miRNA Regulatory Functions in Farm Animal Diseases, and Biomarker Potentials for Effective Therapies

MicroRNAs (miRNAs) are small endogenous RNAs that regulate gene expression post-transcriptionally by targeting either the 3′ untranslated or coding regions of genes. They have been reported to play key roles in a wide range of biological processes. The recent remarkable developments of transcriptomics technologies, especially next-generation sequencing technologies and advanced bioinformatics tools, allow more in-depth exploration of messenger RNAs (mRNAs) and non-coding RNAs (ncRNAs), including miRNAs. These technologies have offered great opportunities for a deeper exploration of miRNA involvement in farm animal diseases, as well as livestock productivity and welfare. In this review, we provide an overview of the current knowledge of miRNA roles in major farm animal diseases with a particular focus on diseases of economic importance. In addition, we discuss the steps and future perspectives of using miRNAs as biomarkers and molecular therapy for livestock disease management as well as the challenges and opportunities for understanding the regulatory mechanisms of miRNAs related to disease pathogenesis.


Introduction
MicroRNAs (miRNAs), defined as short non-coding RNA (ncRNA) molecules of about 22 nucleotides in length, regulate a variety of biological processes through the posttranscriptional regulation of gene expression. They control the expression of protein-coding genes and participate in the regulation of many cellular processes in animals. miRNAs regulate gene expression by inhibiting translation initiation or elongation, and inducing co-translational protein degradation and premature termination of translation [1,2]. In addition to this, they are also known to form miRNA-mRNA and miRNA-lncRNA pairs to influence gene regulation and biological activities [1].
From the identification of the first miRNA (lin-4) in 1993 [3], advances in nextgeneration (NGS) and third-generation sequencing (TGS) technologies in the last decade have heralded a new era and ability to identify various classes of small RNA molecules, including miRNA, in different biological samples at unprecedented speed [4]. The advantages offered by multiple sequencing platforms (e.g., Illumina, Ion Torrent and SOLiD) and bioinformatics data management capabilities support in-depth miRNA sequencing Livestock diseases are responsible for huge economic losses to the livestock industry and cause important issues of animal welfare [14,15]. Moreover, many livestock diseases can be transmitted to humans with the potential to cause health issues and even death. The effective control of livestock diseases is a global challenge for the livestock industry requiring multiple layers of control and intervention [16][17][18]. Many livestock diseases, such as mastitis, paratuberculosis and bovine viral diarrhea (BVD) in cattle, porcine reproductive and respiratory syndrome (PRRS) and African swine fever (ASF) in pigs, and Newcastle disease and Avian influenza in poultry, require multidisciplinary or holistic approaches for effective management and control. Vaccination, therapeutic treatments, and eradication strategies are traditional and routine methods to combat diseases, while modern methods, such as genome editing and RNA interference (RNAi), can lead to the development of alternative strategies for combating disease outbreaks. However, instead of combating diseases, farmers can select animals based on the genetic resistance of health traits, which has been regarded as a sustainable method [14,15]. Together with genetic markers, targeting epigenetic markers and miRNAs have been regarded as a further strategy for combating livestock diseases [19][20][21][22][23].
The potential roles of miRNAs in farm animal diseases have been summarized in several reviews [24][25][26][27][28][29]. These reviews, however, only provided an overview of the changes in miRNA expression profiles during disease progression. In this review, we present an in-depth and up to date review of miRNA roles in the main farm animal diseases and argue for the potential use of miRNAs as biomarkers for animal disease management.

