Whole transcriptome analysis of Mycobacterium avium subsp. paratuberculosis infection and identification of circulating miRNA as biomarkers for resilience

doi:10.21203/rs.3.rs-3581668/v1

Download PDF

Research Article

Whole transcriptome analysis of Mycobacterium avium subsp. paratuberculosis infection and identification of circulating miRNA as biomarkers for resilience

https://doi.org/10.21203/rs.3.rs-3581668/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Mycobacterium avium subsp. paratuberculosis (MAP), the causative agent of Johne’s disease (JD), results in a chronic enteritis affecting ruminants worldwide and presents to severe economic and welfare issues. Pathogenic mycobacteria such as MAP are able to manipulate host transcriptomic pathways to establish infection and cause disease. Following exposure to pathogenic mycobacteria, host microRNA (miRNA) are modulated, effecting downstream gene expression. miRNA have been utilised as circulating biomarkers for several human diseases and cancers, however their applicability to MAP infection in ruminants remains relatively unexplored. Here, we use a combination of small RNA sequencing and microarray data to explore circulating miRNA and mRNA responses of sheep infected with and resilient to MAP. A total of 58 differentially expressed miRNA were identified in only MAP infected sheep and 25 in only resilient sheep. Early and late timepoint specific signatures were also observed between infected and resilient sheep. Together, our data presents a novel miRNA signature of MAP infection, and resilience to infection. Further, we identify key signalling and immune pathways regulated by these miRNA which may contribute to the resilience phenotype and the ability of these sheep to prevent infection.

Pathogenic mycobacteria such as Mycobacterium avium subsp. paratuberculosis (MAP), which is the causative agent of Johne’s disease (JD), a chronic granulomatous disease affecting animals worldwide, are able to manipulate host immune and transcriptomic systems to evade clearance ^1–3. Large economic and production losses occur due to JD, which also presents a significant animal welfare issue with animals suffering malnutrition and wasting ^4–7. Due to long, silent subclinical phases where infected animals are intermittently shedding bacteria, accurate diagnosis is difficult. Current approaches for the detection of MAP such as direct detection in faeces and serum antibody levels have low sensitivity in the early stages of disease due to the transient nature of faecal shedding and serological responses. This highlights the need for new approaches, with disease state-specific biomarkers a desirable option ⁸. One molecule of interest in biomarker studies is microRNA (miRNA), due to their abundance and stability in circulation, and their ability to yield a diagnostic signature ⁹.

miRNA are short (21–25 nt) single stranded molecules which post-transcriptionally regulate gene expression. Mature miRNA are released into the cytoplasm within protein complexes or microvesicles, and bind to the untranslated region (UTR) of target mRNA to reduce protein translation, either by degradation of the bound mRNA or transient suppression of translation ¹⁰. miRNA have been recognised for their role in controlling gene expression during mycobacterial infection, providing specific signatures in a range of infection models and conditions ^11–13. Identification of miRNA in biological fluids such as serum and plasma has led to circulating biomarkers being considered as a potential avenue for diagnostic or prognostic assays.

Circulating miRNA may provide an easily attainable and relatively non-invasive sample for the detection of MAP, which allow for efficient processing and analysis, especially when compared to current gold-standard post-mortem diagnostics. Investigation of miRNA as biomarkers of JD have successfully identified differential regulation during infection, however no discernible diagnostic signature has been uncovered during the subclinical or clinical stages of infection ^14–16. In M. tuberculosis infection miRNA biomarkers have been shown to distinguish infected individuals from healthy controls and differentiate between active and latent disease ^13,17,18. Similarly, identification of potential miRNA biomarkers in M. bovis infection provides further support to the need for exploration in MAP infection ^19,20.

It is necessary to not only identify miRNA dysregulated in infection, but also to delineate between disease states and resilient animals. Following both natural and experimental exposure to MAP, some animals remain uninfected and have been termed resilient to infection ^21,22. While recovery from infection has also been observed ^23,24, it is difficult to discriminate between the two responses. We therefore define resilience in the context of mycobacterial infection as animals that have received an infectious dose of MAP, whereby the infection did not establish or progress, and was unable to be detected. Identification of resilient animals may be key in managing the spread of JD as current control programs based around the identification and culling of infected animals have remained ineffective ^25,26. It is therefore beneficial to not only identify animals that are infected with MAP, but also animals which have been exposed and remain uninfected or resilient.

Previous profiling of MAP infection has identified predictive gene lists for not only specific disease states (i.e. paucibacillary or multibacillary), but also for resilience to MAP infection ²⁷. As mRNA expression and downstream protein translation are ultimately regulated by miRNA, the dysregulation of these molecules during infection may provide further insight into the host response to MAP and the host pathways responsible for conferring protection against infection.

This study aimed to examine the use of small-RNA sequencing to identify subsets of miRNA expressed specifically in MAP infected or MAP exposed resilient animals. Further exploration of gene regulation using bioinformatic target pairing to mRNA datasets from MAP infected and exposed sheep was explored to uncover potential immune pathways which may contribute to resilient animals’ ability to avoid infection.

In silico analysis provides potential miRNA biomarkers of mycobacterial infection

Through a combination of data available from current literature and further in silico predictions using IPA, a list of 126 miRNA was generated as potential markers of mycobacterial infection. These were then grouped according to known biological functions (Figure 1). miRNA with primarily tumour suppressor roles (and potential tuberculosis markers) were the most prominent grouping making up approximately 23% of the list, followed closely by apoptosis/cell death and TLR/NFκβ signalling with 18.52% and 17.59% respectively.

Identification of circulating small RNAs in ovine plasma

miRNA sequencing of archived plasma samples from MAP infection trials confirmed the ability of Next Generation Sequencing to identify potential biomarkers of infection. From the 24 animals sampled and 2 timepoints, approximately 60% of reads aligned to mature miRNA sequences (Figure 2), while 9.9% and 20.99% of reads were non-mappable or discarded based on quality, respectively.

Small RNA sequencing uncovers distinct miRNA profiles of MAP infection and resilience in sheep

Differential expression analysis revealed distinct miRNA profiles between infected and resilient sheep (Figure 3). Following FDR and fold change thresholding, a total of 140 miRNA were found to be differentially expressed between groups, and 83 miRNA either up or downregulated compared to controls. Regardless of the timepoint, 29 upregulated and 29 downregulated miRNA were specifically differentially expressed in infected animals compared to both resilient and control groups, displaying a general miRNA MAP infection profile (Table 1). Likewise, 10 upregulated and 15 downregulated miRNA were exclusively dysregulated in resilient sheep when compared to control and MAP infected animals (Table 2).

Within the infected and resilient miRNA groups, there were also time-specific profiles, with particular miRNA differentially regulated specifically in either early or late infection or exposure. Within the infected group, there were 15 upregulated and 16 downregulated miRNAs in early infection only, and 19 upregulated and 10 downregulated miRNAs in late infection only (Supplementary Table 1). In resilient exposed animals, early exposure showed 14 upregulated and 13 downregulated miRNA, and 6 upregulated and 12 downregulated miRNA specific to late exposure timepoint (Supplementary Table 2).

