calcium alginate beads with granular corn starch or chitin powder as nutrients were able to protect M. pemphigi blastospores encapsulated from drying [168].
To avoid a waste of inoculum, small areas are most suitable for testing. The results of pen trials yielded variable results [161]. Liquid or solid formulations can be applied with fertilizer spreaders [169] and the mycoacaricide can be applied on pastures by aerial spraying [170]. EPFs can be directly applied on the host, but they must be able to overcome several barriers such as skin temperature, pH, sebum and sweat [161].
Lastly, the effectiveness of an integrated control has been reported. The mycoacaricide activity was enhanced, when fungal stages were added with deltamethrin to control pyrethroid-resistant R. microplus [171,172] or with cypermethrin and chlopyriphos [173]. Similarly, I. scapularis was sensitive to fipronil, added to M. anisopliae [174,175], as well as R. sanguineus to M. anisopliae plus cypermethrin [176]. An integrated alternative control of R. microplus by both essential oils and entomopathogenic fungi (M. anisopliae and B. bassiana) indicated the suitability of Pelargonium graveolens essential oil that is not able to inhibit these EPFs and, when B. bassiana is not involved, also Lavandula hybrida [177]. Similarly, an integrated control of I. scapularis, was achieved with a synthetic pyrethroid acaricide, M. anisopliae strain F52 and a mixture of essential oils [178].
Other alternative methods of tick control are the use of entomopathogenic nematodes alone [179] or in oily emulsions [180], plant-derived compounds [181][182][183] or vaccines. Anti-tick vaccine Bm86 is commercially available [184], mostly active on R. annulatus and to a lesser extent on R. microplus [185]. EPFs active against different tick species are summarized in Table 1.

Bacteria
Some bacterial species have been demonstrated to be pathogenic for ticks; thus, they are considered useful for biological control. Among EPBs, B. thuringiensis is the most studied agent with activity against ticks [195] and is largely employed in commercial insecticide formulations. The pathogenic action of B. thuringiensis normally occurs after the ingestion of spores by ticks, and the crystalline inclusions containing insecticidal δ-endotoxins specifically interact with receptors in the insect midgut epithelial cells [196].
Studies about the effectiveness of B. thuringiensis against ticks showed that this property is strongly related to tick species and different tick developmental stages [195,197,198]. In vitro investigations reported the activity of B. thuringiensis against Hyalomma dromedarii, Argas (persicargas) persicus, and R. microplus, I. scapularis, I. ricinus, D. reticulatus [199].
Szcsepanska and coworkers [199] tested four environmental strains of B. thuringiensis and one commercially available product (Vectobac) containing B. thuringienis against ticks of the species I. ricinus and D. reticulatus. Vectobac was not active against both tick species, whereas two environmental B. thuringiensis strains proved to be efficient against I. ricinus and D. reticulatus, with the mortality rate for ticks assessed as being up to 80%. Moreover, D. reticulatus males were the most sensitive ticks to bacteria. The authors found similarity between the most and least efficient B. thuringiensis strains in enzymatic profiles (lipases, phosphatases, proteases, and chitinases), and for this reason they supposed that the detected enzymes have a limited role in the pathogenicity profile of the bacterial strains against ticks.
The effectiveness of B. thuringiensis against other tick species was also observed in further investigations. The pathogenicity of B. thuringiensis variety kurstaki was tested against the black-legged tick I. scapularis. B. thuringiensis was active against engorged larvae with LC50 of 107 spores/mL [198].
The sensitivity of the soft ticks A. persicus and the hard ticks H. dromedarii versus the commercial product Vectobac was assayed by Hassanain et al. [197] and mortality rates over 70% were observed. These results are not fully in accordance with those found by Habeeb and El-Hag [200] that did not record mortality rates after dipping H. dromedarii in Vectobac, but when they injected this commercial insecticide to H. dromedarii hemocoel, increased mortality rates in the next 48 h were observed.
In Mexico, four strains of B. thuringiensis, among 60 tested native strains, caused mortality rates exceeding 90% of adult R. microplus on the 20th day of the immersion test assay [195].
To the best of our knowledge, studies about the sensitivity of R. sanguineus to B. thuringiensis are not present in the literature. However, Renè-Martellet et al. [201] studied the microbiota composition of R. sanguineus ticks collected in different geographical areas (Senegal, France, Arizona) and found that each bacterial microbiota was dominated by three genera: Coxiella, Rickettsia and Bacillus. In particular, Rickettsia and Coxiella were the two main genera detected in females, whereas males had a higher proportion of Bacillus; however, the nature of the association between male R. sanguineus ticks and Bacillus spp. was not characterized.