miRNA Biomarker Development and Potential Therapeutic Application in Livestock Production
The possibility and the practical aspects of using miRNAs as biomarkers have been intensively reviewed in many human diseases, such as cancers [30][31][32], rheumatic diseases [33], diabetes mellitus [34,35], and infectious diseases [36]. In farm animals, discovering biomarkers will be crucial for the management of disease [37].
A biomarker is a measurable indicator of a certain biological state, such as health and disease. Initial biomarkers used in human disease management were protein biomarkers but detecting new and enhanced protein biomarkers has proven to be an expensive and time consuming venture, mainly due to the low availability of clinically relevant proteins, their complex nature and the lack of accurate and repeatable detection methods [32]. Tay-lor [38] recently summarized the common qualifying factors of a biomarker: (1) easily accessible (i.e., discovered and measured easily and using minimal invasive procedures); (2) specific to the condition under investigation (specificity); (3) high sensitivity (easily and accurately detected, ideally before the appearance of clinical symptoms, and potentially vary according to disease stages or response to therapy); and (4) translatable from research/development to application. As compared to other nucleic acids, miRNAs are ideal biomarker candidates, as they are very stable under a wide range of conditions and can be extracted from a variety of liquid biospecimens (e.g., blood, urine, milk, and feces) and tissue samples. miRNAs are also highly specific to tissues and cell types, and their ability to delineate disease stages has been successfully used to differentiate cancer stages and to monitor responsiveness to therapy [39,40]. Specifically, a diagnostic role for miRNA with the ability to identify a disease or a prognostic role with the prospect of developing a specific disease phenotype has been demonstrated through meta-analyses of multiple cancer studies [41,42]. Moreover, miRNA signatures in bone marrow or blood are able to distinguish between Pediatric Acute Lymphoblastic Leukemia subtypes [43]. Moreover, miRNA biomarkers could enhance the specificity of metabolic or protein-based tests. For example, Ali Ahmed et al. showed recently that a combination of matrix metalloprotease protein-2 (MMP2) assay with miR-29a and miR-335 expression profiles demonstrated superior diagnostic detection of breast cancer compared to widely used tests, such as carcinoembryonic antigen (CEA) and cancer antigen 15-3 (CA 15-3) [44]. In farm animals, miRNA biomarker potentials for a range of diseases are reviewed below. miRNA can be used as biomarkers, capable of serving diagnostic, prognostic, or therapeutic purposes, for the management of livestock diseases. The development of a biomarker generally requires three major steps. The first step includes (i) discovery: to identify the potential candidate markers for specific conditions; (ii) confirmation and validation: to test the identified markers in other populations or other samples; (iii) adaptation: development of suitable arrays or tests for the identified markers ( Figure 1A). To date, most miRNA studies in animal diseases have focused on profiling miRNA changes (discovery stage). Although the functions of some miRNAs have been validated as summarized in the sections below, those experiments were mostly in vitro and may not reflect the actual underlying complex biological regulatory mechanisms. To the best of our knowledge, there are no available commercial miRNA biomarkers for use in livestock disease management.
Second, many factors must be considered, including specificity, expression sensitivity, validation in a large cohort to ensure its effectiveness, consistency of results, and impact on both cost (cost-effective) and time (long-term) ( Figure 1B). Therefore, to qualify as an miRNA biomarker, expression specificity and consistency for a specific condition must be established in the initial population, and repeatability in other populations confirmed. Moreover, the long-term cost implications must be favorable to enable adoption by end users.
Third, a validated miRNA biomarker should find application in any of the following: diagnosis, prognosis, therapy, and for predicting the response to therapy ( Figure 1C). For instance, different groups of miRNA biomarkers can be used for the management of Johne's disease (JD) in livestock as follows: (i) for early diagnosis of JD infection, (ii) for classifying JD due to different strains of Mycobacterium avium subsp. paratuberculosis (MAP), and (iii) for predicting the stage of JD and (iv) as an effective treatment (therapeutic).
The potential application of miRNAs as biomarkers in livestock improvement is further strengthened by a plethora of recent reports, summarized in section three below, that have reported differential response patterns of miRNAs to different livestock diseases. The development of miRNA biomarkers for livestock diseases, however, requires collective action and different stakeholders' involvement. Researchers, veterinarians and farmers have the prerogative to identify the issues and the need for biomarkers for specific purposes. The experimental design and data analysis require both the statisticians and bioinformaticians to derive the best analytical tools to deliver robust results in the discovery and validation phases. A crucial link is required between academia (researchers) and industry to develop cost-effective and reliable tests and to select the best option for developing biomarkers. Finally, validated and reliable miRNA-based diagnostic kits and biomarkers for breeding for resistance will support livestock improved productivity. Therefore, successfully developed miRNA biomarkers might serve as new tools that could enhance current methods or lead to the development of new methods or therapies for managing farm animal diseases.
miRNA-based therapeutic approaches for the management of farm animal diseases can be through miRNA inhibition (diminish the expression of disease-induced miRNAs) or miRNA replacement (re-establish the expression of disease repressed miRNAs) [45]. Several approaches have been used for miRNA inhibition, such as antisense anti-miR oligonucleotides, locked nucleic acid anti-miRs, antagomiRs, and miRNA sponges [46]. Similarly, several small molecules, synthetic miRNA mimics, and DNA plasmids have been used in miRNA restoration therapy [47,48]. The success of miRNA therapies majorly depends on suitable, effective, and specific delivery systems [49,50]. Some highly efficient and specific miRNA delivery methods have been via exosome and nanoparticles [51,52]. miRNA therapies to cure diseases in livestock could be an attractive option to complement or replace current disease-management methods. However, understanding the specific functions and mechanisms of interaction between miRNA and other biomolecules in disease pathogenesis is the first step in the development of miRNA therapeutics for livestock diseases.
The advantages of miRNA biomarkers for disease management have been summarized recently [44] and include (1) use in early detection of disease, which is important to improve prognosis and limit the spread of disease pathogens; (2) use to improve pathogen detection at the first sign of appearance of disease symptoms; (3) use to detect the latent phase of infection, which is crucial to limiting the spread of pathogens and loss of income through lower productivity when undetected (e.g., the early and subclinical phases of JD); and (4) use to detect and manage regional differences in disease pathogenicity. miRNA biomarkers are also important, as they can be used to develop effective therapies for livestock diseases.