As groups contained both animals that were vaccinated against MAP infection with the commercially available Gudair™ vaccine, and animals that were not vaccinated, analysis was performed to identify a vaccine signature. Six miRNA were identified as being potentially regulated due to vaccination, and displayed differential regulation between control vaccinated, infected vaccinated, and resilient vaccinated groups (Table 3).

In silico, microarrayand small RNA sequencing dataset crossover

To assess the applicability of in silico biomarker prediction and bioinformatic target pairing, overlap between predicted IPA, sequencing, and microarray datasets was assessed. There was an 18.75% similarity between the lists of miRNA generated using in silico prediction approach and those differentially expressed in the miRNA sequencing dataset (Figure 4). From the 126 miRNA predicted, 42 were identified in sequencing and their broad function categorised (Table 4)

In the MAP infected cohort, 88.7% of miRNA identified as significantly differentially expressed had either experimentally observed or moderate to high predicted mRNA targets that were also present in the mRNA infected cohort data. From the differentially expressed mRNA, 57.9% were targets of miRNA of interest. While all miRNA in the resilient animals had all targets present in the mRNA dataset, only 28.9% of the resilient mRNA were targets of the identified miRNA.

Integrated miRNA and mRNA analysis reveals biological pathways regulated following MAP exposure and infection

Differentially expressed miRNA were further analysed alongside previously published microarray gene expression data to further investigate the biological pathways modulated by these miRNA. Pathway analysis of miRNA and their target mRNA provided immune snapshots at the early and late stages of infection and exposure to MAP.

Across both infected and resilient animals there were common pathways which may be indicative of exposure to or infection with MAP. Within the early timepoint, cytokine-chemokine signalling pathways (CXCR4, IL-1, IL-17A, IL-8, BAFF), T-cell activation pathways (iCOS-iCOSL), and immune related signalling pathways (mTOR signalling, JAK-STAT signalling) were all differentially regulated in both exposed resilient and infected animals compared to control groups. Similarly, in the late timepoint, cytokine/chemokine pathways (CCR5, CXCR4, IL-17A, IL-8) and immune signalling pathways (JAK-STAT) were dysregulated in both MAP exposed and infected animals.

Infected animals

Following exposure to MAP, infected animals displayed a distinct profile compared to control and MAP resilient exposed groups in the early timepoint (Table 5). Infected sheep displayed a classical inflammatory response, with increased signalling through cytokine-chemokine pathways (CCR5, CXCR4, IL-3, IL-15, IL-8, IL-9), increased engulfment of bacteria through phagocytosis, and increased T cell activity (iCOS-iCOSL, CD28). Increased Th1 responses were evident in MAP infected animals with enhanced pro-inflammatory cytokine gene expression, as well as regulation of genes associated with leukocyte extravasation and chemotaxis towards bacterial stimuli. Further, genes involved in haematopoietic and chemotactic pathways via GM-CSF, IL-3, fMLP in neutrophils, and CXCR4 were upregulated, suggesting current inflammatory immune response to invading bacteria following challenge. Immune signalling cascades were also enhanced, including LPS-stimulated MAPK signalling, natural killer cell signalling, and mTOR related signalling pathways, suggesting pathogen recognition and host responses to clear bacterial infection.

In the late stage of infection, a switch to an anti-inflammatory and pro-mycobacterial response is evident (Table 6). A reduction in Th1 associated responses is observed, alongside decreasing antigen recognition and TCR signalling. T cell activation and responses were dampened in late infection (CD40 signalling, CD28 signalling, iCOS-iCOSL signalling), indicative of chronic granulomatous infection. Decreases in inflammatory cytokine-chemokine pathways (CCR5, CXCR4, IL-1, IL-15, IL-2, IL-6, IL-7, IL-9, IL-17A, IL-8, TNFR1/2) further imply a classic anti-inflammatory immune response in the later stages of MAP infection.

Exposed animals

Animals that were exposed to MAP and later appeared to be resilient were distinct from animals that were or became infected, suggesting a divergent, effective host response (Table 7). In the early stages of exposure, resilient sheep exhibited a non-inflammatory immune response, with T cell receptor (TCR) signalling decreased (Nur77, DC maturation, TOB). Genes associated with cytokine-chemokine signalling were also downregulated (CCR5, CXCR4, IL-15, IL-17A, IL-22, IL-23). Of interest, production of reactive oxygen and nitrogen species was decreased, which may indicate a non-damaging or less destructive host-protective immune response. Similarly, the T cell exhaustion pathway was down regulated in resilient animals in the early stages post exposure. This pathway was not present in infected animals at either timepoint and, may contribute to the observed disease resilience.

Conversely, at the late timepoint post exposure, resilient animals exhibited a classical inflammatory profile, similar to that observed in early infected animals (Table 8). Antigen recognition and TCR signalling was increased (Nur77, FcεRI), while T cell function and exhaustion and lymphocyte apoptosis were also increased suggesting bacterial recognition and active host responses. Further, inflammatory cytokine-chemokine associated genes were altered, resulting in an increase in signalling pathways (CCR5, CCR3, CXCR4, IL-17A, IL-22, IL-7, IL-8).

qPCR validation of small RNA sequencing

Eight differentially expressed miRNA were analysed using qPCR and compared to the fold change provided by the sequencing analysis (Figure 5, Supplementary Figures 1-2). miRs -16b and -486 were chosen as endogenous reference genes for normalisation as they displayed the most stable expression among the housekeeping candidates. In both early and late timepoint samples, biological variation in miRNA levels was more apparent in qPCR detection. However, despite outliers, the overall direction of expression was aligned to the observed sequencing fold change in the majority of groups. This level of agreement between the sequencing and qPCR results for miRNA at both timepoints was expected due to the differences in sensitivity and data normalisation method used between the two methods.

In this study we have successfully identified specific miRNA signatures in plasma from sheep infected with and resilient to MAP and uncovered potential biomarkers to identify MAP infected and resilient animals at early and late stages following exposure. We have shown that circulating miRNA are able to provide robust signatures of expression from low-input samples and that MAP infection results in an acute modulation of host miRNA expression.

The use of miRNA as biomarkers for Johne’s disease has been investigated in several studies using a range of biological samples with varying results ^11,14,16,28. Here, we have uncovered distinct miRNA profiles in sheep that are not only infected with MAP, but also those that are resilient to infection, showing that miRNA can be used as biomarkers and may be able to differentiate between early and late stages of infection or exposure to MAP. It is also clear that using miRNA and small RNA-specific RNA extraction and library preparation may improve the applicability of sequencing and the alignment of reads to known miRNA, an issue apparent in similar work involving MAP infected cattle ^14,29. While miRNA profiling alone may provide biomarker candidates, their downstream effect is often unknown and their relevance to infection, particularly from circulating miRs, must also be investigated.

We further analysed the differentially expressed miRNAs with mRNA data from infected and resilient sheep to identify miRNA-mRNA target pairings and immune pathways regulated consistently regulated by these molecules. Through this we have explored the biological relevance of these miRNA during infection and their potential contribution to disease resilience through their downstream target molecules. The opposing results between infected and resilient animals at each timepoint suggests that the temporal expression of miRs also contributes to the outcome of host-pathogen interactions.