The tick pathogenic property of Proteus mirabilis has been observed by Brown et al. [202] in a laboratory population of D. andersoni. In fact, a high rate of mortality was observed among all developmental stages of engorged ticks from which P. mirabilis was cultured. Mortality was preceded by disease in ticks, that had discoloration caused by the release of black decayed blood into the hemocoel, when the gut decomposed. At death, the cuticle was badly decomposed and was easily ruptured. Furthermore, the viscous fluid in the body cavity had a characteristic putrefactive odor.
Even though these findings suggest a relevant activity of P. mirabilis as entomopathogen, its use against ticks is not recommended because this is an opportunistic bacterium able to cause infections in humans and animals [203].
Among endosymbiont bacteria, W. pipientis is the species most frequently found in ticks. It has been detected in a range of tick species of the genera Ixodes, Rhipicephalus, Hyalomma, Amblyomma, Haemaphysalis [204].
The presence of Wolbachia in R. sanguineus ticks that mainly parasitize dogs has raised a concern as to whether the endosymbionts within the ticks can be transmitted into these animals. Currently, Wolbachia is being utilized as a method for vector control in Aedes mosquitoes [205,206]. Previous studies on sera of human participants exposed to multiple bites of Wolbachia-infected Aedes mosquitoes, showed Wolbachia-free residues indicating no transmission to humans [207]. On the other hand, an investigation conducted on blood collected from dogs in Haiti found approximately 22% of dogs PCR-positive for Wolbachia [208]. Furthermore, Wolbachia was detected in blood specimens of dogs and was determined from the filarial nematode Dirofilaria repens [209]. Wolbachia has also been detected in the blood of cats and it is supposed to be related to the heartworm Dirofilaria immitis, which harbors this endosymbiont [210]. Based on these findings, further investigations are required to assess the possibility of Wolbachia being transmitted to mammals, including humans, through the feeding of ticks.

Dermanyssus Gallinae
The genus Dermanyssus comprises hematophagous mite species, parasites of birds. The taxonomy of species within the genus was not clearly defined, until now [211]. Dermanyssus gallinae (poultry red mite) is very common in layer houses and is considered as the most damaging to laying hens worldwide [212]. The mite belongs to order Mesostigmata and mainly live at all stages in the environment, in cracks or crevices near the hosts' resting sites, feeding intermittently (every 2-4 days) for short periods (up to 1 h) on the birds during the dark hours [213][214][215]. Its life cycle (from egg to adult) is completed in one to two weeks and takes place through eggs, larvae, two nymphal stages and adults. Especially females of adult and nymphal stages exert hematophagy. This short life cycle, the wide range of optimum temperatures (10-35 • C) and high relative humidity (>70%), usually occurring in egg-laying facilities, contribute to make the mite a pest [216]. Parasite densities can, in fact, reach 50.000-500.000 mites per bird in caged systems [217]. A relationship between the occurrence of mites and hen mortality has been recorded [218].
Although birds are first choice hosts, D. gallinae feed on humans and other mammals, too [219], and can act as a vector for several pathogens of poultry [220], as well as zoonotic agents [221]. D. gallinae, in fact, feeds on humans too, showing opportunistic feeding habits, in respect to other species within the genus. Immunocompromised people seem to be prone to mites' attacks and to following pathogens transmission, occurring in more than a third of cases [222].
Poultry red mites spend most of their life in environmental refugia and can survive up to 9 months without feeding [216], so the control should be performed in the environment. The control of D. gallinae has been made up by using silicas (in dusts and liquid formu-lations), exerting a physical action [223,224], yielding satisfying results, when associated with the mechanical cleaning of henhouses.
Chemical acaricides in the environment have been widely applied [215,218,225]. However, a widespread resistance to such molecules has been recorded in the last few decades [218,[226][227][228][229][230]; moreover, many acaricides have been withdrawn from the European market [218]. These drugs, in fact, would have a public health impact, occurring as residues in eggs and meat [231,232].
Alternative methods of control have been revised by many authors, encompassing the use of vaccines, pheromones, botanical extracts, natural enemies, acaripathogenic fungi and bacteria, as well as identifying different biological targets for new chemicals [214,215,218,225,233].