miRNA Roles in Farm Animal Diseases
In this section, we present reported changes in miRNA expression profiles during disease pathogenesis in farm animal species, including cattle, pig, chicken, sheep, and goat. We present the most recent findings and also focus on the most important economic diseases of livestock.

Potential Regulatory Roles of miRNAs in Cattle Diseases
Following the first miRNA expression study on bovine adipose, mammary gland, immune-related, and embryonic tissues in 2007 [53], over 870 studies have characterized about 1064 precursors and 1025 mature miRNAs, encoded on all 30 chromosomes, in the Bos taurus genome (Table 1). These studies demonstrate crucial regulatory roles of miRNAs in many biological processes in bovine, including mammary gland development and lactation (reviewed in [29]), bovine immunity (reviewed in [54]) and diseases (reviewed in [24]), and embryo development (reviewed in [55]). This section provides an update on miRNA signatures and potential roles in the major bovine infectious diseases, including mastitis, paratuberculosis, or JD and BVD. Important miRNAs for these diseases are listed in Table 2. Mastitis, a common inflammatory disease of the mammary gland can develop into a clinical or subclinical type of infection depending on the causal pathogen [71]. Mastitis infection is caused by diverse pathogens including, but not limited to, Escherichia coli, Streptococcus uberis, Streptococcus dysgalactiae, Bacillus spp, and Staphylococcus aureus [72]. Dairy cow intramammary infections due to S. aureus have received much attention because of their major economic impact on dairy farms [73,74]. Therefore, more studies have investigated the role of miRNA functions in relation to this pathogen [56,57,60,62,[75][76][77][78][79][80][81]. E. coli generally causes clinical mastitis, while most cases of S. aureus mastitis are chronic in nature. Investigating the miRNA expression profiles due to these pathogens following challenge of MAC-T cells (bovine mammary epithelial cell line) with heat-inactivated S. aureus or E. coli bacteria at 0, 6, 12, 24, and 48 h, Jin et al. [60], reported a pathogen directed miRNA expression pattern whereby four differentially expressed (DE) miRNAs (miR-2339, miR-499, miR-23a, and miR-99b) were unique to S. aureus, while 5 (miR-184, miR-24-3p, miR-148, miR-486, and let-7a-5p) were unique to E. coli. Interestingly, the authors also observed a slower initial response of miRNAs to S. aureus bacteria (only one DE miRNA reported after 6 h of infection) compared to E. coli, which initiated an earlier miRNA response (six DE miRNAs reported after 6 h of infection) [60]. Another report on pathogen specificity identified 108 DE miRNAs as specific to E. coli, while 82 DE miRNAs were specific to cows infected with S. aureus [62]. Meanwhile, several important miRNA candidate biomarkers (e.g., mi-223, miR-1246, miR-142-5p, miR-23a, miR-31, miR-23b-3p, miR-26a, and miR-145) have been associated with S. aureus and/or E. coli mastitis in Chinese Holstein cows [76]. Moreover, Ma et al. [81] recently identified miR-378 and miR-185 as candidate biomarkers of milk infected with S. aureus. These studies and other lines of evidence indicate that miRNAs regulate the host response to S. aureus via different target genes and pathways. For example, miR-223 regulation of S. aureus resistance is via the PI3K/AKT/NF-κB pathway [80]; miR-145 modulation is through pathways related to immune cytokines [78]; and miR-15a regulation is through the inhibition of Interleukin-1 Receptor-Associated Kinase 2 (IRAK2) gene expression (Chen et al.) [79].
S. uberis is among the most prevalent mastitis-causing pathogens throughout Europe and North America [82]. Using real-time quantitative PCR to examine the expression of 14 miRNAs in bovine mammary epithelial cells (BMECs) challenged with S. uberis, Naeem et al. [57] indicated possible roles of miR-181a in intramammary infections via its regulatory function on Fc-gammaR-mediated phagocytosis, toll-like receptor signaling, and antigen processing and presentation pathways, while using a next generation sequencing approach, Lawless et al. identified 21 miRNAs as significantly DE post-infection as well as enrichment in pathways related to innate immunity [56]. Ngo et al. [58] profiled circulating miRNAs in cows' milk with naturally occurring mastitis due to different causative agents and identified 26 miRNAs as generic indicators of clinical mastitis, and suggested seven of them (miR-27b, miR-152, miR-194, miR-200b, miR-222, miR-379, and miR-18397) as early mastitis indicators. The authors identified 27 miRNAs unique to S. uberis mastitis with an emphasis on miR-320a and miR-320b due to their roles in the modulation of trained immune activity [58].
Compared to S. uberis and S. aureus, less attention has been given to miRNA changes during E. coli and S. agalactiae mastitis. Pu et al. [61] identified 35 DE miRNAs in mammary gland tissues from cows with S. agalactiae-type mastitis, with regulatory roles in several immune response and signal transduction pathways, such as the RIG-I-like receptor signaling pathway, cytosolic DNA sensing pathway, and Notch signal pathway. Analyzing miRNA expression profiles from peripheral blood, Li et al. [59] reported the involvement of several miRNAs (miR-25, miR30e-5p, miR-342, miR-191, miR-399b, miR-451, and miR-486) in biological processes involved in mastitis infection, such as the involvement of miR-25 in the development of the immune system through targeting of Krüppel-like factor 4 (KLF4) gene [59]. Meanwhile, Chen et al. [83] suggested miR-16 and miR-223 as new markers for dairy cow mastitis diagnosis. Interestingly, Lai et al. [63] identified five significantly upregulated miRNAs (miR-21, miR-146a, miR-155, miR-222, and miR-383) with the potential to effectively differentiate between California mastitis test-positive milk and non-infected milk. In another study, miR-144-5p and miR-130b-5p were significantly downregulated and upregulated, respectively, in mammary gland tissues infected with mastitis compared to healthy tissues [84].