Further exploration of their biological relevance to MAP infection revealed key genes and potential signalling pathways modulated by these miRNAs. Alteration of immune pathways may be crucial factors in whether the host is able to effectively clear bacteria and prevent establishment of an infection, or whether they succumb to disease. MAP infected animals followed the accepted or conventional immune trajectory i.e., early pro-inflammatory response with increased cytokine production and immune cell infiltration, followed by a shift towards an anti-inflammatory profile allowing for persistent granulomatous infection to establish.

Resilient sheep lacked the expected early inflammatory responses following host cells encountering MAP. A unique response observed in resilient animals, distinct from that of infected animals, was the lack of inflammatory responses in early exposed resilient sheep. Although immune responses to infection are complex and dependant on disease phenotype. Generally, responses to MAP are characterised by an early inflammatory cell-mediated immune response with high levels of cytokine production before shifting to a Th2 dominated anti-inflammatory profile ^30–32. While plasma miRNA expression reflects the overall immune response to infection rather than site-specific, it provides a snapshot of the complex and intense interactions occurring at the infection foci. Contrary to typical host responses to invading pathogens there was no increase in anti-inflammatory genes or pathways, suggesting that this is a differing type of response that is non-inflammatory rather than anti-inflammatory. This is beneficial to the host, reducing localised tissue damage and preventing the bacteria from hijacking the host response, possibly contributing to the resilient phenotype of these animals.

Expression analysis revealed a decrease in genes associated with T cell exhaustion at the early timepoint following exposure to MAP, which was not observed in infected animals (Fig. 6). This may be a significant contributor to the prevention of infection through a sustained successful T cell response. Pathway analysis of miRNA-mRNA interactions revealed an overall decrease in T cell exhaustion, however an increase in T cell migration, memory cell differentiation, T cell hyperactivation, and CD8⁺ responses, implying a successful and heightened T cell response. A switch to increased T cell exhaustion signalling in late resilience may also indicate that these animals maintained a sustained T cell response throughout the length of the trial, possibly by exposure to environmental reservoirs of MAP shed from infected animals and further suggests they have remained resilient to infection despite consistent exposure. The importance of T cell responses in immune responses in resilient animals and in those able to control infection is unsurprising, as variations in T cell signalling have been shown to impact disease progression and outcome ^33–35. Prolonged non-inflammatory T cell responses, combined with the lack of inflammatory signals may ensure minimal tissue damage to the host while increase T cell responses clear bacteria, contributing to the protective effect of resilient animals.

Another key immune interaction which may contribute to early resilience to infection is that of MAP and neutrophils. During early infection and resilience, neutrophil responses were differentiated, with activation of CXCR4 and fMLP signalling in infected animals and a reduction in CXCR4 signalling in resilient animals (Figs. 7–8). We hypothesise that this may be to limit the influx of neutrophils to site of infection therefore reducing macrophage co-influx, ultimately reducing the inflammatory profile, and preventing MAP co-opting host macrophages for survival ³⁶. Further, reduction of potent chemoattractant IL-8 may aid in reducing macrophage recruitment and preventing induction of chronic infection ³⁷. Interestingly, IL-1 signalling was increased and may possibly work in parallel with decreasing levels of IL-8 and neutrophils in a protective manner, facilitating MAP killing and promoting bacterial clearance in the absence of macrophages ^38,39. As macrophages are one of the primary immune cells responding to invading pathogens their interaction with MAP is often a determining factor in the outcome of infection. The hijacking of macrophages by pathogenic mycobacteria to provide an intracellular survival niche and form granulomas can be considered detrimental to the host. The increased neutrophil response and possible reduction in macrophage influx to bacteria in resilient animals may allow the host to effectively kill bacteria through NETosis and IL-1β secretion ^40,41 while depriving the bacteria of the opportunity to co-opt macrophages.

These miRNA biomarkers of disease status or resilience provide additional evidence of miRNA regulation of host responses to MAP and their significance as regulators of host gene expression, however further validation studies are required to assess the repeatability and diagnostic potential of the identified infected and resilient specific miRNA. Important to note is that the length of the experimental trial was only approximately 1 year, and that control animals from the early timepoint were not sequenced. While resilient animals may have developed infection if maintained longer, evasion of infection in the early stages post challenge is still key to conferring resilience to MAP infection and warrants further experimental investigation. The lack of early controls may have also introduced a degree of inherent bias and led to the identification of growth or age associated changes in expression as due to either infection or resilience, where further exploration of miRNA expression at early timepoints may provide greater clarity. Further, while these biomarkers may yet provide identifying signatures, their circulating nature may lend itself to a certain degree of bias, with highly abundant cell types producing a higher proportion of the measured miRNA. Future studies of cell-specific miRNA production and the spatio-temporal expression of miRNA during mycobacterial infection can provide greater resolution into the origins of circulating miRNA and their biological relevance in disease.

Taken together, these results suggest that miRNA signatures may reflect host immune response modulations, while future functional analysis may provide further connections between observed miRNA and mRNA expression. We have identified several promising miRNA candidates for detecting animals exposed and resilient or infected with MAP and have explored the effect on their target mRNAs and downstream contribution to immunity. Several immune signalling pathways were differentially activated or inhibited in resilient animals and may yet prove to be key to conferring resistance to MAP infection. While further work is required to investigate the diagnostic potential of these miRNA, we have demonstrated their potential as non-invasive biomarkers for Johne’s disease and examined how these may confer resilience through functional analysis.

The use of animals (Merino sheep) in this study was approved by the University of Sydney Animal Ethics Committee. All methods were carried out in accordance with relevant guidelines and regulation. The study was carried out in compliance with the ARRIVE guidelines.

in silico biomarker analysis

A list of potential miRNA biomarkers was compiled through a literature search for miRNA that have been reported as differentially expressed or as potential biomarkers of all mycobacterial infections, as well as related inflammatory diseases such as Crohn’s disease (CD) and Ulcerative colitis (UC) involving MAP. The search terms included “Mycobacteria + paratuberculosis + miRNA + microRNA + mycobacterium” and were input into both the Scopus database and Abcam’s Firefly Discovery Engine (accessed April 2018).

This list was uploaded to Ingenuity Pathway Analysis (IPA). The Ingenuity Knowledge base contains a database of curated experimental findings and interactions and was used to investigate further miRNA involved in mycobacterial infection. As the current version of IPA does not support bovine or ovine miRNA searches, all miRNA sequences were converted to their human orthologs, displaying between 90-100% homology due to the broad evolutionary conservation of miRNA. A core analysis was run on the candidate biomarker list and miRNAs added based on associated disease conditions, function, and commonalities between networks using IPA’s build tool (as of April 2018). Once all extra miRNA were added to the original analysis, the molecules were connected to create one large network.

In order to identify biologically significant pathways and key miRNA, smaller focused networks were taken from the original analysis based on relevance to MAP pathogenesis and host response to infection.

miRNA sequencing

Sample collection and study design

Archived ovine plasma samples from MAP exposure trials run at the University of Sydney were used for miRNA sequencing. Plasma from whole blood collected via jugular venepuncture into EDTA tubes was used for RNA extraction and miRNA sequencing and had been stored at -80°C for approximately 5-6 years.