Fungi
Entomopathogenic fungi have been assayed to control the mite population. B. bassiana, M. anisopliae, Trichoderma album, and P. fumosoroseus are the most studied fungal species. The use of fungal entomopathogens to control arthropod pests as biological agents would be suggested considering their easy direct penetration through arthropod tegument, the lack of induction of host resistance, the ability to horizontally transmit from fungus-infected to uninfected arthropods, mostly in moist environments [234] and potential damage to flies, lice, and other pests [235,236]. Among the different stages, nymphs show a lower sensitivity to EPFs [237].
B. bassiana, P. fumosoroseus and M. anisopliae were proven to kill several red mites, when administered in high doses, with a variability depending on the isolate [237][238][239][240], being able to cause high mortality within 5 days [238]. The efficacy of B. bassiana appeared enhanced, when administered in mixture with T. album. These fungi killed up to 80% of treated mites within 10 days [241]. In a more integrated approach, B. bassiana showed a synergistic interaction with desiccant dusts (up to 38% higher), maintaining the effectiveness up to 4 weeks [242], with a marked repellent effect [243], and with some essential oils [244]. Problems with the administration of conidia have recently been overcome using corrugated cardboard, infected with high doses of B. bassiana spores, acting as an autoinoculation device [245].
Different strains of M. anisopliae have been successfully applied to control the mites, under laboratory conditions, demonstrating differences in pathogenicity with a dose-and timedependent effect [246]. A spray of conidia in sunflower oil applied on field in a poultry farm demonstrated that the amount of conidia should be greater than in laboratory, being difficult to maintain temperature and humidity under control [247], as observed for M. brunneum [248]. High temperatures together with low relative humidity negatively affect the efficacy of EPFs, as demonstrated in ticks colonized by M. anisopliae [121]. Anyway, M. anisopliae was able to reduce the mite population after a week which lasted up 3 weeks [247].
Recently, a native isolate of Aspergillus oryzae, previously cultured by a dead D. gallinae, was checked for its activity against poultry red mite, showing a lethal activity by the sixth day after the administration of conidial suspension [249].
The main drawbacks for the use of EPFs in the control of D. gallinae are related to rapid mite regrowth, the time necessary to allow fungi to grow and low persistence in the environment [250,251], along with the stability of selected strains [252]. For these reasons, genetically modified fungi should be selected.
In conclusion, these entomopathogenic microorganisms seem to show an interesting anti-mite effect against D. gallinae. Anyway, considering the complexity of the epidemiology of poultry infection, a multidisciplinary approach would be very advisable [253].

Bacteria
The use of B. thuringiensis has been proposed as an alternative control method to chemical acaricides against D. gallinae in integrated management programs. It has been observed that B. thuringiensis var. kurstaki is able to damage the cuticle of D. galllinae and cause the loss of mobility of this mite in a period of 24 h [252]. Moreover, Torres and Hernandez [254] observed a moderate mortality of D. gallinae from day 2 of application (66%), which increased up to 78% at 7 days at a concentration of 35 mg/mL. Similarly, a previous study by Mullens et al. [255] on the fowl mite Ornithonyssus sylviarum, revealed that this mite was susceptible to B. thuringiensis, and the authors concluded that the entomopathogen had potential for the development of a control preparation for direct application to poultry.
Microbiota present in D. gallinae mites has been studied and four categories of bacteria have been identified: saprophytes, opportunistic pathogens, strict pathogens, and endosymbionts. The last ones are intracellular obligate bacteria able to cause phenotypic and reproductive alterations in their arthropod hosts; they belong to genera Spiroplasma, Cardinium, Schineria, Rickettsiella [39,45].
Studies about the presence of Wolbachia sp. in D. gallinae did not find these bacteria, that are frequently present in other arthropods in which they cause reproductive anomalies [39].
Even though endosymbiotic bacteria living inside D. gallinae were found, the effect of these infections on the poultry red mite is not known.
Some These findings demonstrated that different endosymbionts may be present in D. gallinae and the authors supposed that endosymbionts could cause biological modifications to the poultry red mite [43], similarly to what has been observed in other hosts [63].
Based on these observations, it seems that biological control using endosymbiotic bacteria-derived substances that may induce changes to the reproduction of arthropods may be a viable alternative to traditional methods of control of the poultry red mite [45].

Psoroptes sp.
Psoroptes mites are non-burrowing Acharina, responsible for ear and body mange of herbivores. Psoroptes ovis severely impacts on animal health. It induces an exudative dermatitis in beef cattle and sheep which, when not treated, can lead affected animals to lose condition and, sometimes, to death [256,257]. The parasite of rabbits Psoroptes cuniculi (syn P. ovis var. cuniculi) [258], considered conspecific with P. ovis [259], primarily lives on the inner surface of the pinna [260] and is responsible for otoacariasis.