miRNA and Johne's Disease
Mycobacterium avium subsp. paratuberculosis (MAP) is the causative agent of JD in cattle, sheep, goats, and other domestic and wild animals [85,86]. Johne's disease imposes a substantial economic burden on the dairy industry [85,87], and it is prevalent worldwide [88]. MAP and JD have attracted attention because of their speculated connection to human Crohn's disease [89]. Current JD control strategies are hampered by the lack of accurate and reliable diagnostic tests [90]. Meanwhile, several studies have indicated the potential of miRNAs to serve as diagnostic and prognostic tools of MAP. For example, Malvisi et al. [65] identified seven upregulated (miR-19b, miR-19b-2, miR-1271, miR-100, miR-301a, miR-32, and one novel miRNA) and two downregulated (miR-6517 and miR-7857) miRNAs in the blood of MAP-positive animals compared with unexposed animals. Studying the ileum, a target tissue for MAP infection, Liang et al. [66] reported 14 DE miRNAs as well as the potential role of miR-196b in the proliferation of endothelial cells, the role of miR-146b in bacteria recognition, and involvement of miR-146b in the regulation of the inflammatory response when comparing infected and control ileum tissues from calves. Recently, Gupta et al. [67] developed a prediction model using a combination of four miRNAs (miR-1976, miR-873-3p, miR-520f-3p, and miR-126-3p), which distinguished animals moderately and severely infected with JD from healthy animals. Ileum and ileal lymph nodes are important tissues during MAP infection, as they are the sites of MAP and host immune cell interaction, and where the host initiates immune responses to the pathogen. Wang et al. [91,92] reported the involvement of a different set of miRNAs in the regulation of MAP response in the ileum and ileal lymph as well as the possible involvement of miR-100, miR-330, and miR-2447 in the Th17 cell differentiation pathway during MAP infection in the ileal lymph node [91] or association of miR-370 and miR-383 with lipid metabolism in this tissue [92]. Moreover, Shaughnessy et al. demonstrated the utility of a number of miRNAs in bovine feces in differentiating healthy animals from those with late-stage JD, thereby providing potential biomarkers for MAP infection and disease progression [68].