Inoculation of animals and experimental design was performed as previously described ^21,27. Briefly, 2-4 month old merino lambs were sourced from a MAP-free flock and separated into control and exposed groups in separate paddocks, and further separated into vaccinated and non-vaccinated groups. Absence of MAP infection was confirmed through whole-flock faecal tests. Exposed animals were challenged with three oral doses of MAP inoculum administered over a period of a month (S strain Telford 9.2) resulting in a cumulative dose of 3.18 × 10⁹ viable MAP.

Following detection of clinical disease (loss of >10% body weight) or the termination of the trial (14 months), animals were sacrificed via intravenous injection of barbiturate (Lethabarb™) at 0.5 ml/kg bodyweight, and samples collected during necropsy. Detection of MAP via direct PCR (HTJ) and histopathology analysis confirmed infection status and lesion classification ^42,43.

Animals were retrospectively selected for miRNA profiling based on disease status, histology classification, and vaccination status (Table 9). Samples that had been collected at 13 weeks post infection (early timepoint) and 49 weeks post infection (late timepoint) were selected for small RNA sequencing.

Case definition

Animals were classified based on current case definitions for paratuberculosis from histopathology and tissue culture results ⁴⁴. Briefly, animals with post-mortem gut and associated lymphoid tissue lesions that returned a positive MAP tissue culture result were considered infected. Further histopathology for acid-fast bacilli (AFB) and MAP specific lesions had been performed to delineate paratuberculosis pathology, i.e., paucibacillary lesions (low numbers of AFB in focal lesions), and or multibacillary lesions (diffuse lesions with large numbers of AFB). Animals that were exposed to an infectious dose of MAP but failed to become infected were classified as resilient ²¹. Resilient animals that did not succumb to disease showed evidence of exposure to MAP, however, infection was not able to be detected at the time of necropsy.

RNA extraction

RNA was extracted from 48 frozen archived plasma samples, 24 early timepoint samples and 24 late timepoint samples. All centrifugation steps were performed at room temperature using either an Eppendorf 5418 or 5415 D centrifuge (Eppendorf, Australia). Total RNA was extracted using the Qiagen miRNeasy Serum/Plasma Advanced according to the manufacturer’s instructions. Briefly, 2 mL of plasma from each animal was thawed and aliquots used for multiple extractions. Preliminary experiments showed no added benefit of processing volumes larger than 400 µl of plasma on a single RNeasy UCP column, or by 2 sequential elutions in half volumes. All extractions were performed on 400 µl of plasma, with 200 µl of plasma extracted concurrently and reserved for quantitative (q) PCR validation of downstream data. The miRNeasy Serum/Plasma Advanced protocol was adapted according to each aliquot volume per manufacturer’s instructions. All extracts were stored at -80°C until further processing.

RNA concentration

All RNA extracts of the same sample were combined and concentrated into a small volume suitable for sequencing library preparation using the Qiagen RNeasy MinElute Cleanup Kit following manufacturer’s instructions. The final volume of RNA for each sample was 10 µl and eluates were stored at -80°C until processing for library preparation.

RNA quality control and library preparation

RNA quality control (QC), library generation and sequencing were performed by the Ramaciotti Centre for Genomics, University of New South Wales, with QC performed on an Agilent Bioanalyzer (average RNA integrity number (RIN) = 2.63, average total available RNA = 5.39 ng). Library generation was carried out with the QIAseq miRNA Library Kit (Qiagen, Germany) according to the manufacturer’s protocol. The QIAseq miRNA kit has 6 main steps to the library preparation workflow and is detailed below (summarised in Figure 9).

Sequencing and data analysis

Libraries were sequenced on an Illumina NextSeq 500 with a 1x75bp High Output flowcell, allowing for approximately 10 million reads per sample. Due to low RNA concentrations, no early control and exposed non-vaccinated animals were sequenced.

Primary data analysis was performed using CLC Genomics Workbench (v 12.03) and the Biomedical Genomics Plugin QIAseq miRNA workflow (v 1.2.1) with default parameters. Briefly, raw demultiplexed fastq files were uploaded for quality control. All sequences were quality checked using FastQC (minimum Phred score > 25), and any low-quality reads without adaptor sequences were removed. Next, insert sequences less than 16 bp, or reads with Unique Molecular Indices (UMI) sequences less than 10 bp were discarded and the remaining reads progressed. UMI reads were generated using the QIAseq Small RNA reference set (v 22.) and mapped to Ovis aries and Bos taurus reference sets. Raw reads were also mapped to noncoding RNA databases piRBase (v 2.0) and NONCODE (v 5.0). Prior to differential expression analysis, sequencing data was normalised using the TMM method (trimmed mean of M values).

Differential expression analysis was performed on UMI reads for mature miRNA and analysed either across group with an ANOVA-like analysis, or against a control group for time-point comparisons. P values of significantly differentially expressed miRNA (FC ≥ ± 1.5) were adjusted (q value) to control for multiple testing at FDR of 10% using the Benjamini-Hochberg method. Differentially expressed miRNA (q < 0.05) were then investigated for potential targets and pathway analysis.

Bioinformatics and pathway analysis

To investigate not only the miRNA response to infection but whole transcriptomic responses, gene lists of differentially expressed miRNA for each treatment were uploaded to Ingenuity Pathway Analysis (IPA) for further in silico analysis alongside previously published microarray data of MAP infected and resilient sheep (GEO accession number GSE114384) ²⁷. Due to the lack support for bovine and ovine miRNA sequences in the current version of IPA, human orthologs were uploaded, ensuring 90-100% homology. Owing to this, bovine or ovine specific sequences were not analysed in the pathway analysis.

The combined miRNA and mRNA dataset was analysed to determine the degree of overlap between the datasets and biological pathways of importance in infected and resilient animals. Firstly, a microRNA target filter analysis was performed to provide a list of experimentally observed or highly predicted target mRNAs and filtered to genes found in both the target filter analysis and the published gene lists. The genes were then further filtered based on expression pairing, i.e., opposing directional responses from the miRNA and mRNA target. A core analysis was then performed with the filtered dataset to identify significant canonical pathways altered in infected and resilient animals. Z-scores ≤ 2 or ≥ 2 and -log p values > 1.3 were considered significant. The IPA generated z-score is a statistical representation of the agreement between the overall direction of the pathway or relationship and the observed expression of genes involved in that pathway or relationship ⁴⁵. While z-scores ≤ 2 or ≥ 2 were considered significant, pathways with directionality ≤ 1 or ≥ 1 were considered biologically relevant and were included in the analysis. Immunological pathways were chosen for further investigation to determine the underlying immune basis for resilience to MAP infection and the biological role of the identified miRNA.