The life cycle of P. ovis (egg, larva, two nymphal stages and adults) completely occurs on the host. All parasite stages can pierce the surfaces and feed on tissue fluids. Moreover, the host's skin produces serous exudate because of a delayed hypersensitivity response induced by allergens from mites' fecal pellets [256,261].
Psoroptic mange in ovine (sheep scab) and in rabbit hosts (ear cancer) are the most frequently reported clinical forms characterized by severe pruritus. The consequent itching usually distracts the animals from eating, leading to weight loss, and, in sheep, fleece deterioration and reduction in milk and meat production. In rabbit, otherwise, the disease mainly presents as erythema, extreme pruritus, and crusted lesions in the external ear. Sometimes, the mites can spread to other parts of the host's body, causing generalized scabs in the head, neck, ventral abdomen, and urogenital area [260]. The infections are highly contagious and quickly spread among the animals. Psoroptes can survive and maintain their infectivity after 15 days off the host, while mite eggs were able to hatch for up to 7 days in the same conditions [262].
The control of affection and conventional treatment rely on topical organophosphates or injectable formulations of macrocyclic lactones. However, scab mites have developed resistance to conventional acaricide drugs. Psoroptes mites have quickly developed resistance to all the synthetic pyrethroids and to propetamphos, without side-resistance to diazinon [263], that nowadays would protect sheep against mites for 8-10 weeks [264]. Injectable macrocyclic lactone formulations are more operator friendly, but, recently, a multiple resistance has been reported [265,266] and the selective pressure of long acting moxidectine would enhance helminthic resistance, too. The use of remedies alternative to ivermectin are welcome, because of several negative aspects relative to this drug. First, ivermectin can induce drug resistance, with the consequent loss of its effectiveness. Moreover, it can be neurotoxic and induce central nervous system depression in treated animals. Subcutaneous treatment is painful for animals and in rabbits it is unsuitable because of their natural behavior; in fact, rabbits usually lick each other in the ears as part of their hygiene, and this aspect may affect the pharmacokinetics of ivermectin, and prolonged treatment could cause intoxication [267,268]. Furthermore, it was demonstrated that the repeated administration of ivermectin subcutaneously in male rabbits causes a decrease in the weight of the sexual organs, which is a negative consequence in the animal production [269]. Finally, the use of ivermectin can represent a threat for people, too, because of the residues in rabbits' meat for human consumption [270]. Furthermore, acaricide drugs can pass in the environment and in the food chain, occurring as toxic residues in milk and meat, and are banned in organic farms [264].
Alternative biological control can be achieved by using entomopathogenic microorganisms, such as bacteria and molds, or by administering natural compounds [271].

Fungi
Astigmata mites are soft-bodied and have an unsclerotized tegument. This feature would facilitate fungal colonization [272]. Moreover, in diseased animals, the microenvironment of lesions acts as a favorable microclimate for fungal growth. With Psoroptes not being a burrowing mite, the parasites live in groups, in strict contact, that allows the direct transmission of mycelia [273].
The rate of parasite killing and thermotolerance are of capital importance to allow the molds to carry out their entomopathogenic activity. These features depend on the selected fungal isolate [274]; B. bassiana was reported to show the optimal growth temperature between 25 • C and 28 • C [275], even if some isolates can grow at about 30 • C, with highly reduced activity and may not survive at 34 • C [276].
The first in vitro study on the effects of B. bassiana on Psoroptes recovered from rabbits [273] stressed a strong lethal activity on both infected adults and on the life span of larvae hatched from infected eggs. Then, an in vitro and in vivo study was performed, demonstrating that Psoroptes mites can become infected by entomopathogenic fungi on the skin of sheep, also. These findings showed the feasibility of a direct application of fungal conidia onto the sheep body [277].
In a comparative in vitro study with Hirsutella thompsonii, M. anisopliae was highly pathogenic and suitable for the control of P. ovis [272]. These features were furtherly corroborated by observing the efficiency of the mold in producing fatal infections, as well as the infectiveness of 5-day-old cadavers of mites [278]. M. anisopliae shows a higher thermotolerance, when compared with B. bassiana, with an optimum 30 • C, but growing at 37.5 • C, also, while no infections were observed at 40 • C [279], confirming the statement "B. bassiana and M. anisopliae are known to have their optimum of growth at 25 and 30 • C, respectively" [280].