miRNA and Other Cattle Diseases
miRNA expression is also changed during the progression of some other notable infectious diseases of bovine, including BVD, foot and mouth disease (FMD), and tuberculosis ( [69,93,94]). Like JD, the symptoms of BVD virus (BVDV) infections in cows are subclinical and difficult to detect. Only one study has investigated the miRNA profiles of colostrum infected with BVDV from five neonate Holstein calves inoculated with BVDV at different time points: before infection (day 0) and at 4, 9, and 16 days post-challenge and reported two DE miRNAs (miR-423-5p and miR-151-3p) between BVDV challenged and control groups across time points [69]. However, both miRNAs demonstrated inconsistent expression patterns, whereby miR-423-5p increased until day 4 post-challenge and decreased to the control level by day 16 post-BVDV exposure. At the same time, miR-151-3p remained similar to the control level until day 9 before increasing in BVDV challenged animals compared to control animals on day 16 [69]. Thus, more studies are required to identify miRNA roles in BVDV and miRNA biomarkers of BVDV. FMD is a highly contagious disease of domestic and wild cloven-hoofed animals [93], and its outbreaks incur enormous economic, political, and social ramifications. A recent study has suggested potential roles of miR-17-5p in acute FMD infection, miR-31 in FMD virus (FMDV) persistence, and miR-1281 in both acute and persistent infection [70]. Profiling miRNA expression in alveolar macrophages isolated from the lung of bovine infected with Mycobacterium bovis, the causative agent of bovine tuberculosis, Vegh et al. identified different sets of DE miRNAs, with crucial roles in the tightly controlled balance between pathogen survival strategies and the host immune response [64]. Among them, the authors validated IL-1 receptor-associated kinase 1 (IRAK1) and TGF-β receptor 2 (TGFBR2) as important target genes of miR-146. In addition, Want et al. [95] indicated that miRNA-199a plays important roles in M. bovis infection via the inhibition of cellular autography and downregulation of IFN-β expression, while Iannaccone et al. [94] identified miR-146a as a potential biomarker for the rapid diagnosis of M. bovis infection.

miRNA and Swine Influenza Infection
Influenza is a zoonotic viral disease that represents a health and economic threat to humans and animals worldwide [124]. In mammals, influenza viruses replicate mainly in the respiratory tract, usually accompanied by clinical signs, whereas in avian species, the major replication site is the intestinal tract without clinical symptoms [125]. Several studies with different approaches have investigated the implication of miRNAs in swine influenza infection [104,105]. Using computational procedures, initial research identified 36 pig miRNAs with putative target genes in swine influenza viral sequences isolated in a period of 38 years, which indicated that putative target genes and host miRNA (miR-124a, miR-136, and miR-145) interactions were maintained almost throughout virus evolution [110]. Based on miRNA expression profiling, several authors suggested that DE miRNAs regulate genes involved in innate immunity [111] or apoptosis and cell cycle regulation [120] in pigs infected with influenza virus. In another study, Brogaard et al. suggested five miRNAs (miR-15a, miR-18a, miR-21, miR-29b, and miR-590-3p) as potential modulators of viral pathogen recognition and apoptosis in the lung tissues of pigs on day 3 following challenge with influenza A virus H1N2 [121]. Moreover, Zhang et al. [126] observed that swine influenza virus H1N1/2009 infection could modulate the expression of host miRNAs (miR-204 and miR-4331) to facilitate its replication in the host.

miRNA and Other Pig Diseases
Salmonella species infect many vertebrate species, including pigs. Pigs colonized with Salmonella enterica serovar Typhimurium are usually asymptomatic, making their detection in carrier pigs difficult. Variable fecal shedding of Salmonella is an important cause of foodborne illnesses [127], and effective control of Salmonella-related infections is of public health importance. Studying the role of miRNAs in Salmonella disease pathogenesis, Huang et al. demonstrated the decreased expression of miR-155 in pigs with persistent Salmonella shedding (PS) [112], while Hoeke et al. [113] showed that miR-29a regulates intestinal epithelial cell proliferation by targeting caveolin-2 (CAV2). Li et al. analyzed the miRNA expression profiles in porcine intestines infected with Lawsonia intracellularis and found that associated DE miRNAs could target genes involved in pathways related to the immune response, amino acid metabolism, and cell communication/growth/motility [114]. Huang et al. [128] compared the whole blood miRNA transcriptomes of pigs (Duroc × Landrace × Yorkshire) at 2 days post-inoculation and before Salmonella infection and identified 29 DE miRNAs, including miR-146a-5p, miR-125a, and miR-129a-5p, with roles in Salmonella infection and immunology signaling pathways. Following validation by real time quantitative PCR, the authors concluded that miR-146a-5p in peripheral blood could significantly increase the fecal bacterial load [128].