Validation of sequencing results

To validate gene expression changes observed in the miRNA-seq data, 8 differentially expressed genes were selected for validation using qPCR as previously described ⁴⁶. Briefly, cDNA was generated from 10 µl of RNA using the miScript II RT kit (Qiagen) and HiSpec buffer according to the manufacturer’s instructions. qPCR was carried out on an Mx3000p Real-time PCR system using Quantitect SYBR Green PCR Mastermix, miScript Universal Primer, miScript miRNA specific primer, and 5 µl of 1/11 diluted cDNA. All miRNA primers were obtained from Qiagen and reconstituted in 550 µl of TE prior to use (Supplementary Table 3). Cycling conditions were: 95°C for 15 minutes; 40 cycles of 95°C for 20 seconds, 56°C for 30 seconds, 72°C for 30 seconds with fluorescence data acquisition occurring at the end of each cycle, followed by 1 cycle of 95°C for 1 minute, 65°C for 30 seconds, and 97°C for 30 seconds to obtain dissociation curves. Each sample was run in triplicate and no template control and standard curves included on each plate.

To determine the most appropriate reference miRNA, read counts and UMI data were analysed with geNorm and the top 10 most stable miRNA provided. The top 5 ranked reference miRNA were analysed as potential endogenous reference controls (Figure 10). Raw Cq values were transformed using the common base method ⁴⁷ and the resulting efficiency-weighted Cq values (C(w)q) used for fold change analysis. Two-tailed t-tests were performed on ΔC(w)q values to analyse differences in expression between treatment groups. Data is presented as fold change from control non-vaccinated animals.

a. Ethics approval and consent to participate

b. Consent for publication

not applicable

c. Availability of data and materials

Analysed data are presented within the current manuscript and in the supplementary data. Raw microarray data used for mRNA expression analysis The raw microarray data have been deposited in the National Centre for Biotechnology Information’s Gene Ex- pression Omnibus (GEO) and are accessible through GEO Series accession number GSE114384.

d. Competing interests

The authors declare that they have no competing interests.

e. Funding

This work was supported by Meat and Livestock Australia (P.PSH 0576). The funding body did not contribute to the study design or collection, analysis and interpretation of data, or the writing of the manuscript.

f. Authors' contributions

K.W., K.D.S., K.M.P. and A.C.P. all contributed to the study design and data analysis. K.W. performed experiments and wrote the manuscript. K.D.S., K.M.P. and A.C.P. all supervised the study and contributed to the manuscript. All authors read, edited, and approved the final manuscript.

Acknowledgments

The authors would like to acknowledge all past and current members of Farm Animal Health laboratory and field team for collection, processing, and archiving of samples. We also acknowledge Ms Rebecca Maurer for invaluable laboratory assistance, and the Sydney Informatics Hub for the providing access to data analysis software.