Anyway, although the thermotolerance and virulence of EPFs would depend on the strain, M. anisopliae is able to grow up to 35 • C, sharing this feature with P. farinosus [274].
The higher infectivity of M. anisopliae in comparison with B. bassiana was assessed in vivo, too [277]. The strong parasite killing of M. anisopliae seems to be related to its ability to induce the oxidative damage of mites [281].
Scopulariopsis brevicaulis (teleomorph Microascus brevicaulis) is a soilborne ascomycete, occasionally involved as an EPF. The genus has been revised by Sandoval-Denis et al. [282]. The mold showed a dose-dependent pathogenicity for P.cuniculi in an in vitro study, being able to colonize the mites, leading them to death. The infected parasites appeared debilitated, lost mobility and quickly died [164]. However, in the same paper, the occurrence and the following cultivation of Scopulariopsis sp. from bodies of healthy mites from ear crusts was recorded. This finding would suggest an opportunistic role of this mold versus the tested parasite species. EBFs active versus D. gallinae, as well as Psoroptes sp. are reported in Table 2.

Bacteria
The in vitro acaricidal effect of B. thuringiensis on P. cuniculi has been demonstrated. The bacterium can induce histological alterations of this mite, such as the presence of dilated intercellular spaces in the basal membrane, membrane detachment of the peritrophic matrix and morphological alterations in columnar cells of the intestine [283].
The use of mixtures of B. thuringiensis with other acaricidal compounds has been proposed. For example, it has been proven that a combination of chitinase and soybean trypsin protease inhibitor effectively suppresses population growth in the flour mite Acarus siro [284], and many B. thuringiensis strains have chitinolytic activities [285] that could enhance the efficacy in mite control. Lee and coworkers [286] observed that the combination with other natural products such as naphthoquinones induces a decrease in the induction of long-term resistance, with short-term efficacy, and at a low cost.
Similarly, the combined use of B. thuringiensis and ivermectin has been proposed by some authors to combat Psoroptes sp., in view of a potential synergistic or additive effect with the possibility of lowering the dose of ivermectin [283].
Besides B. thuringiensis, that directly acts against Psoroptes sp., other bacteria may be involved in the survival of mites. Some studies have been carried out to verify the role of some bacterial strains isolated from mites. Serratia marcescens is a Gram-negative bacterium of the family Enterobacteriaceae responsible for infections, including septicemia, in several animal species. It has been proven that this bacterium is pathogenic to several insects, too, including flies and mosquitos with different mechanisms of action [287].
S. marcescens has been frequently cultured from Psoroptes sp. mites, but it is not clear if the bacterium acts as endosymbiont or has anti-mite activity. Perrucci and coworkers [288] observed that P. cuniculi does not need S. marcescens to live and infect healthy rabbits. However, the authors found that only rabbits infested with S. marcescens-free P. cuniculi mites presented crusts in their ears, whereas mites and/or eggs were only detected in the ear cerumen of all rabbits infested with S. marcescens-infected mites. Table 3 summarizes EPBs active against Psoroptes sp, D. gallinae and some tick species.

Varroa destructor
Varroa destructor is a parasite Mesostigmata mite, exerting a huge impact on beekeeping. It has become a global parasite, switching host onto Apis mellifera from Apis cerana. Varroasis is often a threat for colonies, when nearby colonies collapse [289]. Without a treatment, an infected colony dies within 2 years post infection [290]. The life cycle consists of a phoretic phase when adult hosts carry mites within and between colonies and a reproductive phase, when Varroa lays eggs inside the bee brood cells [291]. Mature female daughters are produced in worker brood cells and daughters in drone cells and remain immobile until prepupae are present. Then they climb on them and create a feeding site by puncturing the host's cuticle and feed on the larval fat body, parthenogenic mites develop, then oviposition starts. When bees emerge from the cells, they have mites feeding on them [292]. The affected hosts show weight loss, with a deficit in reproductive fitness. The mites would prefer nurse bees [293], and modify bees' behavior, are able to mimic a host's cuticular hydrocarbons to escape the hygienic behavior of the host [294] and quickly shift to acaricide resistance. Varroa can transmit deformed wing virus and acute bee paralysis virus. Noel et al. [291] have recently reviewed the main control options. Chemical control is based on conventional miticide products acting on Varroa on adult bees or, when administered in strips, acting on mites emerging from the brood cells. However, these drugs leave residues in hive products, and resistance to acaricide is increasing. Organic acids or terpenes such as thymol are used in organic control, but they would decrease worker population, increasing capping brood removal or decreasing sperm quality. Lithium chloride appears as a selective inhibitor of Varroa acetylcholinesterase, such as the use of predators (with interesting laboratory results, but not on colony). Promising is the use of RNA interference to knock down specific genes of Varroa, although still experimental. V. destructor has been reported to be susceptible to the entomopathogenic fungi, M. anisopliae, B. bassiana, Verticillium lecanii, Hirsutella spp. [3,295], Hirsutella thompsonii [296,297], B. bassiana [298] and M. anisopliae [299,300]. Clonostachys rosea (formerly Gliocladium roseum) is an Ascomycete, belonging to Hypocreales, widely distributed in soil, and provided by an endophytic ability in tissues from several plants. The mold produces conidia and chlamydospores. Colonies on potato dextrose agar are greyish white when grown in the dark, while appearing yellow to orange under lighter conditions [301]. C. rosea was able to kill 60% of mites, in comparison with B. bassiana and M. anisopliae, which caused the death of 90% of mites. All the acaripathogenic molds were reported to be able to control V. destructor by preventing the gene suppression of bee immunity, induced by the mite, too [302]. However, the main shortcoming for the use of acaripathogenic fungi in beekeeping is due to the potential pathogenicity of these fungi for insects, too.