miRNA and Marek's Disease Virus Infection
Marek's disease, a highly contagious viral neoplastic disease caused by infection of Gallid herpes virus 2 (GaHV-2) or Marek's disease virus (MDV), has remained a major concern in the poultry industry owing to the continual emergence of new virulent strains [157]. Several in vitro approaches have been employed to unravel the roles played by host-encoded miRNAs in different scenarios of MDV infection [133,[135][136][137][157][158][159][160][161][162]. Using white leghorn chicken experimentally inoculated with the oncogenic RB-1B strain as a model to investigate the connection between chicken miRNA response and the oncogenic nature of MDV, Stik et al. [158] reported the upregulation of miR-21 in chicken inoculated with RB-1B strain as compared to chicken vaccinated with a non-oncogenic strain CVI988. In a similar experiment, Li et al. [135] showed that miR-26a was downregulated in chicken spleens infected with MDV during different phases of tumor formation, while Han et al. [136] indicated that miR-103-3p was downregulated in tumor samples from the spleen and liver of infected chickens. Zhao et al. [137] observed that miR-219b promoted cell apoptosis via the regulation of the expression of genes in the apoptosis pathways during MDV infection. Moreover, Heidari et al. [138] indicated that DE miRNAs are involved in many reactome immune-related pathways, such as cytokine signaling, innate and adaptive immune systems, Toll-like receptors, and interleukin pathways.

miRNA and Other Chicken Diseases
Bursal disease, caused by infectious bursal disease virus (IBDV), is a highly contagious disease that predominantly affects the bursa of Fabricius in birds [164,165]. IBDV targets the host immune system by destroying B lymphocytes, attracting T cells, and activating macrophages [165]. In poultry, vaccination has contributed to the overall reduction of disease burden [166]; however, a comprehensive understanding of the complexity of virus and host interaction is limited [165]. The roles of miRNAs in the regulation of IBDV infection have been the focus of many investigations. Initially, Shen [167] reported that recombinant avian adeno-associated virus (AAAV)-delivered VP1-and VP2-specific miRNAs can inhibit the replication of IBDV efficiently in transducing 8-day-old specific pathogen-free chicken embryos. Two other studies reported the important roles of miR-9a and miR-2127 in IBDV infections. Ouyang et al. [147] found that miR-9 was induced 2, 4, 12, and 24 h after infec-tion with IBDV and that miR-9 can promote IBDV replication by repressing the production of type I IFN. In a subsequent experiment, Ouyang et al. [148] provided evidence that miR-2127 function in IBDV infection is via the downregulation of CHP53 mRNA translation and attenuation of CHP53-mediated antiviral innate immune response against IBDV. More recently, Fu et al. [149] reported that miR-130b suppresses IBDV replication via directly targeting the viral genome and cellular Suppressor Of Cytokine Signaling 5 (SOCS5) gene.
Avian influenza, caused by avian influenza viruses (AIVs), is an important disease for many bird species. Wild waterfowls or aquatic birds are the natural reservoir hosts of all influenza A subtypes [168], except for two novel IAV subtypes, H17N10 and H18N11, in bats [169]. The first study on miRNA gene expression in chickens infected with AIV using a deep sequencing approach was performed by Wang et al. [150], who reported 73 and 36 DE miRNAs between chicken lungs and tracheae infected and not with low pathogenic H5N3, respectively, 4 dpi. Some of the DE miRNAs, such as miR-146, miR-15, and miR-21, function in immune-related signal pathways in mammals. In broiler chicken, Wang et al. [151] suggested miR-34a, miR-122-1, miR-122-2, miR-146a, miR-155, miR-206, miR-1719, miR-1594, miR-1599, and miR-451 as strong candidate miRNAs and MX1, IL-8, IRF-7, and TNFRS19 as strong candidate genes involved in the regulation of host response to AIV infection. Examining chicken embryo fibroblasts infected and not with avian influenza virus H9N2, Peng et al. [152] identified 48 DE miRNAs (e.g., miR-146c, miR-181a, miR-181b, miR-30b, miR-30c, miR-30e, and miR-455), which were predicted to target immune response-related genes. When performing miRNA expression profiling in the spleen, thymus, and bursa in chicken and duck, Li et al. [170] reported divergent changes in the miRNA expression upon H5N1 infection in both breeds and suggested that miRNAs can account for the level of susceptibility upon H5N1 infection. Recently, O'Dowd et al. [171] indicated that miR-146a, miR-146b, miR-205a, miR205b, and miR-449 can be used as miRNA-based antiviral agents or vaccine adjuvants in alternative strategies for the control of AIV in chickens.