Johansen, M. D., de Silva, K., Plain, K. M., Whittington, R. J. & Purdie, A. C. Mycobacterium avium subspecies paratuberculosis is able to manipulate host lipid metabolism and accumulate cholesterol within macrophages. Microbial Pathogenesis 130, 44-53, doi:https://doi.org/10.1016/j.micpath.2019.02.031 (2019).
Purdie, A. C., Plain, K. M., Begg, D. J., de Silva, K. & Whittington, R. J. Expression of genes associated with the antigen presentation and processing pathway are consistently regulated in early Mycobacterium avium subsp. paratuberculosis infection. Comparative Immunology, Microbiology and Infectious Diseases 35, 151-162, doi:http://dx.doi.org/10.1016/j.cimid.2011.12.007 (2012).
Park, H.-E., Park, H.-T., Jung, Y. H. & Yoo, H. S. Gene expression profiles of immune-regulatory genes in whole blood of cattle with a subclinical infection of Mycobacterium avium subsp. paratuberculosis. PLOS ONE 13, e0196502, doi:10.1371/journal.pone.0196502 (2018).
Bush, R. D. The Biological And Financial Impact Of Ovine Johne’s Disease In Australia Doctor of Philosophy(PhD) thesis, University of Sydney, (2005).
Bush, R. D., Windsor, P. A. & Toribio, J. A. Losses of adult sheep due to ovine Johne's disease in 12 infected flocks over a 3-year period. Australian Veterinary Journal 84, 246-253, doi:10.1111/j.1751-0813.2006.00001.x (2006).
Sergeant, E. S. Estimated flock-prevalence and distribution of ovine Johne's disease in Australia at December 2001. Aust Vet J 81, 768-769, doi:10.1111/j.1751-0813.2003.tb14612.x (2003).
Fernández-Silva, J. A., Correa-Valencia, N. M. & Ramírez, N. F. Systematic review of the prevalence of paratuberculosis in cattle, sheep, and goats in Latin America and the Caribbean. Tropical Animal Health and Production 46, 1321-1340, doi:10.1007/s11250-014-0656-8 (2014).
Nielsen, S. S. & Toft, N. Ante mortem diagnosis of paratuberculosis: A review of accuracies of ELISA, interferon-γ assay and faecal culture techniques. Veterinary Microbiology 129, 217-235, doi:https://doi.org/10.1016/j.vetmic.2007.12.011 (2008).
Witwer, K. W. Circulating MicroRNA Biomarker Studies: Pitfalls and Potential Solutions. Clinical Chemistry 61, 56-63, doi:10.1373/clinchem.2014.221341 (2015).
Duchaine, T. F. & Fabian, M. R. Mechanistic Insights into MicroRNA-Mediated Gene Silencing. Cold Spring Harbor Perspectives in Biology (2018).
Wright, K., Plain, K., Purdie, A., Saunders, B. M. & de Silva, K. Biomarkers for Detecting Resilience against Mycobacterial Disease in Animals. Infection and Immunity 88, e00401-00419, doi:10.1128/IAI.00401-19 (2019).
Sabir, N. et al. miRNAs in Tuberculosis: New Avenues for Diagnosis and Host-Directed Therapy. Frontiers in microbiology 9, 602-602, doi:10.3389/fmicb.2018.00602 (2018).
Barry, S. E. et al. Identification of a plasma microRNA profile in untreated pulmonary tuberculosis patients that is modulated by anti-mycobacterial therapy. Journal of Infection 77, 341-348, doi:https://doi.org/10.1016/j.jinf.2018.03.006 (2018).
Farrell, D. et al. The Identification of Circulating MiRNA in Bovine Serum and Their Potential as Novel Biomarkers of Early Mycobacterium avium subsp paratuberculosis Infection. PLoS ONE 10 (2015).
Gupta, S. K. et al. Detection of microRNA in cattle serum and their potential use to diagnose severity of Johne's disease. Journal of Dairy Science 101, 10259-10270, doi:https://doi.org/10.3168/jds.2018-14785 (2018).
Malvisi, M. et al. Responses of Bovine Innate Immunity to Mycobacterium avium subsp. paratuberculosis Infection Revealed by Changes in Gene Expression and Levels of MicroRNA. PLoS One 11, e0164461, doi:10.1371/journal.pone.0164461 (2016).
Ueberberg, B., Kohns, M., Mayatepek, E. & Jacobsen, M. Are microRNAs suitable biomarkers of immunity to tuberculosis? Molecular and Cellular Pediatrics 1, 8, doi:10.1186/s40348-014-0008-9 (2014).
Zhang, X. et al. Screening and Identification of Six Serum microRNAs as Novel Potential Combination Biomarkers for Pulmonary Tuberculosis Diagnosis. PLOS ONE 8, e81076, doi:10.1371/journal.pone.0081076 (2013).
Vegh, P. et al. MicroRNA profiling of the bovine alveolar macrophage response to Mycobacterium bovis infection suggests pathogen survival is enhanced by microRNA regulation of endocytosis and lysosome trafficking. Tuberculosis 95, 60-67, doi:https://doi.org/10.1016/j.tube.2014.10.011 (2015).
Golby, P., Villarreal-Ramos, B., Dean, G., Jones, G. J. & Vordermeier, M. MicroRNA expression profiling of PPD-B stimulated PBMC from M. bovis-challenged unvaccinated and BCG vaccinated cattle. Vaccine 32, 5839-5844, doi:https://doi.org/10.1016/j.vaccine.2014.07.034 (2014).
de Silva, K., Plain, K., Purdie, A., Begg, D. & Whittington, R. Defining resilience to mycobacterial disease: Characteristics of survivors of ovine paratuberculosis. Veterinary Immunology and Immunopathology 195, 56-64, doi:https://doi.org/10.1016/j.vetimm.2017.11.008 (2018).
Stinson, K. J., Baquero, M. M. & Plattner, B. L. Resilience to infection by Mycobacterium avium subspecies paratuberculosis following direct intestinal inoculation in calves. Veterinary Research 49, 58, doi:10.1186/s13567-018-0553-7 (2018).
Begg, D. J. et al. Immunopathological changes and apparent recovery from infection revealed in cattle in an experimental model of Johne’s disease using a lyophilised culture of Mycobacterium avium subspecies paratuberculosis. Veterinary Microbiology (2018).
Dennis, M. M., Reddacliff, L. A. & Whittington, R. J. Longitudinal Study of Clinicopathological Features of Johne’s Disease in Sheep Naturally Exposed to Mycobacterium avium Subspecies Paratuberculosis. Veterinary Pathology 48, 565-575, doi:10.1177/0300985810375049 (2010).
Windsor, P. & Whittington, R. Ovine Paratuberculosis Control in Australia Revisited. Animals 10, 1623 (2020).
Whittington, R. et al. Control of paratuberculosis: who, why and how. A review of 48 countries. BMC Veterinary Research 15, 198, doi:10.1186/s12917-019-1943-4 (2019).
Purdie, A. C., Plain, K. M., Begg, D. J., de Silva, K. & Whittington, R. J. Gene expression profiles during subclinical Mycobacterium avium subspecies paratuberculosis infection in sheep can predict disease outcome. Scientific reports 9, 8245-8245, doi:10.1038/s41598-019-44670-w (2019).
Shaughnessy, R. G. et al. Identification of microRNAs in bovine faeces and their potential as biomarkers of Johne’s Disease. Scientific Reports 10, 5908, doi:10.1038/s41598-020-62843-w (2020).
Coenen-Stass, A. M. L. et al. Evaluation of methodologies for microRNA biomarker detection by next generation sequencing. RNA biology 15, 1133-1145, doi:10.1080/15476286.2018.1514236 (2018).
Sohal, J. S. et al. Immunology of mycobacterial infections: With special reference to Mycobacterium avium subspecies paratuberculosis. Immunobiology 213, 585-598, doi:https://doi.org/10.1016/j.imbio.2007.11.002 (2008).
Stabel, J. R. Host responses to Mycobacterium avium subsp. paratuberculosis: a complex arsenal. Animal Health Research Reviews 7, 61-70, doi:10.1017/S1466252307001168 (2007).
Begg, D. J. et al. Does a th1 over th2 dominancy really exist in the early stages of mycobacterium avium subspecies paratuberculosis infections? Immunobiology 216, 840-846 (2011).
Leite, F. L. et al. ZAP-70, CTLA-4 and proximal T cell receptor signaling in cows infected with Mycobacterium avium subsp. paratuberculosis. Veterinary Immunology and Immunopathology 167, 15-21, doi:https://doi.org/10.1016/j.vetimm.2015.06.017 (2015).
Okagawa, T. et al. Bovine Immunoinhibitory Receptors Contribute to Suppression of Mycobacterium avium subsp. paratuberculosis-Specific T-Cell Responses. Infection and immunity 84, 77-89, doi:10.1128/IAI.01014-15 (2015).
Roussey, J. A., Steibel, J. P. & Coussens, P. M. Regulatory T Cell Activity and Signs of T Cell Unresponsiveness in Bovine Paratuberculosis. Frontiers in Veterinary Science 1, doi:10.3389/fvets.2014.00020 (2014).
Silva, M. T. When two is better than one: macrophages and neutrophils work in concert in innate immunity as complementary and cooperative partners of a myeloid phagocyte system. Journal of Leukocyte Biology 87, 93-106, doi:https://doi.org/10.1189/jlb.0809549 (2010).
Lowe, D. M., Redford, P. S., Wilkinson, R. J., O’Garra, A. & Martineau, A. R. Neutrophils in tuberculosis: friend or foe? Trends in Immunology 33, 14-25, doi:https://doi.org/10.1016/j.it.2011.10.003 (2012).
Kenefick, K. B., Adams, J. L., Steinberg, H. & Czuprynski, C. J. In vivo administration of a monoclonal antibody against the type I IL-1 receptor inhibits the ability of mice to eliminate Mycobacterium paratuberculosis. J Leukoc Biol 55, 719-722, doi:10.1002/jlb.55.6.719 (1994).
Mayer-Barber, K. D. et al. Cutting Edge: Caspase-1 Independent IL-1β Production Is Critical for Host Resistance to Mycobacterium tuberculosis and Does Not Require TLR Signaling In Vivo. The Journal of Immunology 184, 3326, doi:10.4049/jimmunol.0904189 (2010).
Filio-Rodríguez, G. et al. In vivo induction of neutrophil extracellular traps by Mycobacterium tuberculosis in a guinea pig model. Innate Immun 23, 625-637, doi:10.1177/1753425917732406 (2017).
Ladero-Aunon, I. et al. Bovine Neutrophils Release Extracellular Traps and Cooperate With Macrophages in Mycobacterium avium subsp. paratuberculosis clearance In Vitro. Front Immunol 12, 645304, doi:10.3389/fimmu.2021.645304 (2021).
Plain, K. M. et al. High-Throughput Direct Fecal PCR Assay for Detection of Mycobacterium avium subsp. paratuberculosis in Sheep and Cattle. Journal of Clinical Microbiology 52, 745-757, doi:10.1128/JCM.03233-13 (2014).
Pérez, V., Marín, J. F. G. & Badiola, J. J. Description and classification of different types of lesion associated with natural paratuberculosis infection in sheep. Journal of Comparative Pathology 114, 107-122, doi:https://doi.org/10.1016/S0021-9975(96)80001-6 (1996).
Whittington, R. J. et al. Case definition terminology for paratuberculosis (Johne’s disease). BMC Veterinary Research 13, 328, doi:10.1186/s12917-017-1254-6 (2017).
Krämer, A., Green, J., Pollard, J., Jr. & Tugendreich, S. Causal analysis approaches in Ingenuity Pathway Analysis. Bioinformatics 30, 523-530, doi:10.1093/bioinformatics/btt703 (2014).
Wright, K., de Silva, K., Purdie, A. C. & Plain, K. M. Comparison of methods for miRNA isolation and quantification from ovine plasma. Scientific Reports 10, 825, doi:10.1038/s41598-020-57659-7 (2020).
Ganger, M., Dietz, G. & Ewing, S. A common base method for analysis of qPCR data and the application of simple blocking in qPCR experiments. BMC Bioinformatics 18, doi:10.1186/s12859-017-1949-5 (2017).