Alquisira-Ramirez et al. [303] observed that B. thuringiensis could be an effective alternative to control V. destructor, because the bacterium is virulent to the mite but does not cause mortality in bees. In fact, no toxic effects of the proteins of B. thuringiensis have been demonstrated for the larvae and adults of A. mellifera, maybe because the pH of the bee intestine is usually acidic, whereas B. thuringiensis toxins are activated at alkaline pH values [304].

Zoonotic Potential of EPFs
A potential zoonotic activity of EPFs has been reported. M. robertsii, M. guizhouense, M. brunneum and M. pingshaense (specie complex M. anisopliae) were referred as the species involved with human infection, mostly keratitis [305], although two cases of keratitis due to M. anisopliae have been reported in soft contact lens wearers [306]. However, considering a last further recent taxonomic study, comprising the description on new species [20], within the known complexes, it is very hard to state the lack of zoonotic ability of Metarhizium species, used as mycoacaricide. On the other hand, B. bassiana was identified as responsible for mycotic keratitis in a patient involved in occasional agriculture work [307] and, interestingly, in the same study, 14 clinical cases of B. bassiana keratitis were revised. Five out of the twelve patients with anamnestic data were working in agriculture.
Bacillus thuringiensis has been associated to different human infections; it has been cultured from marginal and apical periodontitis, wounds, corneal ulcera and gastrointestinal infections in humans [308]. Even though this EPB is not considered as a traditional zoonotic agent, its presence in different forms of human infections suggests that, at least in immunocompromised patients, it could represent a risk. B. thuringiensis, similarly to B. cereus, produces several virulence factors potentially acting against mammalian cells, such as hemolysins and enterotoxins [309].
For these reasons, the use of B. thuringiensis in pest control should be carried out with attention to avoid possible infections, mainly in operators.
Bacillus cereus has been involved in human periodontitis, too [308], as well as in other human infections, mainly of the gastrointestinal tract [310]. B. cereus is thus a well-known pathogen for humans and animals and for this reason, its use is not recommended in pest control.

Conclusions
The control of ectoparasites requests the development of novel strategies and, among them, the use of entomopathogenic microorganisms appears as a promising tool to achieve an ecofriendly approach. Several studies have been accomplished, both in crops' defense and in sanitary entomology, mostly in fighting mosquitoes. The present study has revised the literature dealing with the application of these organisms to manage some veterinary parasitosis, caused by Acari.
Most of data refer to ticks' control, showing the feasibility of the environmental application of this strategy, as well as important differences in the sensitivity of ticks and pathogenicity of EPFs, that make a preliminary laboratory assay mandatory.
Entomopathogenic microorganisms appear as important for their environmental sustainability, for the lack of resistance induction in parasites and, in general, for their safety towards hosts, proving the ability to break the life cycle of both these pests and of several vector-borne agents, zoonotic, also, in a One Health perspective.
These tools appear promising in an integrate approach, too, and their administration with conventional acaricide drugs or, in a green approach, with different plant extracts is advisable. Finally, the management of D. gallinae should be considered as an ideal candidate for an in-field application of this strategy, considering the withdrawal of several conventional acaricide from the market. Further study and in-field research are needed to improve a large-scale application, considering the possible impact on non-target species, too.