Potential Regulatory Roles of miRNAs in Small Ruminant Diseases
Small ruminants, including sheep and goats, are an important source of meat, milk, and wool throughout the world. miRNA studies in small ruminants have focused on muscle [172][173][174], embryo/ovary [175][176][177][178][179], mammary gland development [180][181][182][183], milk-related phenotypes (yield and composition) [183][184][185], and hair/skin-related phenotypes [186][187][188]. An initial study on miRNA in small ruminants performed by Wenguang et al. [187] used microarray analysis to characterize the expression of 159 miRNAs in skin samples from the body and ear of goats and sheep and identified 105 miRNAs conserved between the two species with significant roles in hair follicle differentiation. Subsequent studies using high-throughput sequencing techniques identified 106 precursors and 153 mature miRNAs in sheep and 267 precursors and 436 mature miRNAs in goats (Table 1). miRNAs with important roles in sheep and goat diseases are shown in Table 5.  [197,198] The functions of miRNAs have been reported for several small ruminant diseases, such as Cystic echinococcosis infection [189], an epithelial tumor induced in goats and sheep by enzootic nasal tumor virus (ENTV) [190], bluetongue virus [191], Peste des petits ruminants (PPR) infection [192], and prion disease [193]. Small ruminants are highly susceptible to Cystic echinococcosis, a chronic zoonotic infection caused by infection with the larval stage of the cestode, Echinococcus granulosus. Jiang et al. [189] profiled the miRNA expression in intestinal tissues of sheep with resistant and non-resistant Major Histocompatibility Complex (MHC) haplotypes after peroral infection with E. granulosus eggs and highlighted miR-21-3p, miR-542-5p, miR-671, miR-134-5p, miR-26b, and miR-27a as NF-κB pathwayresponsive miRNAs during E. granulosus infection. miRNAs also play important roles in enzootic nasal adenocarcinoma, an epithelial tumor induced in goats and sheep by enzootic nasal tumor virus [190]. Wang et al. [190] identified 116 DE miRNAs in the tumor and para-carcinoma nasal tissues of Nanjing yellow goats with enzootic nasal adenocarcinoma and showed the involvement of the predicted target genes in cell proliferation, signal transduction, and other processes associated with cancer.
In an effort to explore the mechanisms of bluetongue virus infection, Du et al. [191] identified 25 known and 240 novel DE miRNA candidates in primary sheep testicular cells infected with bluetongue virus as well as the significant enrichment of target genes in MAPK, PI3K-Akt, endocytosis, Hippo, NF-κB viral carcinogenesis, FoxO, and JAK-STAT signaling pathways. Peste des petits ruminants (PPR) is a highly contagious viral disease characterized by fever, sore mouth, conjunctivitis, gastroenteritis, and pneumonia, and primarily affects goats and sheep. Recently, Pandey et al. [192] found that miR-21-3p, miR-320a, and miR-363 might act cooperatively to enhance viral pathogenesis in the lung and spleen of sheep by downregulating several immune response genes in lung and spleen tissues of sheep infected with PPR virus (PPRV). In goats, Qui et al. [196] identified 316 DE miRNAs between the control and infected peripheral blood mononuclear cells with PPR virus. The authors suggested various biological processes and pathways in which DE miRNAs might participate, such as immune functions (miR-150, miR-146, and let-7) or apoptosis (miR-671-5p and miR-182). In two follow-up studies, the authors validated the immunological roles of miR-1 via targeting Tumor Necrosis Factor-Like Weak (TWEAK) and miR-218 via targeting Signaling Lymphocyte Activation Molecular (SLAM) genes, respectively [197,198]. To identify potential miRNA biomarkers for small ruminant prion diseases, Sanz Rubio et al. [193] analyzed ten potential candidate miRNAs from circulating blood plasma of naturally infected scrapie sheep by quantitative reverse transcription PCR and identified miR-342-3p and miR-21-5p as circulating biomarkers of prion disease. Yang et al. [194] reported miR-150, miR-370-3p, and miR-411b-3p as important for PPR vaccine virus responses in both peripheral blood lymphocyte and primary testicular cells from sheep using time course experiments. Examining seronegative and infected sheep with Small Ruminant Lentiviruses, Bilbao-Arribas et al. [195] reported 52 DE miRNAs and suggested miR-21 as a potential biomarker for the severity of lung lesions or a therapeutic target.