Table 1. Differentially expressed miRNA in MAP infected sheep.

Infected
miRNA	Fold Change	q value	miRNA	Fold Change	q value
oar-let-7a	-1.51	0.0001	bta-miR-326	-2.03	0.0400
oar-let-7b	-1.58	1.13E-10	bta-miR-328	2.13	1.08E-05
oar-let-7c	-1.89	9.39E-09	bta-miR-335	1.68	0.0007
bta-let-7d	2.03	0.0049	bta-miR-338	6.67	0.0018
oar-let-7d	-1.73	5.43E-07	bta-miR-342	1.66	1.82E-09
bta-miR-101	-1.55	1.53E-05	bta-miR-345-3p	2.14	1.77E-06
bta-miR-10175-5p	4.32	0.0052	bta-miR-3613a	-3.81	0.0005
bta-miR-103	1.73	0.0071	oar-miR-362	3.69	0.0300
oar-miR-106b	-1.74	0.0400	bta-miR-365-3p	1.9	0.0003
bta-miR-12034	2.85	1.64E-05	bta-miR-374b	4.26	0.0400
bta-miR-122	1.53	0.0014	oar-miR-376d	-3.19	0.0400
bta-miR-1224	2.18	0.0060	bta-miR-378	1.56	2.29E-05
bta-miR-1246	1.51	0.0079	bta-miR-378c	1.87	0.0100
bta-miR-1306	2.37	1.72E-05	oar-miR-379-5p	-2.01	0.0003
bta-miR-130b	1.93	0.0005	oar-miR-381-3p	4.28	0.0400
bta-miR-144	-1.87	2.99E-06	oar-miR-382-3p	-2.48	0.0400
bta-miR-145	1.5	0.0001	oar-miR-382-5p	-1.82	5.76E-07
bta-miR-149-5p	-4.2	0.0010	oar-miR-409-5p	-4.42	0.0034
oar-miR-152	-1.58	2.93E-08	oar-miR-412-3p	-4.17	0.0057
oar-miR-17-5p	-1.58	0.0100	bta-miR-423-5p	-1.66	5.01E-14
bta-miR-18a	-2.47	0.0400	bta-miR-4286	-5.36	0.0002
bta-miR-192	1.58	1.04E-05	bta-miR-455-5p	-2.58	0.0400
bta-miR-199c	-3.48	0.00E+00	oar-miR-487b-3p	2.02	0.0057
bta-miR-204	1.83	0.0400	bta-miR-491	-4.44	0.0038
bta-miR-205	-1.57	2.50E-06	oar-miR-493-3p	-2.07	0.0500
bta-miR-214	-3.89	0.0100	bta-miR-504	2.3	0.0500
bta-miR-2285cc	-2.06	0.0074	bta-miR-6529a	1.54	0.0006
bta-miR-30b-5p	2.54	4.97E-10	bta-miR-671	-2.3	0.0200
bta-miR-30c	1.97	8.29E-06	bta-miR-874	1.74	0.0004

Table 2. Differentially expressed miRNA in MAP exposed resilient sheep.

Exposed
miRNA	Fold Change	q value	miRNA	Fold Change	q value
bta-let-7f	-1.74	1.31E-05	bta-miR-2285bf	-1.76	0.0300
oar-let-7f	-1.59	4.65E-05	bta-miR-296-5p	1.77	0.0003
oar-miR-103	-1.54	0.0300	bta-miR-29b	2.82	0.0015
bta-miR-143	-2.22	0.0002	bta-miR-30f	-1.94	0.0500
bta-miR-146b	-5.47	0.0001	bta-miR-3120	-3.22	0.0500
bta-miR-151-5p	1.64	0.0072	oar-miR-369-3p	3.24	0.0036
bta-miR-183	2.25	0.0500	bta-miR-429	-2.3	0.0400
bta-miR-1839	4.6	0.0400	bta-miR-450b	-4.28	0.0200
bta-miR-184	3.02	0.0100	bta-miR-6525	1.91	0.0200
bta-miR-199a-5p	-2.61	0.0024	bta-miR-769	-2.91	0.0007
bta-miR-21-5p	4.61	0.0049	bta-miR-92b	-2.31	0.0300
oar-miR-218a	-2.02	0.0500	bta-miR-9-5p	7.39	0.0042
oar-miR-26a	-1.66	1.17E-07

Table 3. miRNA differentially regulated due to vaccination.

miRNA	Fold Change
miRNA	Control	q value	Exposed	q value	Infected	q value
bta-miR-1	5.35	8.13E-09	4.02	3.27E-05	4.48	1.77E-08
bta-miR-199a-5p	3.37	0.0022	3.46	0.0200	-18.60	4.24E-06
bta-miR-532	1.55	0.0014	1.89	1.78E-07	-2.04	0.0001
bta-miR-99a-3p	-3.22	0.0200	7.02	0.0100	4.78	0.0400
oar-miR-103	-2.03	1.06E-06	-1.81	0.0025	2.94	0.0005
oar-miR-106b	-2.14	0.0300	-2.96	0.0086	4.19	0.0007

Table 4. Functional grouping of miRNA observed in both in silico and sequencing analysis.

Functional Grouping	Differentially Expressed miRNA
TLR/NFκβ signalling	4
Apoptosis/Cell death	7
Autophagy/phagocytosis	6
Lipid Pathways	4
Cytokines/Inflammatory response	8
Macrophage activity	2
Tumour suppressor - potential TB marker	11

Table 5. Regulation of biological pathways by miRNA-mRNA interactions in early infection.

Pathway		z-score	-log (P-value)
TCR signalling/Ag recognition
DC/NK cell crosstalk	n	1.866	0.575	n activation
TLR signalling	n	-1.807	0.26	n inhibition
Cytokine-chemokine signalling
fMLP in neutrophils	n	3.569	1.62
CXCR4 signalling	n	3.434	3.14
IL-2 signalling	n	3.411	0.947
CCR3 in eosinophils	n	2.887	1.82
IL-3 signalling	n	2.556	2.01
NFAT in the immune response	n	2.496	2.2
IL-15 signalling	n	2.2	0.764
Th1 pathway	n	2.2	0.921
IL-8 signalling	n	2	2.13
IL-9 signalling	n	1.667	0.55
GM-CSF signalling	n	1.569	1.89
Leukocyte extravasation	n	1.286	0.762
IL-17A signalling in gastric cells	n	-1.134	0.305
TNFR2 signalling	n	-1.134	0.478
IL-22 signalling	n	-1.508	1.48
PKR in interferon induction	n	-1.667	1.04
IL-1 signalling	n	-1.698	1.14
Phagocytosis
Macropinocytosis	n	2.294	1.15
Fcγ receptor mediated phagocytosis	n	1.859	1.43
T cell function
iCOS-iCOSL in Th cells	n	2.959	2.27
CD28 signalling in Th cells	n	2.121	0.783
PKCθ in T lymphocytes	n	2.082	0.735
Signalling cascades
p70S6K signalling	n	2.214	1.14
JAK-STAT signalling	n	2.117	0.928
Natural killer cell signalling	n	1.82	1.32
mTOR signalling	n	1.732	0.996
NfκB signalling	n	1.698	0.597
LPS-stimulated MAPK signalling	n	1.461	1.23
Apoptosis/cell death
Ca²⁺ induced lymphocyte apoptosis	n	1.5281	1.75
Apoptosis signalling	n	-1.151	1.54
PD-1, PD-1L pathway	n	-2.646	1.03
Humoral response
Th2 pathway	n	3.656	1.22
B cell activating factor signalling	n	-1.414	1.77

Table 6. Regulation of biological pathways by miRNA-mRNA interactions in late infection.