Important miRNA Biomarkers in Livestock Diseases
Even though each disease reviewed has its own specific mechanisms, a common feature was the roles of associated miRNAs in the host immune response. miRNAs are essential players in innate immune and inflammatory responses [199,200]. Several notable miRNAs with roles in the innate immune regulations (reviewed in sections above) were associated with three or more livestock diseases, including miR-146, miR-223, and miR-21, etc. (Figure 2).
In humans, miR-146 is known as a key modulator of the immune response in cancers [201]. MiR-146 was associated with most diseases analyzed, such as bovine mastitis, tuberculosis, JD, swine influenza disease, PPRSV, and Avian influenza (Tables 2-4 and Figure 2). MiR-223 is an important miRNA in infection and inflammation [202,203], and it also plays important roles in bovine mastitis and swine influenza (Tables 2 and 3, Figure 2). MiR-21 is an miRNA with multiple roles in human diseases [204], and it is also involved in multiple diseases in different livestock species (Tables 2-5 and Figure 2). For instance, miR-21 modulates the expression of both pro-and anti-inflammatory cytokine responses against influenza A (H1N2) infection in pigs [121] or plays important roles in lymphocyte development and modulation in the lungs of chickens infected with AIV [150]. In addition to the common roles of these miRNAs in the associated diseases, some miRNAs are unique to one disease or a pathogen. For example, some miRNAs could respond to E. coli infections but not to mastitis caused by other pathogens (Table 2). In virus-related diseases, miRNAs are also known to be important for controlling the replication of the virus [205], in which case, they could either impair virus replication (for instance miR-130b [149]) or facilitate virus replication (for instance miR-30c [104]). For bacterial infections, miRNAs can contribute to the host response to the infection via a wide range of pathways and host cells [206]. The miRNAs uniquely associated with a specific disease pathogen or stage of disease pathogenesis are potential targets for the development of biomarkers for prognostic, diagnostic, and therapeutic applications for the management of livestock diseases.

Challenges and Opportunities for Understanding Biological Roles of miRNA
To date, it is well known that miRNAs play important roles in many biological processes related to disease development in farm animals. Therefore, the application of miRNAs to improve disease resistance in farm animals is very promising. miRNAs can be used as direct biomarkers or indirectly through other technologies. As direct biomarkers, such as circulating biomarkers, miRNA in biological fluids, such as blood, milk, saliva, and urine, can facilitate the rapid detection of disease infection status. Indirectly, miRNA can find use in other technologies, such as RNA interference or genome editing. Genome editing using CRISPR/Cas9 technology can robustly, specifically, and stably modify miRNA expression by editing either the seed sequence of miRNAs or the three prime untranslated regions of their target genes [207]. The success of the application of this technology in miRNA-mediated therapy has been proven in diseases in animal models [208][209][210].
Before adopting miRNAs as biomarkers, it is crucial to understand their roles in disease pathogenesis. Although affordable "OMICS"-based technologies have enabled the faster identification of miRNAs, the identification and validation of miRNA functions is still hindered by low sample size and poor reproducibility. Additionally, a holistic approach for exploring and validating miRNA functions, given the complexity of livestock diseases, is lacking. Since many livestock diseases are chronic in nature, miRNA functional studies should consider the different disease stages. Some diseases are also caused by multiple pathogens, such as mastitis or impact numerous tissues, or organs, such as JD; therefore, the spatiotemporal-specific manner of the regulatory function of miRNAs needs to be considered. The lack of sensitive and reliable tools for detecting lowly expressed miRNAs might ignore some potentially important miRNAs with essential functions. Additionally, an miRNA can target hundreds of genes, thus making it difficult, costly, and labor-intensive to validate each miRNA gene target functionally. Lastly, the limited attention to in vivo experiments for miRNA validation is also a significant challenge for understanding miRNA roles in livestock diseases.
Nevertheless, the lower cost of sequencing may lead to an increase in sample sizes in miRNA studies. Moreover, the present downward trend in the cost of sequencing may provide opportunities to sequence multiple types of molecules (miRNAs, lncRNAs, mRNAs, etc.) simultaneously, thereby enhancing the possibility of integrative analyses for the further exploration of miRNA roles in interaction networks. Furthermore, the collaboration by different research groups can significantly improve the power of detection and validation of miRNA functions. Other technologies, such as single-cell sequencing [211], will further the understanding of disease pathogenesis and miRNA functions, while genome editing [212] and RNA interfering technologies [213] could facilitate the identification of the exact target genes and downstream impact of miRNAs on disease pathogenesis. Machine learning and deep learning methods could improve the ability to classify disease pathogens [214] and predict the roles of miRNAs in disease progression [215].

Conclusions
It is without a doubt that miRNAs play significant regulatory roles in livestock disease pathogenesis and have substantial potential as biomarkers for the management of livestock diseases. However, the application of miRNAs in disease management is hindered by many factors, such as inadequate diagnostic tools; lack of assessment for the accuracy, sensitivity, and specificity of miRNAs; and potentially high cost of developing miRNA biomarkers. Nevertheless, given the pressing need to control livestock diseases, a significant increase in miRNA research has been observed in recent years. The lower cost of sequencing or miRNA genotyping and more powerful computing resources and statistical methods are important assets for miRNA studies. Therefore, we believe that miRNA biomarkers will eventually be developed and employed as powerful tools to manage livestock diseases.