Pathway		z-score	-log (P-value)
TCR signalling/Ag recognition
PRRs & bacterial recognition	n	-1.698	1.09	n activation
IRF activation by cytosolic PRRs	n	-1.291	1.48	n inhibition
CD40 signalling	n	-1.279	4.66
Cytokine-chemokine signalling
IL-7 signalling	n	-2.183	1.55
TNFR1 signalling	n	-2.138	2.42
fMLP in neutrophils	n	-2.058	5.37
IL-1 signalling	n	-2	3.41
Th1 pathway	n	-1.8	1.63
IL-9 signalling	n	-1.667	1.81
CCR5 in macrophages	n	-1.508	1.85
TNFR2 signalling	n	-1.414	2.66
IL-6 signalling	n	-1.406	6.74
IL-8 signalling	n	-1.313	8.72
IL-15 signalling	n	-1.279	3.15
NFAT in the immune response	n	-1.265	5.14
CXCR4 signalling	n	-1.151	4.72
IL-2 signalling	n	-1.147	4.11
IL-17A signalling in gastric cells	n	-1.134	2.2
Phagocytosis
Macropinocytosis	n	-1.069	1.97
T cell function
CD28 signalling in Th cells	n	-2.785	3.46
PKCθ in T lymphocytes	n	-2.535	4.16
iCOS-iCOSL in Th cells	n	-2.2	3.03
Signalling cascades
LPS-stimulated MAPK signalling	n	-1.8	3.79
PI3K/AKT signalling	n	-1.333	5.22
JAK-STAT signalling	n	-1.225	3.54
Apoptosis/cell death
PD-1, PD-1L pathway	n	1.279	1.54
Necroptosis signalling	n	-1.372	2.01
Apoptosis signalling	n	-1.622	8.97
Humoral response
B cell receptor signalling	n	-1.938	3.89
B cell activating factor signalling	n	-1.897	1.95
Pi3K in B lymphocytes	n	-1.298	4.69
FcγRIIIβ in B lymphocytes	n	-1.155	1.71

Table 7. Regulation of biological pathways by miRNA-mRNA interactions in early resilience.

Pathway		z-score	-log (P-value)
TCR signalling/Ag recognition
Nur77 in T lymphocytes	n	-1.807	0.754	n activation
Antiproliferative TOB in T cell signalling	n	-1.265	1.1	n inhibition
Dendritic cell maturation	n	-1.183	1.58
Cytokine-chemokine signalling
IL-8 signalling	n	1.938	3.2
iNOS signalling	n	1.414	0.53
Th1 pathway	n	1.279	1.33
IL-2 signalling	n	1.265	0
IL-1 signalling	n	1.213	1.24
IL-15 production	n	-1.091	0.362
Leukocyte extravasation signalling	n	-1.219	0.588
CXCR4 signalling	n	-1.569	0.311
IL-23 signalling	n	-1.633	0
CCR5 in macrophages	n	-1.667	0.851
IL-17A signalling in gastric cells	n	-1.89	2.04
PKR in interferon induction	n	-2.403	1.72
NO & RO production	n	-2.475	0
IL-22 signalling	n	-2.646	0.775
T cell function
iCOS-iCOSL in Th cells	n	1.4	2.09
T cell exhaustion	n	-1.392	1.38
Signalling cascades
JAK/STAT signalling	n	1.886	0.527
mTOR signalling	n	-1.134	0
Humoral response
Th2 pathway	n	1.8	1.72
B cell receptor signalling	n	-1.508	2.2

Table 8. Regulation of biological pathways by miRNA-mRNA interactions in late resilience.

Pathway		z-score	-log (P-value)
TCR signalling/Ag recognition
Fc epsilon RI signalling	n	2.065	0.475	n activation
Nur77 in T lymphocytes	n	1.147	2.68	n inhibition
Cytokine-chemokine signalling
eNOS signalling	n	2.353	1.32
CCR5 in macrophages	n	1.941	2.43
IL-8 signalling	n	1.938	3.2
IL-22 signalling	n	1.667	2.04
IL-17A signalling in gastric cells	n	1.414	2.09
CXCR4 signalling	n	1.347	0.866
IL-7 signalling	n	1.291	1.88
CCR3 in eosinophils	n	1.291	0.429
MIF regulation of innate immunity	n	1.265	1.25
Leukocyte extravasation signalling	n	1.225	0.466
TGF-β signalling	n	1.147	1.53
iNOS signalling	n	-1.265	0.866
Interferon signalling	n	-1.342	0.349
IL-23 signalling	n	-1.633	0
T cell function
T cell exhaustion	n	1.095	4.14
4-1BB signalling in T lymphocytes	n	1.265	1.44
Th17 activation pathway	n	-1.667	0.266
Signalling cascades
p70S6K signalling	n	1.877	0.664
JAK/STAT signalling	n	-1.155	0
Apoptosis/cell death
Ca²⁺ induced lymphocyte apoptosis	n	1.508	0.578

Table 9. Animals selected for miRNA sequencing.

MAP Exposure	Disease Status	Lesion Pathologies	Vaccination Status	No. of animals (n)
Non-exposed	Control	None	Non-vaccinated	4
Non-exposed	Control	None	Vaccinated	4
Exposed	Infected	Multibacillary	Non-vaccinated	5
Exposed	Infected	Paucibacillary-Multibacillary	Vaccinated	3
Exposed	Resilient	None-Paucibacillary	Non-vaccinated	4
Exposed	Resilient	Paucibacillary	Vaccinated	4

No competing interests reported.

SeqManuscriptsupplementarydata.docx

Download PDF

Editorial decision: Revision requested
04 Dec, 2023
Submission checks completed at journal
10 Nov, 2023
Editor assigned by journal
10 Nov, 2023
First submitted to journal
08 Nov, 2023

You are reading this latest preprint version

Whole transcriptome analysis of Mycobacterium avium subsp. paratuberculosis infection and identification of circulating miRNA as biomarkers for resilience

Status:

Version 1

Abstract

Figures

Introduction

Results

Discussion

Methods

Declarations

References

Tables

Additional Declarations

Supplementary Files

Status:

Version 1